Available under License .
209kB
Official URL: http://www.hivnursing.net/past_articles/volume_sea...
Writing and research can be challenging for nurses at undergraduate and postgraduate level; however, understanding the process and developing the skills to conduct a literature review with a staged strategy will positively affect care delivery. Nurses have a responsibility to deliver care based on the best evidence available. Therefore, developing the necessary skills to conduct a literature review is beneficial to both nurses and those in their care.
Repository Staff Only: item control page
Item Type: | Article |
---|---|
Creators (Authors or editors): | |
Subjects: | > |
Schools: | > |
ID Code: | 24898 |
Depositing User ID: | |
Date Deposited: | 21 Nov 2018 12:27 |
Last Modified: | 02 Sep 2024 13:47 |
Research guides, research in education.
General literature review guides.
Libraries & Cultural Resources
BMC Medical Informatics and Decision Making volume 24 , Article number: 241 ( 2024 ) Cite this article
Metrics details
Successful deployment of clinical prediction models for clinical deterioration relates not only to predictive performance but to integration into the decision making process. Models may demonstrate good discrimination and calibration, but fail to match the needs of practising acute care clinicians who receive, interpret, and act upon model outputs or alerts. We sought to understand how prediction models for clinical deterioration, also known as early warning scores (EWS), influence the decision-making of clinicians who regularly use them and elicit their perspectives on model design to guide future deterioration model development and implementation.
Nurses and doctors who regularly receive or respond to EWS alerts in two digital metropolitan hospitals were interviewed for up to one hour between February 2022 and March 2023 using semi-structured formats. We grouped interview data into sub-themes and then into general themes using reflexive thematic analysis. Themes were then mapped to a model of clinical decision making using deductive framework mapping to develop a set of practical recommendations for future deterioration model development and deployment.
Fifteen nurses ( n = 8) and doctors ( n = 7) were interviewed for a mean duration of 42 min. Participants emphasised the importance of using predictive tools for supporting rather than supplanting critical thinking, avoiding over-protocolising care, incorporating important contextual information and focusing on how clinicians generate, test, and select diagnostic hypotheses when managing deteriorating patients. These themes were incorporated into a conceptual model which informed recommendations that clinical deterioration prediction models demonstrate transparency and interactivity, generate outputs tailored to the tasks and responsibilities of end-users, avoid priming clinicians with potential diagnoses before patients were physically assessed, and support the process of deciding upon subsequent management.
Prediction models for deteriorating inpatients may be more impactful if they are designed in accordance with the decision-making processes of acute care clinicians. Models should produce actionable outputs that assist with, rather than supplant, critical thinking.
• This article explored decision-making processes of clinicians using a clinical prediction model for deteriorating patients, also known as an early warning score.
• Our study identified that the clinical utility of deterioration models may lie in their assistance in generating, evaluating, and selecting diagnostic hypotheses, an important part of clinical decision making that is underrepresented in the prediction modelling literature.
• Nurses in particular stressed the need for models that encourage critical thinking and further investigation rather than prescribe strict care protocols.
Peer Review reports
The number of ‘clinical prediction model’ articles published on PubMed has grown rapidly over the past two decades, from 1,918 articles identified with these search terms published in 2002 to 26,326 published in 2022. A clinical prediction model is defined as any multivariable model that provides patient-level estimates of the probability or risk of a disease, condition or future event [ 1 , 2 , 3 ].
Recent systematic and scoping reviews report a lack of evidence that clinical decision support systems based on prediction models are associated with improved patient outcomes once implemented in acute care [ 4 , 5 , 6 , 7 ]. One potential reason may be that some models are not superior to clinical judgment in reducing missed diagnoses or correctly classifying non-diseased patients [ 8 ]. While improving predictive accuracy is important, this appears insufficient for improving patient outcomes, suggesting that more attention should be paid to the process and justification of how prediction models are designed and deployed [ 9 , 10 ].
If model predictions are to influence clinical decision-making, they must not only demonstrate acceptable accuracy, but also be implemented and adopted at scale in clinical settings. This requires consideration of how they are integrated into clinical workflows, how they generate value for users, and how clinicians perceive and respond to their outputs of predicted risks [ 11 , 12 ]. These concepts are tenets of user-centred design, which focuses on building systems based on the needs and responsibilities of those who will use them. User-centred decision support tools can be designed in a variety of ways, but may benefit from understanding the characteristics of the users and the local environment in which tools are implemented, [ 13 ] the nature of the tasks end-users are expected to perform, [ 14 ] and the interface between the user and the tools [ 15 ].
A common task for prediction models integrated into clinical decision support systems is in predicting or recognising clinical deterioration, also known as early warning scores. Clinical deterioration is defined as the transition of a patient from their current health state to a worse one that puts them at greater risk of adverse events and death [ 16 ]. Early warning scores were initially designed to get the attention of skilled clinicians when patients began to deteriorate, but have since morphed into complex multivariable prediction models [ 17 ]. As with many other clinical prediction models, early warning scores often fail to demonstrate better patient outcomes once deployed [ 4 , 18 ]. The clinical utility of early warning scores likely rests on two key contextual elements: the presence of uncertainty, both in terms of diagnosis and prognosis, and the potential for undesirable patient outcomes if an appropriate care pathway is delayed or an inappropriate one is chosen [ 19 ].
The overarching goal of this qualitative study was to determine how prediction models for clinical deterioration, or early warning scores, could be better tailored to the needs of end-users to improve inpatient care. This study had three aims. First, to understand the experiences and perspectives of nurses and doctors who use early warning scores. Second, to identify the tasks these clinicians performed when managing deteriorating patients, the decision-making processes that guided these tasks, and how these could be conceptualised schematically. Finally, to address these tasks and needs with actionable, practical recommendations for enhancing future deterioration prediction model development and deployment.
To achieve our study aims, we conducted semi-structured interviews of nurses and doctors at two large, digitally mature hospitals. We first asked clinicians to describe their backgrounds, perspectives, and experience with early warning scores to give context to our analysis. We then examined the tasks and responsibilities of participants and the decision-making processes that guided these tasks using reflexive thematic analysis, an inductive method that facilitated the identification of general themes. We then identified a conceptual decision-making framework from the literature to which we mapped these themes to understand how they may lead to better decision support tools. Finally, we used this framework to formulate recommendations for deterioration prediction model design and deployment. These steps are presented graphically in a flow diagram (Fig. 1 ).
Schema of study goal, aims and methods
The study was conducted at one large tertiary and one medium-sized metropolitan hospital in Brisbane, Australia. The large hospital contained over 1,000 beds, handling over 116,000 admissions and approximately 150,000 deterioration alerts per year in 2019. Over the same period, the medium hospital contained 175 beds, handling over 31,000 admissions and approximately 42,000 deterioration alerts per year. These facilities had a high level of digital maturity, including fully integrated electronic medical records.
The deterioration monitoring system used at both hospitals was the Queensland Adult Deterioration Detection System (Q-ADDS) [ 20 , 21 ]. Q-ADDS uses an underlying prediction model to convert patient-level vital signs from a single time of observation into an ordinal risk score describing an adult patient’s risk of acute deterioration. Vital signs collected are respiratory rate (breaths/minute), oxygen flow rate (L/minute), arterial oxygen saturation (percent), blood pressure (mmHg), heart rate (beats/minute), temperature (degrees Celsius), level of consciousness (Alert-Voice-Pain-Unresponsive) and increased or new onset agitation. Increased pain and urine output are collected but not used for score calculation [ 21 ]. The Q-ADDS tool is included in the supplementary material.
Vital signs are entered into the patient’s electronic medical record, either imported from the vital signs monitoring device at the patient’s bedside or from manual entry by nurses. Calculations are made automatically within Q-ADDS to generate an ordinal risk score per patient observation. Scores can be elevated to levels requiring a tiered escalation response if a single vital sign is greatly deranged, or if several observations are deranged by varying degrees. Scores range from 0 to 8+, with automated alerts and escalation protocols ranging from more frequent observations for lower scores to immediate activation of the medical emergency team (MET) at higher scores.
The escalation process for Q-ADDS is highly structured, mandated and well documented [ 21 ]. Briefly, when a patient’s vital signs meet a required alert threshold, the patient’s nurse is required to physically assess the patient and, depending on the level of severity predicted by Q-ADDS, notify the patient’s doctor (escalation). The doctor is then required to be notified of the patient’s Q-ADDS score, potentially review the patient, and discuss any potential changes to care with the nurse. Both nurses and doctors can escalate straight to MET calls or an emergency ‘code blue’ call (requiring cardiopulmonary resuscitation or assisted ventilation) at any time if necessary.
Participant recruitment began in February 2022 and concluded in March 2023, disrupted by the COVID-19 pandemic. Eligibility criteria were nurses or doctors at each hospital with direct patient contact who either receive or respond, respectively, to Q-ADDS alerts. An anticipated target sample size of 15 participants was established prior to recruitment, based on expected constraints in recruitment due to clinician workloads and the expected length of interviews relative to their scope, as guided by prior research [ 22 ]. As the analysis plan involved coding interviews iteratively as they were conducted, the main justification for ceasing recruitment was when no new themes relating to the study objectives were generated during successive interviews as the target sample size was approached [ 23 ].
Study information was broadly distributed via email to nurses and doctors in patient-facing roles across hospitals. Nurse unit managers were followed up during regular nursing committee meetings to participate or assist with recruitment within their assigned wards. Doctors were followed up by face-to-face rounding. Snowball sampling, in which participants were encouraged to refer their colleagues for study participation, was employed whenever possible. In all cases, study authors explained study goals and distributed participant consent forms prior to interview scheduling with the explicit proviso that participation was completely voluntary and anonymous to all but two study authors (RB and SN).
We used a reflexive framework method to develop an open-ended interview template [ 24 ] that aligned with our study aims. Interview questions were informed by the non-adoption, abandonment, scale-up, spread and sustainability (NASSS) framework [ 25 ]. The NASSS framework relates the end-user perceptions of the technology being evaluated to its value proposition for the clinical situation to which it is being applied. We selected a reflexive method based on the NASSS for our study as we wanted to allow end-users to speak freely about the barriers they faced when using prediction models for clinical deterioration, but did not limit participants to discussing only topics that could fit within the NASSS framework.
Participants were first asked about their background and clinical expertise. They were then invited to share their experiences and perspectives with using early warning scores to manage deteriorating patients. This was used as a segue for participants to describe the primary tasks required of them when evaluating and treating a deteriorating patient. Participants were encouraged to talk through their decision-making process when fulfilling these tasks, and to identify any barriers or obstacles to achieving those tasks that were related to prediction models for deteriorating patients. Participants were specifically encouraged to identify any sources of information that were useful for managing deteriorating patients, including prediction models for other, related disease groups like sepsis, and to think of any barriers or facilitators for making that information more accessible. Finally, participants were invited to suggest ways to improve early warning scores, and how those changes may lead to benefits for patients and clinicians.
As we employed a reflexive methodology to allow clinicians to speak freely about their perspectives and opinions, answers to interview questions were optional and open-ended, allowing participants to discuss relevant tangents. Separate interview guides were developed for nurses and doctors as the responsibilities and information needs of these two disciplines in managing deteriorating patients often differ. Nurses are generally charged with receiving and passing on deterioration alerts, while doctors are generally charged with responding to alerts and making any required changes to patient care plans [ 4 ]. Interview guides are contained in the supplement.
Due to clinician workloads, member checking, a form of post-interview validation in which participants retrospectively confirm their interview answers, was not used. To ensure participants perceived the interviewers as being impartial, two study authors not employed by the hospital network and not involved in direct patient care (RB and SN) were solely responsible for conducting interviews and interrogating interview transcripts. Interviews were recorded and transcribed verbatim, then re-checked for accuracy.
Transcripts were analysed using a reflexive thematic methodology informed by Braun and Clarke [ 26 ]. This method was selected because it facilitated exploring the research objectives rather than being restricted to the domains of a specific technology adoption framework, which may limit generalisability [ 27 ]. Interviews were analysed over five steps to identify emergent themes.
Each interview was broken down into segments by RB and SN, where segments corresponded to a distinct opinion.
Whenever appropriate, representative quotes for each distinct concept were extracted.
Segments were grouped into sub-themes.
Sub-themes were grouped into higher-order themes, or general concepts.
Steps 1 through 4 were iteratively repeated by RB and supervised by SN.
As reflexive methods incorporate the experiences and expertise of the analysts, our goal was to extract any sub-themes relevant to the study aims and able to be analysed in the context of early warning scores, prediction models, or decision support tools for clinical deterioration. The concepts explored during this process were not exhaustive, but repeated analysis and re-analysis of participant transcripts helped to ensure all themes could be interpreted in the context of our three study aims: background and perspectives, tasks and decision-making, and recommendations for future practice.
Once the emergent themes from the inductive analysis were defined, we conducted a brief scan of PubMed for English-language studies that investigated how the design of clinical decision support systems relate to clinical decision-making frameworks. The purpose of this exercise was to identify a framework against which we could map the previously elicited contexts, tasks, and decision-making of end-users in developing a decision-making model that could then be used to support the third aim of formulating recommendations to enhance prediction model development and deployment.
RB and SN then mapped higher-order themes from the inductive analysis to the decision-making model based on whether there was a clear relationship between each theme and a node in the model (see Results).
Recommendations for improving prediction model design were derived by reformatting the inductive themes based on the stated preferences of the participants. These recommendations were then assessed by the remaining authors and the process repeated iteratively until authors were confident that all recommendations were concordant with the decision-making model.
Our sample included 8 nurses and 7 doctors of varying levels of expertise and clinical specialties; further information is contained in the supplement. Compared to doctors, nurse participants were generally more experienced, often participating in training or mentoring less experienced staff. Clinical specialities of nurses were diverse, including orthopaedics, cancer services, medical assessment and planning unit, general medicine, and pain management services. Doctor participants ranged from interns with less than a year of clinical experience up to consultant level, including three doctors doing training rotations and two surgical registrars. Clinical specialties of doctors included geriatric medicine, colorectal surgery, and medical education.
Eleven interviews were conducted jointly by RB and SN, one conducted by RB, and three by SN. Interviews were scheduled for up to one hour, with a mean duration of 42 min. Six higher-order themes were identified. These were: added value of more information; communication of model outputs; validation of clinical intuition; capability for objective measurement; over-protocolisation of care; and model transparency and interactivity (Table 1 ). Some aspects of care, including the need for critical thinking and the informational value of discerning trends in patient observations, were discussed in several contexts, making them relevant to more than one higher-order theme.
Clinicians identified that additional data or variables important for decision making were often omitted from the Q-ADDS digital interface. Such variables included current medical conditions, prescribed medications and prior observations, which were important for interpreting current patient data in the context of their baseline observations under normal circumstances (e.g., habitually low arterial oxygen saturation due to chronic obstructive pulmonary disease) or in response to an acute stimulus (e.g., expected hypotension for next 4 to 8 h while treatment for septic shock is underway).
“The trend is the biggest thing [when] looking at the data , because sometimes people’s observations are deranged forever and it’s not abnormal for them to be tachycardic , whereas for someone else , if it’s new and acute , then that’s a worry.” – Registrar.
Participants frequently emphasised the critical importance of looking at patients holistically, or that patients were more than the sum of the variables used to predict risk. Senior nurses stressed that prediction models were only one part of patient evaluation, and clinicians should be encouraged to incorporate both model outputs and their own knowledge and experiences in decision making rather than trust models implicitly. Doctors also emphasised this holistic approach, adding that they placed more importance on hearing a nurse was concerned for the patient than seeing the model output. Critical thinking about future management was frequently raised in this context, with both nurses and doctors insisting that model predictions and the information required for contextualising risk scores should be communicated together when escalating the patient’s care to more senior clinicians.
Model outputs were discussed in two contexts. First, doctors perceived that ordinal risk scores generated by Q-ADDS felt arbitrary compared to receiving probabilities of a future event, for example cardiorespiratory decompensation, that required a response such as resuscitation or high-level treatment. However, nurses did not wholly embrace probabilities as outputs, instead suggesting that recommendations for how they should respond to different Q-ADDS scores were more important. This difference may reflect the different roles of alert receivers (nurses) and alert responders (doctors).
“[It’s helpful] if you use probabilities… If your patient has a sedation score of 2 and a respiratory rate of 10 , [giving them] a probability of respiratory depression would be helpful. However , I don’t find many clinicians , and certainly beginning practitioners , think in terms of probabilities.” – Clinical nurse consultant.
Second, there was frequent mention of alert fatigue in the context of model outputs. One doctor and two nurses felt there was insufficient leeway for nurses to exercise discretion in responding to risk scores, leading to many unnecessary alert-initiated actions. More nuance in the way Q-ADDS outputs were delivered to clinicians with different roles was deemed important to avoid model alerts being perceived as repetitive and unwarranted. However, three other doctors warned against altering MET call criteria in response to repetitive and seemingly unchanging risk scores and that at-risk patients should, as a standard of care, remain under frequent observation. Frustrations centred more often around rigidly tying repetitive Q-ADDS outputs to certain mandated actions, leading to multiple clinical reviews in a row for a patient whose trajectory was predictable, for example a patient with stable heart failure having a constantly low blood pressure. This led to duplication of nursing effort (e.g., repeatedly checking the blood pressure) and the perception that prediction models were overly sensitive.
“It takes away a lot of nurses’ critical judgement. If someone’s baseline systolic [blood pressure] is 95 [mmHg] , they’re asymptomatic and I would never hear about it previously. We’re all aware that this is where they sit and that’s fine. Now they are required to notify me in the middle of the night , “Just so you know , they’ve dropped to 89 [below an alert threshold of 90mmHg].“” – Junior doctor.
Clinicians identified the ability of prediction models to validate their clinical intuition as both a benefit and a hindrance, depending on how outputs were interpreted and acted upon. Junior clinicians appreciated early warning scores giving them more support to escalate care to senior clinicians, as a conversation starter or framing a request for discussion. Clinicians described how assessing the patient holistically first, then obtaining model outputs to add context and validate their diagnostic hypotheses, was very useful in deciding what care should be initiated and when.
“You kind of rule [hypotheses] out… you go to the worst extreme: is it something you need to really be concerned about , especially if their [score] is quite high? You’re thinking of common complications like blood clots , so that presents as tachycardic… I’m thinking of a PE [pulmonary embolism] , then you do the nursing interventions.” – Clinical nurse manager.
While deterioration alerts were often seen as triggers to think about potential causes for deterioration, participants noted that decision making could be compromised if clinicians were primed by model outputs to think of different diagnoses before they had fully assessed the patient at the bedside. Clinicians described the dangers of tunnel vision or, before considering all available clinical information, investigating favoured diagnoses to the exclusion of more likely causes.
“[Diagnosis-specific warnings are] great , [but] that’s one of those things that can lead to a bit of confirmation bias… It’s a good trigger to articulate , “I need to look for sources of infection when I go to escalate"… but then , people can get a little bit sidetracked with that and ignore something more blatant in front of them. I’ve seen people go down this rabbit warren of being obsessed with the “fact” that it was sepsis , but it was something very , very unrelated.” – Nurse educator.
Clinicians perceived that prediction models were useful as more objective measures of patients’ clinical status that could ameliorate clinical uncertainty or mitigate cognitive biases. In contrast to the risk of confirmation bias arising from front-loading model outputs suggesting specific diagnoses, prediction models could offer a second opinion that could help clinicians recognise opposing signals in noisy data that, in particular, assisted in considering serious diagnoses that shouldn’t be missed (e.g., sepsis), or more frequent and easily treated diagnoses (e.g., dehydration). Prediction models were also useful when they disclosed several small, early changes in patient status that provided an opportunity for early intervention.
“Maybe [the patient has] a low grade fever , they’re a bit tachycardic. Maybe [sepsis] isn’t completely out of the blue for this person. If there was some sort of tool , that said there’s a reasonable chance that they could have sepsis here , I would use that to justify the option of going for blood cultures and maybe a full septic screen. If [I’m indecisive] , that sort of information could certainly push me in that direction.” – Junior doctor.
Clinicians frequently mentioned that prediction models would have been more useful when first starting clinical practice, but become less useful with experience. However, clinicians noted that at any experience level, risk scoring was considered most useful as a triage/prioritisation tool, helping decide which patients to see first, or which clinical concerns to address first.
“[Doctors] can easily triage a patient who’s scoring 4 to 5 versus 1 to 3. If they’re swamped , they can change the escalation process , or triage appropriately with better communication.” – Clinical nurse manager.
Clinicians also stressed that predictions were not necessarily accurate because measurement error or random variation, especially one-off outlier values for certain variables, was a significant contributor to false alerts and inappropriate responses. For example, a single unusually high respiratory rate generated an unusually high risk score, prompting an unnecessary alert.
The sentiment most commonly expressed by all experienced nursing participants and some doctors was that nurses were increasingly being trained to solely react to model outputs with fixed response protocols, rather than think critically about what is happening to patients and why. It was perceived that prediction models may actually reduce the capacity for clinicians to process and internalise important information. For example, several nurses observed their staff failing to act on their own clinical suspicions that patients were deteriorating because the risk score had not exceeded a response threshold.
“We’ve had patients on the ward that have had quite a high tachycardia , but it’s not triggering because it’s below the threshold to trigger… [I often need to make my staff] make the clinical decision that they can call the MET anyway , because they have clinical concern with the patient.” – Clinical nurse consultant.
A source of great frustration for many nurses was the lack of critical thinking by their colleagues of possible causes when assessing deteriorating patients. They wanted their staff to investigate whether early warning score outputs or other changes in patient status were caused by simple, easily fixable issues such as fitting the oxygen mask properly and helping the patient sit up to breathe more easily, or whether they indicated more serious underlying pathophysiology. Nurses repeatedly referenced the need for clinicians to always be asking why something was happening, not simply reacting to what was happening.
“[Models should also be] trying to get back to critical thinking. What I’m seeing doesn’t add up with the monitor , so I should investigate further than just simply calling the code.” – Clinical nurse educator.
Clinicians frequently requested more transparent and interactive prediction models. These included a desire to receive more training in how prediction models worked and how risk estimates were generated mathematically, and being able to visualise important predictors of deterioration and the absolute magnitude of their effects (effect sizes) in intuitive ways. For example, despite receiving training in Q-ADDS, nurses expressed frustrations that nobody at the hospital seemed to understand how it worked in generating risk scores. Doctors were interested in being able to visualise the relative size and direction of effect of different model variables, potentially using colour-coding, combined with other contextual patient data like current vital sign trends and medications, and presented on one single screen.
The ability to modify threshold values for model variables and see how this impacted risk scores, and what this may then mean for altering MET calling criteria, was also discussed. For example, in an older patient with an acute ischaemic stroke, a persistently high, asymptomatic blood pressure value is an expected bodily response to this acute insult over the first 24–48 h. In the absence of any change to alert criteria, recurrent alerts would be triggered which may encourage overtreatment and precipitous lowering of the blood pressure with potential to cause harm. Altering the criteria to an acceptable or “normal” value for this clinical scenario (i.e. a higher than normal blood pressure) may generate a lower, more patient-centred risk estimate and less propensity to overtreat. This ability to tinker with the model may also enhance understanding of how it works.
“I wish I could alter criteria and see what the score is after that , with another set of observations. A lot of the time… I wonder what they’re sitting at , now that I’ve [altered] the bit that I’m not concerned about… It would be quite helpful to refresh it and have their score refreshed as the new score.” – Junior doctor.
Guided by the responses of our participants regarding their decision-making processes, our literature search identified a narrative review by Banning (2008) that reported previous work by O’Neill et al. (2005) [ 28 , 29 ]. While these studies referred to models of nurse decision-making, we selected a model (Fig. 2 ) that also appropriately described the responses of doctors in our participant group and matched the context of using clinical decision support systems to support clinical judgement. As an example, when clinicians referenced needing to look for certain data points to give context to a patient assessment, this was mapped to nodes relating to “Current patient data,” “Changes to patient status/data,” and “Hypothesis-driven assessment.”
Decision-making model(Adapted from Neill’s clinical decision making framework [2005] and modified by Banning [2006]) with sequential decision nodes
The themes from Table 1 were mapped to the nodes in the decision-making model based on close alignment with participant responses (see Fig. 3 ). This mapping is further explained below, where the nodes in the model are described in parentheses.
Value of additional information for decision-making : participants stressed the importance of understanding not only the data going into the prediction model, but also how that data changed over time as trends, and the data that were not included in the model. (Current patient data, changes to patient status/data)
Format, frequency, and relevance of outputs : participants suggested a change in patient data should not always lead to an alert. Doctors, but not necessarily nurses, proposed outputs displayed as probabilities rather than scores, tying model predictions to potential diagnoses or prognoses. (Changes to patient status/data, hypothesis generation)
Using models to validate but not supersede clinical intuition : Depending on the exact timing of model outputs within the pathway of patient assessment, participants found predictions could either augment or hinder the hypothesis generation process. (Hypothesis generation)
Measuring risks objectively : Risk scores can assist with triaging or prioritising patients by urgency or prognostic risk, thereby potentially leading to early intervention to identify and/or prevent adverse events. (Clinician concerns, hypothesis generation)
Supporting critical thinking and reducing over-protocolised care : by acting as triggers for further assessment, participants suggested prediction models can support or discount diagnostic hypotheses, lead to root-cause identification, and facilitate interim cares, for example by ensuring good fit of nasal prongs. (Provision of interim care, hypothesis generation, hypothesis-driven assessment)
Model transparency and interactivity : understanding how prediction models worked, being able to modify or add necessary context to model predictions, and understanding the relative contribution of different predictors could better assist the generation and selection of different hypotheses that may explain a given risk score. (Hypothesis generation, recognition of clinical pattern and hypothesis selection)
Mapping of the perceived relationships between higher-order themes and nodes in the decision-making model shown in Fig. 2
Based on the mapping of themes to the decision-making model, we formulated four recommendations for enhancing the development and deployment of prediction models for clinical deterioration.
Improve accessibility and transparency of data included in the model. Provide an interface that allows end-users to see what predictor variables are included in the model, their relative contributions to model outputs, and facilitate easy access to data not included in the model but still relevant for model-informed decisions, e.g., trends of predictor variables over time.
Present model outputs that are relevant to the end-user receiving those outputs, their responsibilities, and the tasks they may be obliged to perform, while preserving the ability of clinicians to apply their own discretionary judgement.
In situations associated with diagnostic uncertainty, avoid tunnel vision from priming clinicians with possible diagnostic explanations based on model outputs, prior to more detailed clinical assessment of the patient.
Support critical thinking whereby clinicians can apply a more holistic view of the patient’s condition, take all relevant contextual factors into account, and be more thoughtful in generating and selecting causal hypotheses.
This qualitative study involving front-line acute care clinicians who respond to early warning score alerts has generated several insights into how clinicians perceive the use of prediction models for clinical deterioration. Clinicians preferred models that facilitated critical thinking, allowed an understanding of the impact of variables included and excluded from the model, provided model outputs specific to the tasks and responsibilities of different disciplines of clinicians, and supported decision-making processes in terms of hypotheses and choice of management, rather than simply responding to alerts in a pre-specified, mandated manner. In particular, preventing prediction models from supplanting critical thinking was repeatedly emphasised.
Reduced staffing ratios, less time spent with patients, greater reliance on more junior workforce, and increasing dependence on automated activation of protocolised management are all pressures that could lead to a decline in clinical reasoning skills. This problem could be exacerbated by adding yet more predictive algorithms and accompanying protocols for other clinical scenarios, which may intensify alert fatigue and disrupt essential clinical care. However, extrapolating our results to areas other than clinical deterioration should be done with caution. An opposing view may be that using prediction models to reduce the burden of routine surveillance may allow redirection of critical thinking skills towards more useful tasks, a question that has not been explored in depth in the clinical informatics literature.
Clinicians expressed interest in models capable of providing causal insights into clinical deterioration. This is neither a function nor capability of most risk prediction models, requiring different assumptions and theoretical frameworks [ 30 ]. Despite this limitation, risk nomograms, visualisations of changes in risk with changes in predictor variables, and other interactive tools for estimating risk may be useful adjuncts for clinical decision-making due to the ease with which input values can be manipulated.
Our research supports and extends the literature on the acceptability of risk prediction models within clinical decision support systems. Common themes in the literature supporting good practices in clinical informatics and which are also reflected in our study include: alert fatigue; the delivery of more relevant contextual information; [ 31 ] the value of patient histories; [ 32 , 33 ] ranking relevant information by clinical importance, including colour-coding; [ 34 , 35 ] not using computerised tools to replace clinical judgement; [ 32 , 36 , 37 ] and understanding the analytic methods underpinning the tool [ 38 ]. One other study has investigated the perspectives of clinicians of relatively simple, rules-based prediction models similar to Q-ADDS. Kappen et al [ 12 ] conducted an impact study of a prediction model for postoperative nausea and vomiting and also found that clinicians frequently made decisions in an intuitive manner that incorporated information both included and absent from prediction models. However, the authors recommended a more directive than assistive approach to model-based recommendations, possibly due to a greater focus on timely prescribing of effective prophylaxis or treatment.
The unique contribution of our study is a better understanding of how clinicians may use prediction models to generate and validate diagnostic hypotheses. The central role of critical thinking and back-and-forth interactions between clinician and model in our results provide a basis for future research using more direct investigative approaches like cognitive task analysis [ 39 ]. Our study has yielded a set of cognitive insights into decision making that can be applied in tandem with statistical best practice in designing, validating and implementing prediction models. [ 19 , 40 , 41 ].
Our results may generalise to prediction models based on machine learning (ML) and artificial intelligence (AI), according to results of several recent studies. Tonekaboni et al [ 42 ] investigated clinician preferences for ML models in the intensive care unit and emergency department using hypothetical scenarios. Several themes appear both in our results and theirs: a need to understand the impact of both included and excluded predictors on model performance; the role of uncertain or noisy data in prediction accuracy; and the influence of trends or patient trajectories in decision making. Their recommendations for more transparent models and the delivery of model outputs designed for the task at hand align closely with ours. The authors’ focus on clinicians’ trust in the model was not echoed by our participants.
Eini-Porat et al [ 43 ] conducted a comprehensive case study of ML models in both adult and paediatric critical care. Their results present several findings supported by our participants despite differences in clinical environments: the value of trends and smaller changes in several vital signs that could cumulatively signal future deterioration; the utility of triage and prioritisation in time-poor settings; and the use of models as triggers for investigating the cause of deterioration.
As ML/AI models proliferate in the clinical deterioration prediction space, [ 44 ] it is important to deeply understand the factors that may influence clinician acceptance of more complex approaches. As a general principle, these methods often strive to input as many variables or transformations of those variables as possible into the model development process to improve predictive accuracy, incorporating dynamic updating to refine model performance. While this functionality may be powerful, highly complex models are not easily explainable, require careful consideration of generalisability, and can prevent clinicians from knowing when a model is producing inaccurate predictions, with potential for patient harm when critical healthcare decisions are being made [ 45 , 46 , 47 ]. Given that our clinicians emphasised the need to understand the model, know which variables are included and excluded, and correctly interpret the format of the output, ML/AI models in the future will need to be transparent in their development and their outputs easily interpretable.
The primary limitations of our study were that our sample was drawn from two hospitals with high levels of digital maturity in a metropolitan region of a developed country, with a context specific to clinical deterioration. Our sample of 15 participants may be considered small but is similar to that of other studies with a narrow focus on clinical perspectives [ 42 , 43 ]. All these factors can limit generalisability to other settings or to other prediction models. As described in the methods, we used open-ended interview templates and generated our inductive themes reflexively, which is vulnerable to different types of biases compared to more structured preference elicitation methods with rigidly defined analysis plans. Member checking may have mitigated this bias, but was not possible due to the time required from busy clinical staff.
Our study does not directly deal with methodological issues in prediction model development, [ 41 , 48 ] nor does it provide explicit guidance on how model predictions should be used in clinical practice. Our findings should also not be considered an exhaustive list of concerns clinicians have with prediction models for clinical deterioration, nor may they necessarily apply to highly specialised clinical areas, such as critical care. Our choice of decision making framework was selected because it demonstrated a clear, intuitive causal pathway for model developers to support the clinical decision-making process. However, other, equally valid frameworks may have led to different conclusions, and we encourage more research in this area.
This study elicited clinician perspectives of models designed to predict and manage impending clinical deterioration. Applying these perspectives to a decision-making model, we formulated four recommendations for the design of future prediction models for deteriorating patients: improved transparency and interactivity, tailoring models to the tasks and responsibilities of different end-users, avoiding priming clinicians with diagnostic predictions prior to in-depth clinical review, and finally, facilitating the diagnostic hypothesis generation and assessment process.
Due to privacy concerns and the potential identifiability of participants, interview transcripts are not available. However, interview guides are available in the supplement.
Jenkins DA, Martin GP, Sperrin M, Riley RD, Debray TPA, Collins GS, Peek N. Continual updating and monitoring of clinical prediction models: time for dynamic prediction systems? Diagn Prognostic Res. 2021;5(1):1.
Article Google Scholar
Collins GS, Reitsma JB, Altman DG, Moons KG. Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): the TRIPOD statement. BMJ. 2015;350:g7594.
Article PubMed Google Scholar
Moons KG, Altman DG, Reitsma JB, Ioannidis JP, Macaskill P, Steyerberg EW, et al. Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): explanation and elaboration. Ann Intern Med. 2015;162(1):W1–73.
Blythe R, Parsons R, White NM, Cook D, McPhail SM. A scoping review of real-time automated clinical deterioration alerts and evidence of impacts on hospitalised patient outcomes. BMJ Qual Saf. 2022;31(10):725–34.
Fahey M, Crayton E, Wolfe C, Douiri A. Clinical prediction models for mortality and functional outcome following ischemic stroke: a systematic review and meta-analysis. PLoS ONE. 2018;13(1):e0185402.
Article PubMed PubMed Central Google Scholar
Fleuren LM, Klausch TLT, Zwager CL, Schoonmade LJ, Guo T, Roggeveen LF, et al. Machine learning for the prediction of sepsis: a systematic review and meta-analysis of diagnostic test accuracy. Intensive Care Med. 2020;46(3):383–400.
White NM, Carter HE, Kularatna S, Borg DN, Brain DC, Tariq A, et al. Evaluating the costs and consequences of computerized clinical decision support systems in hospitals: a scoping review and recommendations for future practice. J Am Med Inform Assoc. 2023;30(6):1205–18.
Sanders S, Doust J, Glasziou P. A systematic review of studies comparing diagnostic clinical prediction rules with clinical judgment. PLoS ONE. 2015;10(6):e0128233.
Abell B, Naicker S, Rodwell D, Donovan T, Tariq A, Baysari M, et al. Identifying barriers and facilitators to successful implementation of computerized clinical decision support systems in hospitals: a NASSS framework-informed scoping review. Implement Sci. 2023;18(1):32.
van der Vegt AH, Campbell V, Mitchell I, Malycha J, Simpson J, Flenady T, et al. Systematic review and longitudinal analysis of implementing Artificial Intelligence to predict clinical deterioration in adult hospitals: what is known and what remains uncertain. J Am Med Inf Assoc. 2024;31(2):509–24.
Davenport T, Kalakota R. The potential for artificial intelligence in healthcare. Future Healthc J. 2019;6(2):94–8.
Kappen TH, van Loon K, Kappen MA, van Wolfswinkel L, Vergouwe Y, van Klei WA, et al. Barriers and facilitators perceived by physicians when using prediction models in practice. J Clin Epidemiol. 2016;70:136–45.
Witteman HO, Dansokho SC, Colquhoun H, Coulter A, Dugas M, Fagerlin A, Giguere AM, Glouberman S, Haslett L, Hoffman A, Ivers N. User-centered design and the development of patient decision aids: protocol for a systematic review. Systematic reviews. 2015;4:1−8.
Zhang J, Norman DA. Representations in distributed cognitive tasks. Cogn Sci. 1994;18(1):87–122.
Johnson CM, Johnson TR, Zhang J. A user-centered framework for redesigning health care interfaces. J Biomed Inf. 2005;38(1):75–87.
Jones D, Mitchell I, Hillman K, Story D. Defining clinical deterioration. Resuscitation. 2013;84(8):1029–34.
Morgan RJ, Wright MM. In defence of early warning scores. Br J Anaesth. 2007;99(5):747–8.
Article CAS PubMed Google Scholar
Smith ME, Chiovaro JC, O’Neil M, Kansagara D, Quinones AR, Freeman M, et al. Early warning system scores for clinical deterioration in hospitalized patients: a systematic review. Annals Am Thorac Soc. 2014;11(9):1454–65.
Baker T, Gerdin M. The clinical usefulness of prognostic prediction models in critical illness. Eur J Intern Med. 2017;45:37–40.
Campbell V, Conway R, Carey K, Tran K, Visser A, Gifford S, et al. Predicting clinical deterioration with Q-ADDS compared to NEWS, between the flags, and eCART track and trigger tools. Resuscitation. 2020;153:28–34.
The Australian Commission on Safety and Quality in Health is the publisher, and the publisher location is Sydney, Australia. https://www.safetyandquality.gov.au/sites/default/files/migrated/35981-ChartDevelopment.pdf .
Vasileiou K, Barnett J, Thorpe S, Young T. Characterising and justifying sample size sufficiency in interview-based studies: systematic analysis of qualitative health research over a 15-year period. BMC Med Res Methodol. 2018;18(1):148.
Hennink MM, Kaiser BN, Marconi VC. Code saturation versus meaning saturation: how many interviews are Enough? Qual Health Res. 2017;27(4):591–608.
Gale NK, Heath G, Cameron E, Rashid S, Redwood S. Using the framework method for the analysis of qualitative data in multi-disciplinary health research. BMC Med Res Methodol. 2013;13(1):1–8.
Greenhalgh T, Wherton J, Papoutsi C, Lynch J, Hughes G, A’Court C, et al. Beyond adoption: a New Framework for Theorizing and evaluating nonadoption, abandonment, and challenges to the Scale-Up, Spread, and sustainability of Health and Care technologies. J Med Internet Res. 2017;19(11):e367.
Braun V, Clarke V. One size fits all? What counts as quality practice in (reflexive) thematic analysis? Qualitative Res Psychol. 2021;18(3):328–52.
Campbell KA, Orr E, Durepos P, Nguyen L, Li L, Whitmore C, et al. Reflexive thematic analysis for applied qualitative health research. Qualitative Rep. 2021;26(6):2011–28.
Google Scholar
Banning M. A review of clinical decision making: models and current research. J Clin Nurs. 2008;17(2):187–95.
O’Neill ES, Dluhy NM, Chin E. Modelling novice clinical reasoning for a computerized decision support system. J Adv Nurs. 2005;49(1):68–77.
Arnold KF, Davies V, de Kamps M, Tennant PWG, Mbotwa J, Gilthorpe MS. Reflection on modern methods: generalized linear models for prognosis and intervention—theory, practice and implications for machine learning. Int J Epidemiol. 2020;49(6):2074–82.
Article PubMed Central Google Scholar
Westerbeek L, Ploegmakers KJ, de Bruijn GJ, Linn AJ, van Weert JCM, Daams JG, et al. Barriers and facilitators influencing medication-related CDSS acceptance according to clinicians: a systematic review. Int J Med Informatics. 2021;152:104506.
Henshall C, Marzano L, Smith K, Attenburrow MJ, Puntis S, Zlodre J, et al. A web-based clinical decision tool to support treatment decision-making in psychiatry: a pilot focus group study with clinicians, patients and carers. BMC Psychiatry. 2017;17(1):265.
Weingart SN, Simchowitz B, Shiman L, Brouillard D, Cyrulik A, Davis RB, et al. Clinicians’ assessments of electronic medication safety alerts in ambulatory care. Arch Intern Med. 2009;169(17):1627–32.
Baysari MT, Zheng WY, Van Dort B, Reid-Anderson H, Gronski M, Kenny E. A late attempt to involve end users in the design of medication-related alerts: Survey Study. J Med Internet Res. 2020;22(3):e14855.
Trafton J, Martins S, Michel M, Lewis E, Wang D, Combs A, et al. Evaluation of the acceptability and usability of a decision support system to encourage safe and effective use of opioid therapy for chronic, noncancer pain by primary care providers. Pain Med. 2010;11(4):575–85.
Wipfli R, Betrancourt M, Guardia A, Lovis C. A qualitative analysis of prescription activity and alert usage in a computerized physician order entry system. Stud Health Technol Inform. 2011;169:940–4.
PubMed Google Scholar
Cornu P, Steurbaut S, De Beukeleer M, Putman K, van de Velde R, Dupont AG. Physician’s expectations regarding prescribing clinical decision support systems in a Belgian hospital. Acta Clin Belg. 2014;69(3):157–64.
Ahearn MD, Kerr SJ. General practitioners’ perceptions of the pharmaceutical decision-support tools in their prescribing software. Med J Australia. 2003;179(1):34–7.
Swaby L, Shu P, Hind D, Sutherland K. The use of cognitive task analysis in clinical and health services research - a systematic review. Pilot Feasibility Stud. 2022;8(1):57.
Steyerberg EW. Applications of prediction models. In: Steyerberg EW, editor. Clinical prediction models. New York: Springer-; 2009. pp. 11–31.
Chapter Google Scholar
Steyerberg EW, Vergouwe Y. Towards better clinical prediction models: seven steps for development and an ABCD for validation. Eur Heart J. 2014;35(29):1925–31.
Tonekaboni S, Joshi S, McCradden MD, Goldenberg A. What Clinicians Want: Contextualizing Explainable Machine Learning for Clinical End Use. In: Doshi-Velez F, Fackler J, Jung K, Kale D, Ranganath R, Wallace B, Wiens J, editors. Proceedings of the 4th Machine Learning for Healthcare Conference; Proceedings of Machine Learning Research: PMLR; 2019;106:359–80.
Eini-Porat B, Amir O, Eytan D, Shalit U. Tell me something interesting: clinical utility of machine learning prediction models in the ICU. J Biomed Inform. 2022;132:104107.
Muralitharan S, Nelson W, Di S, McGillion M, Devereaux PJ, Barr NG, Petch J. Machine learning-based early warning systems for clinical deterioration: systematic scoping review. J Med Internet Res. 2021;23(2):e25187.
Rudin C. Stop Explaining Black Box Machine Learning Models for high stakes decisions and use interpretable models instead. Nat Mach Intell. 2019;1(5):206–15.
Blythe R, Parsons R, Barnett AG, McPhail SM, White NM. Vital signs-based deterioration prediction model assumptions can lead to losses in prediction performance. J Clin Epidemiol. 2023;159:106–15.
Futoma J, Simons M, Panch T, Doshi-Velez F, Celi LA. The myth of generalisability in clinical research and machine learning in health care. Lancet Digit Health. 2020;2(9):e489–92.
Steyerberg EW, Uno H, Ioannidis JPA, van Calster B, Collaborators. Poor performance of clinical prediction models: the harm of commonly applied methods. J Clin Epidemiol. 2018;98:133–43.
Download references
We would like to thank the participants who made time in their busy clinical schedules to speak to us and offer their support in recruitment.
This work was supported by the Digital Health Cooperative Research Centre (“DHCRC”). DHCRC is funded under the Commonwealth’s Cooperative Research Centres (CRC) Program. SMM was supported by an NHMRC-administered fellowships (#1181138).
Authors and affiliations.
Australian Centre for Health Services Innovation and Centre for Healthcare Transformation, School of Public Health and Social Work, Faculty of Health, Queensland University of Technology, 60 Musk Ave, Kelvin Grove, Brisbane, QLD, 4059, Australia
Robin Blythe, Sundresan Naicker, Nicole White & Steven M McPhail
Princess Alexandra Hospital, Metro South Health, Woolloongabba, QLD, Australia
Raelene Donovan & Andrew McKelliget
Queensland Digital Health Centre, Faculty of Medicine, University of Queensland, Brisbane, QLD, Australia
Ian A. Scott
Digital Health and Informatics Directorate, Metro South Health, Woolloongabba, QLD, Australia
Ian A. Scott & Steven M McPhail
You can also search for this author in PubMed Google Scholar
RB: conceptualisation, data acquisition, analysis, interpretation, writing. SN: data acquisition, analysis, interpretation, writing. NW: interpretation, writing. RD: data acquisition, interpretation, writing. IS: data acquisition, analysis, interpretation, writing. AM: data acquisition, interpretation, writing. SM: conceptualisation, data acquisition, analysis, interpretation, writing. All authors have approved the submitted version and agree to be accountable for the integrity and accuracy of the work.
Correspondence to Robin Blythe .
Ethics approval and consent to participate.
This study was approved by the Metro South Human Research Ethics Committee (HREC/2022/QMS/84205). Informed consent was obtained prior to interview scheduling, with all participants filling out a participant information and consent form. Consent forms were approved by the ethics committee. Participation was entirely voluntary, and could be withdrawn at any time. All responses were explicitly deemed confidential, with only the first two study authors and the participant privy to the research data. Interviews were then conducted in accordance with Metro South Health and Queensland University of Technology qualitative research regulations. For further information, please contact the corresponding author.
Not applicable.
The authors declare no competing interests.
Publisher’s note.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary material 1., rights and permissions.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .
Reprints and permissions
Cite this article.
Blythe, R., Naicker, S., White, N. et al. Clinician perspectives and recommendations regarding design of clinical prediction models for deteriorating patients in acute care. BMC Med Inform Decis Mak 24 , 241 (2024). https://doi.org/10.1186/s12911-024-02647-4
Download citation
Received : 06 September 2023
Accepted : 23 August 2024
Published : 02 September 2024
DOI : https://doi.org/10.1186/s12911-024-02647-4
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative
ISSN: 1472-6947
Loading metrics
Open Access
Peer-reviewed
Research Article
Roles Conceptualization, Data curation, Formal analysis, Methodology, Software, Validation, Writing – original draft, Writing – review & editing
* E-mail: [email protected]
Affiliation Lawrence S. Bloomberg Faculty of Nursing, University of Toronto, Toronto, Ontario, Canada
Roles Data curation, Formal analysis, Investigation, Validation, Writing – review & editing
Affiliation Princess Margaret Cancer Centre, Toronto, Canada
Roles Conceptualization, Investigation, Methodology, Validation, Writing – review & editing
Roles Formal analysis, Investigation, Writing – original draft
Roles Conceptualization, Formal analysis, Investigation, Methodology, Supervision, Validation, Writing – review & editing
In healthcare settings worldwide, workplace violence (WPV) has been extensively studied. However, significantly less is known about gender-based WPV and the characteristics of perpetrators. We conducted a comprehensive scoping review on Type II (directed by consumers) and Type III (perpetuated by healthcare workers) gender based-WPV among nurses and physicians globally. For the review, we followed the Preferred Reporting Items for Systematic and Meta Analyses extension for Scoping Review (PRISMA-ScR). The protocol for the comprehensive review was registered on the Open Science Framework on January 14, 2022, at https://osf.io/t4pfb/ . A systematic search in five health and social science databases yielded 178 relevant studies that indicated types of perpetrators, with only 34 providing descriptive data for perpetrators’ gender. Across both types of WPV, men (65.1%) were more frequently responsible for perpetuating WPV compared to women (28.2%) and both genders (6.7%). Type II WPV, demonstrated a higher incidence of violence against women; linked to the gendered roles, stereotypes, and societal expectations that allocate specific responsibilities based on gender. Type III WPV was further categorized into Type III-A (horizontal) and Type III-B (vertical). With Type III WPV, gendered power structures and stereotypes contributed to a permissive environment for violence by men and women that victimized more women. These revelations emphasize the pressing need for gender-sensitive strategies for addressing WPV within the healthcare sector. Policymakers must prioritize the security of healthcare workers, especially women, through reforms and zero-tolerance policies. Promoting gender equality and empowerment within the workforce and leadership is pivotal. Additionally, creating a culture of inclusivity, support, and respect, led by senior leadership, acknowledging WPV as a structural issue and enabling an open dialogue across all levels are essential for combating this pervasive problem.
Citation: Ayaz B, Dozois G, Baumann AL, Fuseini A, Nelson S (2024) Perpetrators of gender-based workplace violence amongst nurses and physicians–A scoping review of the literature. PLOS Glob Public Health 4(9): e0003646. https://doi.org/10.1371/journal.pgph.0003646
Editor: Tanmay Bagade, The University of Newcastle Australia: University of Newcastle, AUSTRALIA
Received: May 22, 2024; Accepted: August 2, 2024; Published: September 6, 2024
Copyright: © 2024 Ayaz et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are included in the manuscript and its supporting information files.
Funding: The authors received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
The International Labour Office (ILO), the International Council of Nurses (ICN), the World Health Organization (WHO), and Public Services International (PSI) defined WPV as "incidents where staff are abused, threatened or assaulted in circumstances related to their work, including commuting to and from work, involving an explicit or implicit challenge to their safety, well-being or health" [ 1 ]. The ILO [ 2 ] further defined "gender-based violence and harassment means violence and harassment directed at persons because of their sex or gender, or affecting persons of a particular sex or gender disproportionately".
Irrespective of industry, workplace violence (WPV) can cause lasting trauma and injuries and is a serious threat to human and health resources. WPV includes physical and psychological violence, including physical assault, verbal abuse, sexual harassment, and bullying. Gender-based workplace violence (GB-WPV), which is experienced across operational layers of an organization (horizontal) and organizational hierarchy (vertical), reinforces the differential risk for exposure and outcomes of violence for men and women [ 3 ]. Despite extensive research on workplace violence in healthcare, GB-WPV, its perpetrators, and its impact on healthcare professionals remains understudied. We presented the sex-segregated prevalence and risk factors for WPV somewhere else [ 4 ]. The earlier paper focused on the scope and scale of workplace violence (WPV), risk factors and its impact on men and women. As part of the same scoping review protocol, this paper reports on GB-WPV perpetrators. It specifically focuses on explaining the root causes of violent acts by individuals and the triggers and circumstances to provide gender-sensitive recommendations.
A systematic review [ 5 ] of the consequences of exposure to WPV in the healthcare setting based on 68 studies reported psychological and emotional effects such as post-traumatic stress, depression, anger, and fear. These effects impact work productivity, leading to increased sick leaves, poor job satisfaction, burnout, and higher attrition rates, particularly for women [ 5 ]. Studies have also shown that men are more likely to commit physical violence [ 6 ] and sexual harassment [ 7 , 8 ], while women are more often engaged in verbal abuse [ 9 ]. Moreover, gender stereotypes and inequalities in the distribution of roles and responsibilities can worsen power imbalances [ 3 ]. By recognizing and understanding these issues, employers and organizations can more effectively prevent and deal with gender-based workplace violence, ensuring a safer and more equitable work environment for everyone.
The classification of workplace violence has evolved, delineating distinct categories shaped by the nature of its perpetrators. The current working taxonomy categorizes WPV into four types based on the perpetrators of violence. This typology, as shown in Table 1 , emerged from a collaborative effort of a workshop on workplace violence intervention research held in Washington, DC, in 2000. The findings of this endeavour were subsequently published by the U.S. Department of Justice in 2001 [ 13 ]. Since then, this framework has been adopted by multiple organizations [ 10 , 11 , 14 ], and by researchers [ 12 , 15 ].
https://doi.org/10.1371/journal.pgph.0003646.t001
This paper explores the dynamics of workplace violence by categorizing and summarizing both Type II WPV, from patients and significant others, and Type III WPV (horizontal and vertical), which pertain to violence perpetuated by colleagues, supervisors, and administrators within the organization. Additionally, we explore the nuances of GB-WPV, considering both the perpetrators and nurses and physicians as victims of WPV. We summarized perpetrators based on their gender and synthesized the factors attributed to Type II and III, which are prevalent forms of violence reported in the literature. Type I and Type IV violence are beyond the scope of this paper as they focus on a security-based rather than workplace culture interventions. Understanding the factors contributing to these types of WPV is crucial to developing effective preventive interventions and strategies. Currently, there is a dearth of information identifying the characteristics of individuals who are more likely to commit GB-WPV and the characteristics of those targeted by such offences. This review addresses this gap by synthesizing data from studies that reported on the gender/sex data for various forms (please see S1 Text : Definitions of the Forms of Violence) of WPV and perpetrators of WPV among nurses and physicians.
While WPV affects individuals across the gender spectrum and in different professional groups, women are disproportionately affected. Some studies attributed it to their preponderance in the health workforce, their overrepresentation in lower positions in organizational and professional hierarchies, and societal gender norms in most cultures that subjugate women [ 9 , 16 ]. Recognizing that workplace violence is fundamentally intertwined with broader societal structures rooted in socioeconomic, cultural, and institutional factors, we underscore the necessity of a systematic approach to address this issue—one that is integrated, participatory, culturally and gender-sensitive, and non-discriminatory [ 1 ]. While current interventions aimed at addressing WPV primarily focus on assessing the effectiveness of training interventions to prevent and manage WPV in healthcare settings [ 17 – 19 ], they often lack gender-segregated findings for their effectiveness. Clarifying the existing situation on the gender of victims and perpetrators for specific Type/s of violence would help develop gender-sensitive interventions and policies to more effectively address WPV. This scoping review focuses on understanding GB-WPV and its perpetrators in the global health workforce, including nurses and physicians. Our preliminary search for a scoping review revealed that GB-WPV affects men, women, and non-binary persons. However, most studies included in this review reported gender as binary (men and women); only a few studies included non-binary personnel (for sample-see Table 2 , in results section). Therefore, we defined gender as a binary for this review and deliberated on it in the discussion section.
https://doi.org/10.1371/journal.pgph.0003646.t002
Our specific objectives set out for this paper were:
Following the Joanna Briggs Institute (JBI) revised guidelines, we conducted a scoping review. The protocol for this review was registered on the Open Science Framework on January 14, 2022, and is accessible at https://osf.io/t4pfb/ and S2 Text : Registered Protocol. The scoping literature review design addressed the research questions and accommodated the heterogeneous and complex nature of the literature. This method is appropriate for exploring the extent of the literature, mapping and summarizing the evidence, and identifying and analyzing knowledge gaps to inform future research. The framework used for this review consists of eight steps; they are built upon the seminal framework of Arksey and O’Malley’s scoping review, which was further developed by Levac and colleagues. The revised guidelines of JBI align these eight steps with the Preferred Reporting Items for Systematic and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR), ensuring rigour, transparency, and trustworthiness in Reporting the conduct of the scoping review. The first step of the scoping review framework is to align with research objectives, the title, and the inclusion criteria, as well as the exclusion criteria (see Box 1 ). Please see S1 Checklist : PRISMA-ScR Checklist.
Inclusion Criteria for Studies
1. The study participants included nurses and/ or physicians who experienced WPV during their careers.
2. Provided sex-segregated data for any form of violence and any type of perpetrators among nurses and physicians, including students, globally.
3. Published in English and after 2010.
Exclusion criteria
4. Studies that did not provide sex-segregated data and information for perpetrators.
5. Exclude systematic/ scoping reviews, concept or theoretical papers, and theses.
The research team collaborated with a health sciences librarian to develop a comprehensive search strategy. The systematic search focused on published literature in various databases, including Ovid MEDLINE: Epub Ahead of Print, In-Process and Other Non-Indexed Citations, which were translated in CINAHL Plus, APA PsycINFO, Web of Sciences, and Gender Studies Databases, Applied Social Sciences Index & Abstracts (ASSIA), and Sociological Abstracts ( S3 Text : Ovid MEDLINE search strategy, which was translated in all other databases). The search terms related to the population (midwifery, nursing, and physicians), concepts (violence and gender-based violence), and context (healthcare) were combined appropriately based on the scoping review objectives. These terms were identified through a preliminary literature search on various aspects of workplace violence in Google Scholar. The final search results were exported to EndNote, a citation manager, to de-duplicate sources from multiple databases. After de-duplication, the sources were imported into the Covidence online software program that streamlined the screening process by two independent reviewers. The final search of the literature review for this study was conducted on 11 th February 2024.
The identified sources were screened based on the inclusion criteria ( S4 Text : Excluded Sources). Two independent reviewers screened the titles and abstracts to shortlist the sources. Discrepancies were resolved through discussion and consensus, with a complete source review conducted, if necessary, followed by a full-text review against the inclusion criteria by two reviewers. The selection process is presented in the PRISMA diagram ( Fig 1 ). Given the overall objective of the review to map the most frequent forms and prevalence of GB-WPV for midwives, nurses, and physicians in different contexts and clinical settings, a quality assessment of the identified sources was not conducted.
https://doi.org/10.1371/journal.pgph.0003646.g001
This paper is a component of a multi-part scoping review; it reports on the perpetrators of WPV from gender-segregated prevalence data reported from a global context among the health workforce, including nurses and physicians. The prevalence and risk factors have been reported elsewhere [ 4 ]. This paper reports on Type II and Type III (vertical and horizontal) WPV perpetrators. Data from all sources ( S1 Data ) that reported sex/gender segregated findings and provided information for the types of perpetrators were included in mapping the prevalence of GB-WPV (See Table 2 ) for several types/forms of WPV and the clinical setting across countries/special regions. We could not calculate a mean score for various forms of violence based on gender for all the studies that provided information on perpetrators because of the wide variability in the operational definitions of the terms and the concepts in these studies. These studies also did not consistently provide quantifiable data for the Types of perpetrators. Only 34 studies (19%) provided the gender of perpetrators. We summarized the proportion of male and female perpetrators in those studies for Type II, Type III-A (horizontal) and Type III-B (Vertical) violence (see Table 3 ).
https://doi.org/10.1371/journal.pgph.0003646.t003
After de-duplication, 8435 possible references were imported for screening in the Covidence. These studies were screened against the title by one person, 1551 were shortlisted to be screened (for title and abstract) by two independent reviewers, and 402 were assessed for full-text eligibility. After applying the inclusion and exclusion criteria, 178 [ 6 – 9 , 15 , 20 – 125 ] studies were retained (PRISMA diagram, Fig 1 ) and analyzed to report on perpetrators that provided gender-segregated findings for WPV and information on various types of perpetrators ( Table 2 ). We included studies published between 2010–2024. The most common study design was quantitative, cross-sectional (n = 168), mixed methods (n = 4), and qualitative methods (n = 6).
A total of 178 studies provided information on the perpetrators of either Type II (consumers/patients, including patients’ companions), Type III-A (from colleagues), and Type III-B (from administrators and superior authorities within and between professions) violence. Studies included in this review did not consistently provide data for all types of violence and perpetrators; instead, they provided data for any Type/s. Of 178 studies, 141 (79%) reported perpetrators for Type II violence, followed by 93 (52.2%) for Type III-B (vertical) and 92 (51.6%) for Type III-A (horizontal) violence. Only 40 (22.5%) studies [ 9 , 21 – 53 , 164 , 168 , 172 , 177 , 178 , 191 ] reported information about all three types of violence.
While the search terms yielded many studies, there was significantly less information on the gender of perpetrators of WPV. Of the 178 studies reported on perpetrators, only 34 studies provided data for perpetrators’ gender (detailed in Table 3 ). Across the three types of violence, more men (65%) were responsible for perpetrating WPV compared to women (28%). Both men and women perpetuated violence in the remaining 7% of cases. Of the 34 studies, 25 studies reported on Type II violence, predominantly perpetrated by men, encompassing general violence [ 37 , 40 , 53 – 60 ], physical violence [ 6 , 35 , 44 , 61 – 63 , 178 ], verbal violence [ 6 , 35 , 44 , 53 , 63 , 172 , 178 ], and sexual harassment [ 8 , 41 , 44 , 63 , 64 ]. In most of these studies, women experienced a higher prevalence of violence than men. Gender-based workplace violence against nurses emerged as a pressing issue for Type II (56.2%) violence in ten studies [ 6 , 8 , 35 , 40 , 53 – 56 , 168 , 178 ]; men perpetrated 80% of the violence while women were responsible for only 19% violence, and almost all studies reported a higher prevalence of WPV against female nurses. A recent study [ 168 ] from 79 countries, though reported gender was not significant for WPV, being a nurse had higher odds of experiencing WPV (OR = 1.95; 95% CI 1.46 to 2.59, p<0.001) than a physician (OR = 1.70; 95% CI 1.33 to 2.18, p<0.001). In this study, most perpetrators were consumers (56%), followed by supervisors (16%) and colleagues (9%), or a combination of all (19%).
Violence perpetrated by colleagues (Type III-A) was reported by 15 studies, including seven for physicians [ 41 , 43 , 65 – 68 , 172 ], three studies for nurses [ 40 , 53 , 178 ], and five that included both professionals [ 29 , 37 , 44 , 60 , 63 ]. Approximately 24% of violence was perpetrated by colleagues (Type III-A) among nurses and physicians. More perpetrators were men (63.5%) than women (23%), and some violence by colleagues was reported as perpetrated by both men and women (13.5%). Only one study [ 29 ] reported higher rates of bullying by women (37.9%) than men (10.5%) and by both genders (51.6%). Two other studies reported higher mobbing behaviours (20% Vs. 69%) (192) and (8%vs.93%) (72) by women. In these studies, most perpetrators (40.7%) were supervisors and senior colleagues (Type III-B). Victims were both physicians (53.1%) and nurses (53.6%) with similar intensity, but a higher number of women (n = 195, 56.4%) were exposed to bullying than men (n = 18, 36%). Additionally, those who experienced bullying had lower levels of psychological health status. Bullying from colleagues (26.4%) and patients/consumers (7.7%) was perceived as less harmful than bullying from supervisors (Type III-B), which was also less reported because of the fear of consequences.
Of the 34 studies reporting on the gendered perpetuation of WPV ( Table 3 ), 24 reported on Type III-B (vertical) violence, which was more prevalent among physicians (51.5%) than nurses (16%). When it did occur among nurses, more men (77%) perpetuated Type III-B violence than women (18%) and both men and women (5%). Several studies highlighted physicians as perpetrators of WPV against nurses regardless of gender [ 8 , 51 , 53 ]. Similarly, more men (67.5%) than women (24.2%) and both genders (8.2%) perpetuated Type III-B violence among physicians. In seven of ten studies (70%) for Type III-B violence among physicians, male supervisors and administrators perpetuated sexual harassment [ 41 , 64 – 69 ]. Four studies reported bullying [ 43 , 70 , 172 ] and emotional abuse [ 71 ], which was also perpetrated by men.
Medical residents appear to be particularly vulnerable to Type III-B violence, with more than 60% of studies [ 41 , 43 , 64 , 66 , 67 , 71 , 172 , 175 ] reporting this type of violence in medical residency programs. Furthermore, several studies highlighted that the perpetrator of sexual harassment was most often of the opposite sex [ 63 , 64 , 66 ]. For instance, Freedman-Weiss et al. [ 66 ] reported that male residents experienced 65.9% of harassment from men compared to 81.8% from women. On the other hand, female residents reported experiencing more harassment from men (97.7%) compared to women (42.4%). In the same study, the main perpetrators for female resident victims were attending physicians (72.9%), followed by nurses (68.5%), senior colleagues (44.7%) and same-level residents (23.5%). Among male residents, nurses were the most common perpetrator of WPV (69%), followed by attending physicians (62%), senior colleagues (41.9%) and same-level residents (25.6%).
Healthcare professionals in lower hierarchical positions, such as nurses and residents, often contend with stressful conditions and managerial or administrative abuse and harassment, posing challenges to patient care, institutional integrity, and the healthcare system. These experiences also detrimentally impact the victims’ health and career progression. For instance, Tekin and Bulut [ 51 ] found that Turkish nurses who experienced Type III-B violence reported feelings of anger, humiliation, confusion and sadness. Moreover, these experiences also led to strained relationships with others, decreased performance, and caused them to consider leaving the profession. Although this study did not specify the gender of the offender, women experienced significantly higher verbal abuse. The highest perpetuation for all forms of abuse, including verbal (85.7%), physical (46.4%) and sexual (94.4%), was from physicians. In these cases, gender and status within the organizational hierarchy played a critical role in perpetuating Type III-A and III-B WPV, which requires serious attention from employers and health organizations to address GB-WPV through a gender-sensitive approach.
Our examination explores the complexities of gender dynamics concerning both the perpetrator and the victims of workplace violence within the global healthcare community, mainly focusing on nurses and physicians. While 178 studies provided information about perpetrators and sex-segregated findings for workplace violence, only 34 studies (19%) reported the gender of the perpetrator for Type II and Type III violence. These findings provided insights into how gender and an individual’s position within the organization create unique vulnerabilities to WPV. The consequences of such violence against health workers not only affect patient care but also have broader implications for healthcare organizations and workforce landscapes. In our review, men were found to be the primary instigators, accounting for 65% of incidences of WPV, while women were responsible for 28% of instances. Both men and women perpetrated the remaining 7% of incidents. Additionally, our analysis identified distinctive behaviour patterns among male and female offenders. Recognizing that each type of violence requires a different approach for its management and prevention, we will discuss the divergent behavioural patterns of men and women perpetrators of Type II and Type III violence. We examine the underlying factors contributing to these differences and discuss the implications of adopting gender-sensitive approaches to prevent and manage GB-WPV.
Of the 34 studies that provided the gender of perpetrators for any type/s of violence, the majority (74%, n = 25) reported on Type II WPV perpetrated by patients, their families, or visitors. In this context, male perpetrators were more prevalent, targeting both nurses (77.9%) and physicians (70%). The majority of studies that reported on Type II violence indicated a higher prevalence of various forms of violence against female nurses and physicians. The higher perpetration of WPV by men can be linked to societal norms associating aggression and dominance with masculinity [ 193 ]. At the same time, violence against a feminized nursing workforce is normalized as part of the job [ 24 , 75 , 98 , 193 ]. This link between societal norms and assigned roles was evident in several studies [ 76 , 125 ], which is deliberated in the following section.
Type II violence typically targets healthcare providers in the performance of their professional duties and is characterized by acts of physical violence [ 6 , 35 , 44 , 61 – 63 , 166 ]; verbal violence [ 6 , 35 , 44 , 53 , 63 ], and sexual harassment/ violence [ 8 , 41 , 44 , 63 , 64 , 167 ]. Most of these studies reported a higher prevalence of WPV for women for all forms of violence [ 8 , 9 , 44 , 53 , 62 , 64 , 169 , 174 , 176 ]. The social norms, which stem from social relations dictate gender roles and responsibilities, and healthcare institutions are no exception to these forces. For example, a study conducted in Italy that included all areas of practice and the entire health workforce, investigating determinants of aggression against the health workforce reported women were 1.37 times more likely to experience aggression from consumers and colleagues. In this study, nurses experienced the highest number of episodes of violence (64%). Most of these aggressive acts occurred during assistance and supportive care to patients (38%) [ 125 ]. On the other hand, men were not immune to WPV, particularly physical [ 44 , 61 , 166 , 185 ] and both physical and verbal violence in the emergency department in Saudi Arabia, Turkey and China [ 37 , 59 , 189 ]. In Turkey, male physicians experienced higher violence (62.4%) in contrast to their female counterparts (37.6%) [ 59 ]. A similar pattern emerged in Saudi Arabia, with male physicians and nurses reporting a higher prevalence (57.8%), than their female counterparts (42.8%) [ 37 ]. These three studies identified several factors for the high occurrence of WPV from patients and their relatives, including dissatisfaction with the treatment, long wait times and lack of staff [ 37 , 59 ], overcrowding and lack of security [ 37 ]. Though these highlighted factors are important to explain the occurrence of workplace violence for both men and women in the workforce, in the Saudi context, culture seems to have a protective factor for women, where public abuse from men is socially unacceptable [ 88 ]. Similarly, three other studies in Jordan attributed the higher prevalence among male physicians to culture and the existence of laws that intensify legal penalties against women abusers [ 87 ], the cultural norm of altruism and tolerance towards females, particularly physical violence [ 42 ], and a lack of encouragement for reporting WPV by females as part of the male-dominant culture [ 150 ]. Additionally, the higher occurrence of physical violence for men can also be explained by the cultural expectation of masculinity.
In contrast, women’s experience of severe sexual harassment was associated with pregnancy, family responsibilities, and occupational segregation [ 63 ]. Newman et al. [ 63 ] explained that occupational segregation also creates a vertical hierarchy where women are assigned to lower-level tasks (typically front-line care providers). The WHO report analyzed gender and equity in the health and social workforce ‘delivered by women, led by men’ (2019) and acknowledged occupational segregation as universal, which is reinforced by the broader societal norms and creates discriminatory practices with regard to gender and occupational roles [ 194 ]. In these lower positions, women experience sexual harassment from male colleagues, male patients and community members [ 16 , 194 ]. Considering the prevalence of Type II violence for both men and women linked to sex-segregated responsibilities and societal structures. Jafree [ 195 ] calls on policymakers to ensure security and protection for the health workforce, particularly women; legislative reforms for healthcare governance and zero-tolerance policies for violence were also recommended. Several other sources, too, advocate for zero-tolerance policies and emphasize the need for a managerial approach that takes all complaints seriously, reports investigation outcomes, and enforces sanctions to eliminate impunity [ 9 , 92 , 131 ]. Collaborative community efforts are required to acknowledge and alter the patriarchal culture and reduce violence against women by creating awareness about the public role through various forums, including the media [ 28 , 79 , 94 , 195 ].
Several contributing factors have been identified in the context of Type II WPV, such as noise levels, inadequate communication skills [ 74 ], perceived/actual staff incompetence or unsympathetic attitudes, dissatisfaction with service provision, prolonged wait times, and poor communication [ 53 , 196 ]. These circumstances can escalate emotions and increase the likelihood of violent encounters. Furthermore, specific treatment specialties, such as emergency departments [ 35 , 75 , 191 ], psychiatric units [ 76 , 77 ], and geriatric care [ 26 , 76 ], have demonstrated a higher risk of Type II workplace violence. Factors specific to these settings include a lack of privacy and personal space, unrealistic expectations of clients, insufficient staffing and resources, poor staff skills mix, healthcare systems and processes not understood by clients, perceived favouritism, overcrowding in emergency departments, delays in providing analgesia, and inflexible visiting hours [ 196 ]. These challenges, compounded by a shortage of skilled professionals, unclear expectations and communication, scheduling issues, and environmental stressors can generate increased stress and, thus, uncertainty. Addressing these factors constitutes the initial step in decreasing or eliminating the risk of violence for both men and women [ 197 ].
Both primary research and systematic reviews have acknowledged the difficulty associated with addressing multifactorial violence, given the diversity in population and setting and the types/classifications of violence [ 94 , 95 , 102 , 198 ]. However, these sources did not provide information about perpetuators, particularly gendered nature. For instance, a recent umbrella review examined 32 systematic reviews for WPV prevalence and characteristics. This comprehensive assessment reported that the overall prevalence from the meta-analysis of 11 reviews was 57.9%, ranging from 34.1% to 78.9% among healthcare providers and most affected were nurses working in psychiatric wards [ 198 ]. This prevalence aligns with the findings of this review. Of note, the umbrella review too did not provide information on perpetrators and prevalence based on gender and stated that the included reviews had reported variable results for men and women; however, it did underscore how gender imbalances in emergency departments could increase the risk of violence among women. Several studies in our review recommended ensuring gender equality in the health workforce and leadership positions to reduce the prevalence of WPV among women [ 9 , 30 , 63 , 80 ].
Type III-A (Horizontal or lateral) workplace violence perpetrated by one healthcare worker against another may stem from interpersonal conflicts, workplace stress [ 12 ], or other factors contributing to a hostile work environment. Among studies that provided data on Type III-A violence, most perpetrators were men (63.5%) compared to women (23%). Horizontal WPV was reported more frequently by physicians [ 41 , 43 , 65 – 69 ] than among nurses [ 40 , 53 ]. The studies that sampled both nurses and physicians [ 29 , 37 , 44 , 60 , 63 ] also reported that men perpetuated all forms of violence in most cases for both male and female victims [ 37 , 44 , 63 ]. In some instances, women experienced violence from both men and women [ 63 ].
Type III violence is also rooted in cultural norms and societal expectations that allocate roles and responsibilities based on gender [ 63 ]; in most cultures, women are responsible for childbearing and rearing and men hold decision-making positions. This phenomenon transcends the household and is also seen in the workplace and healthcare institutions [ 9 , 68 , 78 ]. These gendered roles and responsibilities often position men in leadership positions while women are assigned to caring roles with less authority and responsibility, perpetuating discriminatory practices that negatively impact women [ 9 , 63 , 70 ]. This dynamic prevails in both wealthy and lower- and middle-income countries in varied behaviors. For instance, in Australia and New Zealand, women experienced significantly higher discrimination (31% vs. 8%) and sexual harassment (23% vs. 0.5%) than men, primarily due to family responsibilities, lack of mentorship and rigid promotion criteria [ 70 ]. In Rwanda, women’s experiences of childbearing and care, including managing pregnancy, motherhood and work, and the widespread negative stereotyping of women at work led to discrimination that co-occurred with sexual harassment within health workplaces [ 63 ]. Jacobson et al. [ 12 ] report on Type III-A violence in medical residency programs, and women experienced a significantly higher frequency of work-related incidents from colleagues and support staff, explaining the higher workload for women due to the coexistence of family responsibilities. Additionally, relational and managerial issues, including organizational affairs within large, complex health organizations, shifting duties and cohabitation of various teams on the same unit, were identified as factors contributing to Type III-A violence in Italy [ 53 ].
This type of interpersonal violence, including violence against women, is prevalent in science, technology, engineering and math (STEM), which are considered male-dominant disciplines [ 199 , 200 ], unlike healthcare, where 70% of the workforce globally are women and higher rates of violence are associated with their roles and responsibilities and the gendered workplace hierarchy [ 194 ]. In STEM, violence against women can be explained by the backlash effect, in which gender equality is associated with higher prevalence [ 200 ].
Given the social reality of women’s lives and career development in healthcare, flexible human resource development and management policies could empower women to balance their work and family responsibilities. Zampieroni et al. [ 53 ] recommend adopting realistic workloads and skill-mixed staffing, promoting gender equality in staff allocation, and participatory leadership to overcome relational conflict and managerial actions that enhance working conditions. Nurse managers must play the role of cultural gatekeepers, hold individuals accountable and foster staff empowerment; utilizing research-informed methods such as ‘cognitive rehearsal and crucial conversations’ [ 20 ] and conducting team-building workshops will assist in mitigating the impact of horizontal violence [ 21 ].
Type III-B (vertical) violence is primarily perpetrated by senior colleagues, supervisors, and administrative personnel occupying higher positions in the organizational hierarchy than the victim. Among the 34 studies, 66% reported perpetrators’ gender for Type III-B violence; men perpetuated 77% among nurses and 67.5% among physicians. The causative factors for vertical violence included organizational structure, leadership and administrative authorities, and power struggles in the health workplaces. These factors not only perpetuated WPV but also prohibited reporting of the instances due to fear of reprisal [ 29 , 63 , 66 ]. Two prevalent forms of violence linked to hierarchical/ organizational structure were sexual harassment and bullying/mobbing. The majority of studies reporting Type III-B violence reported sexual violence from male supervisors and administrators [ 8 , 41 , 44 , 63 – 69 ], particularly in medical residency programs—placing these trainee residents in a vulnerable position [ 41 , 66 , 67 ]. Additionally, vertical violence was the only type reported to be perpetuated by women at higher levels in the organizational hierarchy, particularly bullying (women: 37.9% vs. men: 10.5%) among nurses and physicians [ 29 ]. Additionally, two studies reported higher rates of mobbing behaviours by women than men among healthcare professionals, including nurses and physicians [ 72 , 192 ].
Type III-B violence is emblematic of the hierarchical and inflexible organizational culture historically dominated by male medical professionals. This stemmed from beliefs and negative stereotypes, such as women being weak, unwilling to speak up, indecisive and incompetent [ 63 ]. Additionally, perceived competence was expressed as a predictor for bullying among women [ 42 , 153 ]. Such perceptions reinforce the structural power held by men, particularly with male managers and physicians. The patriarchal institutional structures provide power domination among women as well, who could use their power to oppress individuals under their control. A qualitative study [ 201 ] from Estonia exposed this dynamic of domination and sexual harassment among nurses; it highlighted the association between power and the use of sexualized language. A female nurse stated, “I am more disturbed by their patronizing behaviour"; the nurse characterizes physicians’ attitude as: “I am a man, I am a physician, I can do and say whatever comes to my mind” (Nurse 18, p.30). Lamesoo [ 201 ] further explained that nurses placed themselves in the hospital hierarchy between physicians and patients and acknowledged that they could not challenge a physician’s incivility. However, these nurses can easily ask patients to refrain from such behaviour without hesitation because patients have less power than nurses, and patients are expected to follow hospital rules [ 201 ] dictated by nurses. These instances explain organizational power as a protective factor for offenders. However, women’s underrepresentation in positions of power places them in a vulnerable position.
Another qualitative study in Uganda by Newman et al. [ 9 ] reported from key informant interviews in the Uganda health system that "we have women over-represented in the bottom of any organization and for the men, it is an upward or inverted pyramid whereby as you go up the power ladder…. There is a tendency to abuse that power and they don’t even think that they are abusing it because they have grown up thinking they may be flattering the women…". The authors further stated that "Sexual coercion started during recruitment of health workers and continued after hiring, perpetrated by men in hierarchically superior decision-making positions supervisors, senior managers (including human resources) or medical superintendents” [ 9 ]. These severe human rights violations necessitate a transformation in the mindset of individuals in the workforce and a cultural shift at organizational levels to rectify the dominant, hierarchical and permissive environment [ 65 ]. Ensuring gender equality at the upper echelons of healthcare organizations and in decision-making positions is crucial to establishing a secure and equitable environment for all, regardless of gender. A scoping review of three evidence-based guidelines and 33 systematic reviews on strategies to prevent and manage WPV in healthcare settings reported a correlation between strong leadership to cultivate and enforce a culture of inclusivity, support and respect as a prerequisite for successful prevention of WPV [ 202 ]. Therefore, healthcare organizations’ leadership must proactively seek organizational solutions to end gender-based WPV and prioritize gender equality and protecting employees’ rights as part of their human resources for health (HRH) policy [ 9 ].
Sexual harassment in academia was found to be an issue across various contexts, particularly among women in medical residency programs. A study [ 78 ] in a U.S. medical college reported that one-third of respondents experienced sexual harassment, including medical students (51.7%), residents/fellows (31%) and faculty members (25%), which was inversely proportional to their position in the program. Similarly, sexual harassment was more prevalent among women in vascular surgery in the U.S. [ 67 ], ophthalmology in Australia and New Zealand [ 79 ] and cardiothoracic surgery, reported by a global survey [ 28 ], and rates of sexual harassment in almost all contexts were higher among female trainees. In one instance, in the U.S., male (70%) and female (69%) residents [ 41 ] in obstetrics and gynecology residents experienced sexual harassment at similar levels [ 41 ]. Additionally, one study in the U.S. with a large, representative sample (n = 6000) from a national survey reported that higher women’s representation within a specialty was associated with lower sexual harassment for both men and women from coworkers and patients [ 80 ]. This observation held true in the Canadian context where reporting of sexual harassment incidents was low (2.9%) in a study with female participants constituting 53% of the sample [ 33 ]. These women did report slightly higher rates of intimidation, harassment, and discrimination (IHD) based on gender (males 40.4%; females 48.0%). Hence, findings underscore the recurring recommendation of gender equality in the health workforce and leadership positions and the role of leadership in preventing Type II and Type III violence, including harassment.
Acknowledging sexual harassment as a prevalent problem is the crucial initial step in formulating a successful strategy to prevent its occurrence [ 65 , 203 ]; a comprehensive strategy should encompass a zero-tolerance statement across the specialty with a transparent and fair mechanism for reporting sexual harassment [ 65 , 78 ]. Moreover, it is essential to provide trainees with both direct face-to-face and electronic routes for anonymous and confidential reporting to alleviate concerns related to personal reattribution and academic detriment [ 64 , 78 ]. Standardized, transparent reporting mechanisms with well-delineated consequences for the offender must be established. Additionally, the institutions should ensure the availability of links to all the required resources is the first hit on online searches, displaying posters/presentations/ads [ 78 ].
Recognizing harassment as an institutional structural issue, senior leadership can have a protective role by serving as role models. A qualitative study conducted in Germany [ 197 ] representing women nurses (50%) and physicians (50%) explored preventive options for sexual harassment in academia. The findings revealed that leadership commitment and clear statements can significantly influence multiple levels by demonstrating openness to address taboo topics, raise awareness, and place the issue at the decision table. A participant stated, "A culture of political correctness is communicated from the top down, with the management committee and senior management acting as role models” (p.12). Another participant stated, "It is the senior staff that creates a team culture that should be supportive and transparent, with clear boundaries… .. I have an open door and open eyes policy and try to initiate rituals that allow us to work together in the correct way” (p.12). While commitment and stated actions are essential, meaningful cultural change necessitates the consistent, active, structured, and continued engagement of all health workforce members, including students and trainees, staff and especially from senior leadership. Senior leadership must be actively engaged in this process, particularly male leaders. Therefore, engaging individuals at various levels in open, nonjudgmental conversations is paramount to breaking the silence [ 30 ] and ingraining these principles into the organizational culture.
First, in our comprehensive review of workplace violence (WPV), not all studies reported on perpetrators of WPV. Therefore, we included all studies that indicated perpetrators/ sources of violence. We categorized these sources into distinctive categories of Type II and Type III WPV. Limitations to this approach include the heterogeneity of the forms of violence reported by the included studies according to gender. While studies reported victims’ exposure to Types II, and/or III, the gender of perpetrators in each case was not specified. As a result, we presented the prevalence of the various forms and categorized the perpetrators’ type for all the studies (178) in Table 2 . The final set of studies (19%) that reported on the gender of the perpetrators was analyzed. Since fewer studies provided information about the gender of perpetrators across the types/forms of violence, future research must focus on conducting and reporting gender-segregated findings for perpetrators that will strengthen recognition of the gender-based WPV and could lead to gender-sensitive strategies at the local and international levels. Another limitation of our review was that most of the included studies operationalized gender as a binary. A few studies included either non-binary (less than 4%) [ 80 , 98 ] individuals or mentioned as others (less than 4%) [ 28 , 30 ] or unknown (less than 9%) [ 85 , 128 ], in the analysis of the total population, reported in Table 2 . Even these studies did not report findings for those minority populations or address it as a limitation. Therefore, we reported findings based on gender binary. All these studies, which represented non-binary individuals, were conducted in the USA [ 30 , 80 , 98 , 128 ] and Canada [ 85 ]; in these contexts, gender diversity and inclusion are acknowledged as compared to most Low-and -middle-income countries where sex is equated with gender. These studies did not recognize it as a limitation; only one study, which reported on survey data from the Association of American Medical Colleges (AAMC) National Sample Survey of Physicians (NSSP), expressed excluding the non-binary data because of the lower sample [ 80 ]. Considering this limitation, we recommend that future research include gender-diverse populations.
The review revealed a higher prevalence of Type II and Type III WPV among women compared to men. In parallel, it was observed that men predominantly perpetrated all forms of violence against both men and women healthcare providers. Only Type III-B violence, including bullying/ mobbing, was occasionally perpetuated by women. Both Types II and Type III violence have roots in societal structures, and women were more frequently victimized. This increased victimization of women can be attributed to their lower status in society and in the healthcare settings that assign roles and responsibilities based on this status. Additionally, women’s reproductive realities, including managing pregnancy, motherhood and work, and widespread negative stereotyping contributed to their vulnerability to gender-based WPV.
Conversely, men’s domination in leadership, decision-making and supervisory positions in most contexts creates a hierarchical and permissive environment that perpetuates violence against women. Therefore, understanding gender implications concerning both the victim and perpetrator among the critical health workforce of nurses and physicians across the globe is essential. Healthcare organizations and professional stakeholders must seriously consider zero-tolerance policies, transparent mechanisms for handling violent incidents, and the provision of appropriate support to victims. These measures will empower individual professionals, enhance patient care, and positively impact healthcare institutions and society as a whole.
S1 checklist. prisma-scr checklist..
https://doi.org/10.1371/journal.pgph.0003646.s001
https://doi.org/10.1371/journal.pgph.0003646.s002
https://doi.org/10.1371/journal.pgph.0003646.s003
https://doi.org/10.1371/journal.pgph.0003646.s004
https://doi.org/10.1371/journal.pgph.0003646.s005
https://doi.org/10.1371/journal.pgph.0003646.s006
You are accessing a machine-readable page. In order to be human-readable, please install an RSS reader.
All articles published by MDPI are made immediately available worldwide under an open access license. No special permission is required to reuse all or part of the article published by MDPI, including figures and tables. For articles published under an open access Creative Common CC BY license, any part of the article may be reused without permission provided that the original article is clearly cited. For more information, please refer to https://www.mdpi.com/openaccess .
Feature papers represent the most advanced research with significant potential for high impact in the field. A Feature Paper should be a substantial original Article that involves several techniques or approaches, provides an outlook for future research directions and describes possible research applications.
Feature papers are submitted upon individual invitation or recommendation by the scientific editors and must receive positive feedback from the reviewers.
Editor’s Choice articles are based on recommendations by the scientific editors of MDPI journals from around the world. Editors select a small number of articles recently published in the journal that they believe will be particularly interesting to readers, or important in the respective research area. The aim is to provide a snapshot of some of the most exciting work published in the various research areas of the journal.
Original Submission Date Received: .
Find support for a specific problem in the support section of our website.
Please let us know what you think of our products and services.
Visit our dedicated information section to learn more about MDPI.
Literature review on collaborative project delivery for sustainable construction: bibliometric analysis.
2. literature review, 2.1. collaborative project delivery, 2.2. design build (db), 2.3. construction manager at risk (cmar), 2.4. integrated project delivery method (ipd), 2.5. sustainability, 2.6. sustainable construction, 2.7. benefits of eci comparing case studies, 2.8. collaborative delivery models, 3. methodology, 3.1. research methods, 3.2. database research, 4.1. ipd, design-build, and cmar overview, 4.1.1. yearly publication distribution of db cmar and ipd, 4.1.2. major country analysis, 4.1.3. most relevant and influential journals, 4.1.4. corresponding author countries, 4.2. keyword analysis, 4.2.1. high-frequency keyword analysis, 4.2.2. co-occurrence network analysis, 4.2.3. analysis of keywords’ frequency over time, 5. discussion, 5.1. findings of advantages and disadvantages of ipd, db, and cmar for sustainable construction, 5.1.1. advantages of ipd, 5.1.2. advantages of design-build, 5.1.3. advantages of construction manager at risk, 5.1.4. disadvantages of ipd, 5.1.5. disadvantages of design-build, 5.1.6. disadvantages of construction manager at risk, 5.2. most suitable cpd technique for sustainable construction based on literature review, 5.2.1. limitations, 5.2.2. recommendations for future research, 6. future trend, 6.1. enhancing innovation through collaborative project delivery, 6.2. open communication and block chain technology, 6.3. multi-party agreement, 6.4. utilizing artificial intelligence in decision support systems, 7. conclusions, author contributions, institutional review board statement, informed consent statement, data availability statement, conflicts of interest.
Click here to enlarge figure
Paper | Reference | Total Citation TC | TC Per Year | Normalized TC |
---|---|---|---|---|
Kent D.C., 2010, J Constr Eng Manage | (Kent and Becerik-Gerber, 2010) [ ] | 300 | 21.43 | 7.67 |
Ugwu O.O., 2007, Build Environ | (Ugwu and Haupt, 2007) [ ] | 269 | 15.82 | 7.69 |
Kines P., 2010, J Saf Res | (Kines et al., 2010) [ ] | 238 | 17.00 | 6.08 |
Asmar M., 2013, J Constr Eng Manag | (Asmar et al., 2013) [ ] | 226 | 20.55 | 5.01 |
Ballard G., 2008, Lean Constr J | (Ballard, 2008) [ ] | 221 | 13.81 | 6.85 |
Hale D.R., 2009, J Constr Eng Manag | (Hale et al., 2009) [ ] | 211 | 14.07 | 6.95 |
Bynum P., 2013, J Constr Eng Manag | (Bynum et al., 2013) [ ] | 185 | 16.82 | 4.11 |
Ibbs C.W., 2003, J Constr Eng Manag | (Ibbs et al., 2003) [ ] | 183 | 8.71 | 8.58 |
Choudry R.M., 2009, J Constr Eng Manag | (Choudhry et al., 2009) [ ] | 182 | 12.13 | 6.00 |
Mollaoglu-Korkmaz S., 2013, J Manage Eng | (Mollaoglu-Korkmaz et al., 2013) [ ] | 152 | 13.82 | 3.37 |
El Wardani M.A., 2006, J Constr Eng Manag | (El Wardani et al., 2006) [ ] | 144 | 8.00 | 4.65 |
Ghassemi R., 2011, Lean Constr J | (Ghassemi and Becerik-Gerber, 2011) [ ] | 143 | 11.00 | 5.54 |
Liu J., 2016, J Constr Eng Manag | (Liu et al., 2016) [ ] | 140 | 17.50 | 5.12 |
El-Sayegh S.M., 2015, J Manag Eng | (El-Sayegh and Mansour, 2015) [ ] | 135 | 15.00 | 6.59 |
Fang C., 2012, Reliab Eng Syst Saf | (Fang et al., 2012) [ ] | 131 | 10.92 | 4.05 |
Franz B., 2017, J Constr Eng Manag | (Franz et al., 2017) [ ] | 126 | 18.00 | 5.56 |
Kim H., 2016, J Comput Civ Eng | (Kim et al., 2016) [ ] | 125 | 15.63 | 4.57 |
Ding L.Y., 2013, Autom Constr | (Ding and Zhou, 2013) [ ] | 118 | 10.73 | 2.62 |
Wanberg J., 2013, J Constr Eng Manag | (Wanberg et al., 2013) [ ] | 116 | 10.55 | 2.57 |
Shrestha, P.P., 2012, J Constr Eng Manag | (Shrestha et al., 2012) [ ] | 112 | 9.33 | 3.47 |
Torabi S.A., 2009, Int J Prod Res | (Torabi and Hassini, 2009) [ ] | 105 | 7.00 | 3.46 |
Baradan S., 2006, J Constr Eng Manag | (Baradan and Usmen, 2006) [ ] | 99 | 5.50 | 3.20 |
Levitt R.E., 2007, J Constr Eng Manag | (Levitt, 2007) [ ] | 97 | 5.71 | 2.77 |
Sullivan J., 2017, J Constr Eng Manag | (Sullivan et al., 2017) [ ] | 93 | 13.29 | 4.11 |
Araya F., 2021, Saf Sci | (Araya, 2021) [ ] | 92 | 30.67 | 9.5 |
Country | Frequency |
---|---|
USA | 584 |
CHINA | 167 |
UK | 101 |
AUSTRALIA | 71 |
SOUTH KOREA | 56 |
CANADA | 51 |
IRAN | 39 |
MALAYSIA | 39 |
INDIA | 30 |
SOUTH AFRICA | 22 |
SPAIN | 22 |
FINLAND | 18 |
FRANCE | 17 |
DENMARK | 16 |
EGYPT | 16 |
SWEDEN | 16 |
INDONESIA | 15 |
NETHERLANDS | 14 |
NEW ZEALAND | 14 |
BRAZIL | 13 |
GERMANY | 13 |
NIGERIA | 13 |
UNITED ARAB ENIRATES | 13 |
JORDAN | 12 |
SAUDI ARABIA | 12 |
Country | TC | Average Article Citations |
---|---|---|
USA | 4933 | 23.70 |
CHINA | 1106 | 18.10 |
UNITED KINGDOM | 763 | 19.10 |
HONG KONG | 703 | 37.00 |
AUSTRALIA | 494 | 21.50 |
SOUTH KOREA | 312 | 16.00 |
IRAN | 198 | 52.00 |
SPAIN | 191 | 15.20 |
SWEDEN | 188 | 21.20 |
PAKISTAN | 182 | 20.90 |
FRANCE | 164 | 182.00 |
UNITED ARAB EMIRATES | 163 | 32.80 |
MALAYSIA | 154 | 32.60 |
INDIA | 145 | 15.40 |
SINGAPORE | 130 | 13.20 |
CANADA | 107 | 43.30 |
ITALY | 92 | 7.60 |
LEBANON | 92 | 18.40 |
NETHERLANDS | 91 | 18.40 |
NORWAY | 74 | 18.20 |
IPD Advantages | ||
---|---|---|
Advantages | % Percentage of Advantages from Ordered List of Publication | Publication List |
Collaborative atmosphere and fairness | 79 | B = [ ] C = [ ] D = [ ] E = [ ] F = [ ] G = [ ] H = [ ] I = [ ] J = [ ] K = [ ] L = [ ] M = [ ] N = [ ] O = [ ] P = [ ] Q = [ ] R = [ ] S = [ ] T = [ ] U = [ ] V = [ ] |
Early involvement of stakeholders | 63 | B = [ ] C = [ ] D = [ ] E = [ ] F = [ ] G = [ ] H = [ ] I = [ ] J = [ ] L = [ ] M = [ ] N = [ ] O U = [ ] V = [ ] W = [ ] |
Promoting trust | 25 | R = [ ] S = [ ] U = [ ] V = [ ] W = [ ] X = [ ] |
Reduce schedule time | 42 | C = [ ] D = [ ] E = [ ] F = [ ] G = [ ] H = [ ] I = [ ] J = [ ] S = [ ] T = [ ] |
Reduce waste | 42 | C = [ ] D = [ ] E = [ ] F = [ ] G = [ ] H = [ ] I = [ ] J = [ ] S = [ ] T = [ ] |
Shared cost, risk reward, and responsibilities | 75 | C = [ ] D = [ ] E = [ ] F = [ ] G = [ ] H = [ ] I = [ ] J = [ ] S = [ ] T = [ ] U = [ ] V = [ ] W = [ ] X = [ ] |
Multi-party agreement and noncompetitive bidding | 54 | C = [ ] D = [ ] E = [ ] F = [ ] G = [ ] H = [ ] I = [ ] J = [ ] K = [ ] N = [ ] Q = [ ] T = [ ] V = [ ] |
Integrated decision-making for designs and shared design responsibilities | 38 | C = [ ] D = [ ] E = [ ] H = [ ] I = [ ] J = [ ] L = [ ] P = [ ] T = [ ] |
Open communication and time management | 38 | D = [ ] E = [ ] F = [ ] O = [ ] R = [ ] S = [ ] T = [ ] U = [ ] V = [ ] |
Reduce project duration and liability by fast-tracking design and construction | 25 | F = [ ] G = [ ] L = [ ] O = [ ] S = V |
Shared manpower and changes in SOW, equipment rentage, and change orders | 17 | A = [ ] F = [ ] G = [ ] Q = [ ] |
Information sharing and technological impact | 38 | A = [ ] D = [ ] G = KLMPRV |
Fast problem resolution through an integrated approach | 21 | B = [ ] C = [ ] D = [ ] E = [ ] S = [ ] |
Lowest cost delivery and project cost | 33 | A = [ ] C = [ ] F = [ ] G = [ ] L = [ ] P = [ ] Q = [ ] S = [ ] T = [ ] U = [ ] |
Improved efficiency and reduced errors | 29 | B = [ ] C = [ ] F = [ ] L = [ ] Q = [ ] S = [ ] T = [ ] |
Combined risk pool estimated maximum price (allowable cost) | 17 | A = [ ] L = [ ] P = [ ] Q = [ ] |
Cooperation innovation and coordination | 46 | CEFLPQRSTUV |
Combined labor material cost estimation, budgeting, and profits | 25 | A = [ ] D = [ ] P = [ ] S = [ ] T = [ ] U = [ ] V = [ ] |
Strengthened relationship and self-governance | 17 | C = [ ] D = [ ] F = [ ] |
Fewer change orders, Schedules, and request for information | 21 | L = [ ] O = [ ] Q = [ ] T = [ ] V = [ ] |
Ordered list of publication A = [ ] B = [ ] C = [ ] D = [ ] E = [ ] F = [ ] G = [ ] H = [ ] I = [ ] J = [ ] K = [ ] L = [ ] M = [ ] N = [ ] O = [ ] P = [ ] Q = [ ] R = [ ] S = [ ] T = [ ] U = [ ] V = [ ] W = [ ] X = [ ] |
DB Advantages | ||
---|---|---|
Disadvantages | %Percentage of Advantages from Ordered List of Publication | Publication List |
Single point of accountability for the design and construction | 39 | CDIJMOQRT C = [ ] D = [ ] I = [ ] J = [ ] M = [ ] O = [ ] Q = [ ] R = [ ] T = [ ] |
Produces time saving schedule | 52 | CDHJKLMORSTV C = [ ] D = [ ] H = [ ] J = [ ] K = [ ] L = [ ] M = [ ] O = [ ] R = [ ] S = [ ] T = [ ] V = [ ] |
Cost effective projects | 39 | CKLMNOPQSV C = [ ] K = [ ] L = [ ] M = [ ] N = [ ] O = [ ] P = [ ] Q = [ ] S = [ ] V = [ ] |
Design build functions as a single Entity | 8 | DF D = [ ] F = [ ] |
Enhances quality and mitigates design errors | 21 | F = [ ] J = [ ] S = [ ] V = [ ] W = [ ] F = [ ] |
Facilitates teamwork between owner and design builder | 30 | J = [ ] N = [ ] P = [ ] S = [ ] U = [ ] V = [ ] W = [ ] |
Insight into constructability of the design build contractor (Early involvement of contractor) | 13 | H = [ ] I = [ ] T = [ ] |
Enhances fast tracking | 4 | R = [ ] |
Good coordination and decision-making | 27 | C = [ ] D = [ ] E = [ ] M = [ ] O = [ ] Q = [ ] |
Clients’ owner credibility | 13 | A = [ ] C = [ ] G = [ ] |
Dispute reduction mitigates disputes | 21 | B = [ ] H = [ ] I = [ ] J = [ ] Q = [ ] |
Ordered list of publication A = [ ] B = [ ] C = [ ] D = [ ] E = [ ] F = [ ] G = [ ] H = [ ] I = [ ] J = [ ] K = [ ] L = [ ] M = [ ] N = [ ] O = [ ] P = [ ] Q = [ ] R = [ ] S = [ ] T = [ ] U = [ ] V = [ ] W = [ ] |
CMAR Advantages | ||
---|---|---|
Advantages | Percentage of Advantages from the Ordered List of Publication | Publication List |
Early stakeholder involvement | 31 | H = [ ] I = [ ] L = [ ] M = [ ] O = [ ] |
Fast-tracking cost savings and delivery within budget | 50 | A = [ ] B = [ ] C = [ ] D = [ ] F = [ ] I = [ ] M = [ ] O = [ ] |
Reduce project duration by fast-tracking design and construction | 6 | C = [ ] |
Clients have control over the design details and early knowledge of costs | 50 | B = [ ] C = [ ] D = [ ] H = [ ] I = [ ] K = [ ] M = [ ] P = [ ] |
Mitigates against change order | 50 | A = [ ] C = [ ] E = [ ] H = [ ] I = [ ] K = [ ] M = [ ] P = [ ] |
Provides a GMP by considering the risk of price | 31 | A = [ ] B = [ ] C = [ ] M = [ ] O = [ ] |
Reduces design cost and redesigning cost | 25 | C = [ ] D = [ ] E = [ ] H = [ ] |
Facilitates schedule management | 75 | B = [ ] C = [ ] D = [ ] E = [ ] F = [ ] G = [ ] H = [ ] I = [ ] J = [ ] K = [ ] M = [ ] N = [ ] |
Facilitates cost control and transparency | 69 | C = [ ] D = [ ] E = [ ] F = [ ] G = [ ] H = [ ] I = [ ] J = [ ] K = [ ] M = [ ] N = [ ] |
Single point of responsibility for construction and joint team orientation for accountability | 44 | A = [ ] B = [ ] E = [ ] F = [ ] I = [ ] M = [ ] N = [ ] |
Facilitates Collaboration | 25 | E = [ ] F = [ ] I = [ ] J = [ ] |
Ordered list of publication A = [ ] B = [ ] C = [ ] D = [ ] E = [ ] F = [ ] G = [ ] H = [ ] I = [ ] J = [ ] K = [ ] L = [ ] M = [ ] N = [ ] O = [ ] P = [ ] |
IPD Disadvantages | ||
---|---|---|
Disadvantages | % Percentage of Disadvantages from Ordered List of Publication | Publication List |
Impossibility of being sued internally over disputes and mistrust, alongside complexities in compensation and resource distribution | 42 | C = [ ] E = [ ] F = [ ] I = [ ] L = [ ] |
Skepticism of the added value of IPD and impossibility of owners’ inability to tap into financial reserves from shared risk funds | 50 | E = [ ] F = [ ] G = [ ] J = [ ] K = [ ] L = [ ] |
Difficulty in deciding scope | 17 | A = [ ] H = [ ] |
Difficulty in deciding target cost/Budgeting | 25 | A = [ ] D = [ ] H = [ ] |
Adversarial team relationships and legality issues | 50 | B = [ ] C = [ ] D = [ ] F = [ ] K = [ ] L = [ ] |
Immature insurance policy for IPD and uneasiness to produce a coordinating document | 25 | A = [ ] J = [ ] K = [ ] |
Fabricated drawings in place of engineering drawings because of too early interactions | 8 | F = [ ] |
High initial cost of investment in setting up IPD team and difficulty in replacing a member of IPD team | 16 | J = [ ] L = [ ] |
Inexperience in initiating/developing an IPD team and knowledge level | 16 | K = [ ] L = [ ] |
Low adoption of IPD due to cultural, financial, and technological barriers | 33 | E = [ ] F = [ ] K = [ ] L = [ ] |
High degree of risks amongst teams coming together for IPD and owners responsible for claims, damages, and expenses (liabilities) | 25 | D = [ ] F = [ ] L = [ ] |
Issues with poor collaboration | 8 | H = [ ] |
Non-adaptability to IPD environment | 42 | E = [ ] G = [ ] J = [ ] K = [ ] L = [ ] |
Ordered list of publication A = [ ] B = [ ] C = [ ] D = [ ] E = [ ] F = [ ] G = [ ] H = [ ] I = [ ] J = [ ] K = [ ] L = [ ] |
DB Disadvantages | ||
---|---|---|
Disadvantages | Percentage of Disadvantages from Ordered List of Publication | Publication List |
Non-competitive selection of team not dependent on best designs of professionals and general contractors | 35 | B = [ ] C = [ ] D = [ ] E = [ ] G = [ ] I = [ ] J = [ ] K = [ ] L = [ ] M = [ ] O = [ ] P = [ ] Q = [ ] R = [ ] S = [ ] |
Deficient checks, balances, and insurance among the designer, general contractor, and owner | 30 | A = [ ] B = [ ] C = [ ] D = [ ] E = [ ] F = [ ] G = [ ] H = [ ] I = [ ] J = [ ] L = [ ] M = [ ] N = [ ] U = V |
Unfair allocation of risk and high startup cost | 40 | R = [ ] C = [ ] S = [ ] |
Architect/Engineer(A/E) not related to clients/owners with no control over the design requirements. A/E has less control or influence over the final design and project requirements | 60 | C = [ ] D = [ ] E = [ ] F = [ ] G = [ ] H = [ ] I = [ ] J = [ ] S = [ ] |
Owner cannot guarantee the quality of the finished project | 35 | C = [ ] D = [ ] E = [ ] F = [ ] G = [ ] H = [ ] I = [ ] J = [ ] S = [ ] |
Difficulty in defining SOW, and alterations in the designs after the contract and during construction with decrease in time | 35 | C = [ ] D = [ ] E = [ ] F = [ ] G = [ ] H = [ ] I = [ ] J = [ ] K = [ ] M = [ ] N = [ ] |
Difficulty in providing track record for design and construction | 40 | C = [ ] D = [ ] E = [ ] F = [ ] G = [ ] H = [ ] I = [ ] J = [ ] K = [ ] N = [ ] |
Discrepancy in quality control and testing intensive of owner’s viewpoint | 25 | C = [ ] D = [ ] E = [ ] H = [ ] I = [ ] J = [ ] K = [ ] N = [ ] |
Delay in design changes, inflexibility, and the absence of a detailed design | 35 | D = [ ] E = [ ] F = [ ] O = [ ] R = [ ] S = [ ] |
Owner/client needs external support to develop SOW/preliminary design of the project | 10 | E = [ ] F = [ ] L = [ ] O = [ ] S = [ ] |
Increased labour costs and tender prices | 5 | A = [ ] F = [ ] G = [ ] Q = [ ] |
Guaranteed maximum price is established with Incomplete designs and work requirement | 25 | A = [ ] D = [ ] G = [ ] K = [ ] L = [ ] M = [ ] P = [ ] R = [ ] |
Responsibility of contractor for omission and changes in design | 20 | A = [ ] B = [ ] C = [ ] D = [ ] S = [ ] |
Ordered list of publication A = [ ] B = [ ] C = [ ] D = [ ] E = [ ] F = [ ] G = [ ] H = [ ] I = [ ] J = [ ] K = [ ] L = [ ] M = [ ] N = [ ] O = [ ] P = [ ] Q = [ ] R = [ ] S = [ ] |
CMAR Disadvantages | ||
---|---|---|
Disadvantages | % Percentage of Advantages from Ordered List of Publication | Publication List |
Unclear definition and relationship of roles and responsibilities of CM and design professionals | 78 | A = [ ] B = [ ] C = [ ] D = [ ] G = [ ] H = [ ] I = [ ] |
Difficult to enforce GMP, SOW, and construction based on incomplete documents | 67 | A = [ ] D = [ ] E = [ ] G = [ ] H = [ ] I = [ ] |
Not suitable for small projects or hold trade contractors over GMP tradeoffs and prices | 56 | B = [ ] C = [ ] G = [ ] H = [ ] I = [ ] |
Improper education on CMAR methodology, polices, and regulations | 56 | E = [ ] F = [ ] G = [ ] H = [ ] I = [ ] |
Knowledge, conflicts, and communication issues between the designer and the CM | 56 | B = [ ] E = [ ] F = [ ] G = [ ] H = [ ] |
Shift of responsibilities (including money) from owners/clients to CM | 44 | A = [ ] B = [ ] E = [ ] I = [ ] |
Additional cost due to design and construction and design defects | 56 | A = [ ] C = [ ] D = [ ] G = [ ] H = [ ] |
Inability of CMAR to self-perform during preconstruction | 11 | C = [ ] |
Disputes/issues concerning construction quality and the completeness of the design | 22 | A = [ ] D = [ ] |
No information exchange/alignment between the A/E with the CMAR | 11 | A = [ ] |
Ordered list of publication A = [ ] B = [ ] C = [ ] D = [ ] E = [ ] F = [ ] G = [ ] H = [ ] I = [ ] |
Critical Success Factors for Sustainable Construction | ||
---|---|---|
Advantages | Percentage of Advantages from Ordered List of Publication % | Publication List |
Collaborative atmosphere | 47 | A = [ ] C = [ ] G = [ ] H = [ ] K = [ ] N = [ ] O = [ ] |
Early stakeholder involvement | 26 | N = [ ] J = [ ] I = [ ] |
Reduce design errors | 13 | N = [ ] O = [ ] |
Cost savings and delivery within budget/Client representative | 33 | ABCEF A = [ ] B = [ ] C = [ ] |
Influence of client | 13 | B = [ ] J = [ ] |
Ordered list of publication A = [ ] B = [ ] C = [ ] D = [ ] E = [ ] F = [ ] G = [ ] H = [ ] I = [ ] J = [ ] K = [ ] L = [ ] M = [ ] N = [ ] O = [ ] P = [ ] Q = [ ] |
The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
Babalola, O.G.; Alam Bhuiyan, M.M.; Hammad, A. Literature Review on Collaborative Project Delivery for Sustainable Construction: Bibliometric Analysis. Sustainability 2024 , 16 , 7707. https://doi.org/10.3390/su16177707
Babalola OG, Alam Bhuiyan MM, Hammad A. Literature Review on Collaborative Project Delivery for Sustainable Construction: Bibliometric Analysis. Sustainability . 2024; 16(17):7707. https://doi.org/10.3390/su16177707
Babalola, Olabode Gafar, Mohammad Masfiqul Alam Bhuiyan, and Ahmed Hammad. 2024. "Literature Review on Collaborative Project Delivery for Sustainable Construction: Bibliometric Analysis" Sustainability 16, no. 17: 7707. https://doi.org/10.3390/su16177707
Article access statistics, further information, mdpi initiatives, follow mdpi.
Subscribe to receive issue release notifications and newsletters from MDPI journals
You have full access to this open access article
1 Altmetric
Succession planning is a vital organizational process that facilitates the smooth transition of leadership and safeguarding organizational knowledge. This research paper seeks to determine the most efficient and effective method for identifying and selecting potential leaders within an organization by examining succession planning policies and their effectiveness in identifying and selecting potential leaders in diverse organizational contexts. This paper utilizes policy analysis methodology to analyze succession planning policies' components, objectives, and outcomes across various industries and sectors. The analysis is based on an extensive review of literature, policy documents, and case studies. This study addresses the methods and tools used to identify and select potential leaders using a scorecard structure. In conclusion, the nine-box grid is a valuable tool that assesses employees' performance and potential for leadership positions.
Succession planning for large and small organizations: a practical review of professional business corporations, succession planning models, conceptual maps: ethical considerations and best practices, explore related subjects.
Avoid common mistakes on your manuscript.
The healthcare environment is known as dynamic and unstable, and healthcare is intensely affected by the change. So, healthcare leadership is becoming more complex. The shortage of competent workers affects service recipients and long-term growth [ 1 , 2 ]. Several countries lack trained healthcare professionals, particularly nurses. The World Health Organization (2016) reports a global shortage of 14.5 million doctors, nurses, and midwives. This shortage could harm vital programs in the health sector by 2030.
A needs-based study by the World Health Organization (WHO) (2016) anticipates a 7.6 million nurse shortage by 2030 [ 1 ]. That requires healthcare leadership to understand and build strategies to improve staff retention and ensure a smooth transition when staff retires [ 3 ]. Establish workforce planning based on current and emerging health service and community needs. Those needs require a crucial and effective strategy or policy to maintain the best organizational performance, as recommendation seven calls for nurses to prepare and enable nurses to lead change to advance health [ 4 ].
Nurse leaders prepare and enable nurses to lead change to advance health. This recommendation insists that "nurses, nursing education programs, and nursing associations should prepare the nursing workforce to assume leadership positions across all levels, while public, private, and governmental health care decision-makers should ensure that leadership positions are available to and filled by nurses" [ 4 ]. So today, nursing leadership should adopt new strategies to improve the retention of the nursing workforce and prepare and enable nurses to lead change to advance health [ 5 ]. Succession planning can be utilized in every area of healthcare succession planning [ 6 ].
Developing a succession plan policy usually starts with developing a strategic plan, identifying and assessing the key positions, and then identifying the potential leaders [ 7 ]. These employees are assessed as having leadership ability, organizational commitment, and motivation to grow and succeed in more senior positions. A systematic method for candidate selection is one of the most crucial components in developing a solid succession plan [ 8 ]. Even if an employee is working well in their current position, that does not guarantee that they will be successful in a more senior role [ 9 ].
There are distinct between leadership potential and performance; the assessment of leadership potential includes an assessment of an individual's potential for assuming future leadership positions, typically focusing on characteristics such as adaptability and strategic thinking. In contrast, a performance assessment is a process that assesses an individual's present job-related accomplishments, with a particular focus on past and current achievements. Both variables are crucial when assessing an individual's preparedness for leadership roles, but with different objectives [ 10 ]. It's critical to have a strategy in place for accurately identifying high-performing employees. Therefore, we must consider this when selecting potential succession candidates. So, determine measurable criteria to evaluate each candidate's potential [ 11 ].
The identifying and selecting criteria on which a potential leader will be nominated were not addressed in the policy for succession planning. Identifying a high-potential leader is an ideal strategic objective for succession planning [ 12 ]. Additionally, it is an essential step in succession planning [ 7 , 12 ]. So, in our policy analysis Here, the question is, "What is the most effective and efficient method of identifying and selecting those potential leaders in the organization?".
Organizations usually struggle to identify potential leaders for further development. As healthcare is a dynamic work environment, it is crucial to provide a solid leadership pool for the future [ 13 ]. Identifying and selecting the most successful potential leaders are crucial strategic objectives for maintaining an organization's viability and competitiveness [ 14 ].
The breakdown of leadership is crucial for any organization's survival and long-term sustainability [ 15 ]. The breakdown of leaders and managers can be a terrible occurrence with far-reaching effects within and beyond the organization. In other words, succession planning, strategy, leadership, and culture are interconnected and comprise many aspects of an organization's vision [ 7 ]. A significant aspect of an organization's long-term viability is how succession planning is integrated with organizational strategy and culture, future leader training and development, and change management to ensure business continuity [ 16 , 17 ]. Succession planning is used to identify, select, and develop future leaders; it must be planned, implemented, and measured to ensure a positive conclusion [ 18 ].
The literature shows that most organizations do not have formal or written succession plans [ 19 ]. According to the literature, succession planning, and an organization's future leader's selection, development, and growth should be a focused, clearly articulated approach to benefit the firm, including all stakeholders [ 20 ]. This kind of planning might be important for developing leaders in different stakeholders.
Creating an effective succession plan for healthcare leadership allows for the development and support of future leaders through training programs. On the other hand, the employee must demonstrate leadership interest and potential [ 16 ]. Incorporating methods established in other healthcare departments, the nursing industry may be able to serve as a model for other healthcare leaders' succession planning strategies [ 21 ]. Suppose organizations do not effectively select, develop, and retain future leaders. In that case, they will encounter an unpredictable future characterized by the breakdown of leadership, which is crucial for any organization's survival and long-term sustainability. By leadership vacancies, the loss of key individuals, and possibly unstable operations [ 22 ]. Even though there is a huge and varied body of knowledge on succession planning, there is still much to learn. The literature also reveals that nursing has a well-established succession process and that healthcare leaders, in general, can benefit from modeling how nursing prepares for leadership transitions. Nursing leaders are crucial in healthcare organizations as they link leadership and nurses [ 23 ]. A lack of succession planning can be expensive and raise hospital expenses when healthcare institutions seek senior positions [ 24 ].
The research literature argues that the process of identifying future leaders includes the utilization of a variety of criteria and methodologies. In accordance with the Knoll study from 2021, it is possible to assess global leadership potential (GLP) by considering a variety of traits, attitudes, and competencies. Identifying GLP entails three stages: nomination, assessment, and confirmation [ 25 ]. According to Norman (2020), the assessment of leadership potential can be achieved by utilizing a comprehensive approach that incorporates behavioral simulations, psychometrics, and processing tests [ 26 ].
In a study conducted by Panait in 2017, a self-assessment questionnaire was Created to identify and evaluate leadership abilities and characteristics [ 27 ]. According to Knaub (2018), the exclusive reliance on social network analysis may not be adequate for identifying leaders[ 28 ]. Therefore, overall, the reviews indicate that identifying and selecting future leaders requires a comprehensive strategy considering different criteria and methods. Still, the articles didn't identify which methods are recommended. The author shows that using different methods can result in inconsistencies and unsuccessful utilization. Therefore, it is important to identify and select the best approach to ensure its effectiveness in implementation.
This paper aims to identify the most effective method for identifying and selecting potential organizational leaders. The main objective is to assess different methods and measure their effectiveness and efficiency in this context. We aim to offer practical recommendations to improve the organization's leadership development and succession planning.
The creativity and motivation of a future leader will contribute to the success of the organization; recognizing these employees' abilities enables department heads to focus on strategic development and evaluation, which could happen based on pre-defined qualifying criteria and appraise the right candidate to determine their strengths and development requirements.
Identifying and selecting willing employees based on measurable abilities using a procedure that ensures every person with leadership potential is examined fairly and fully for potential leadership roles.
In our policy analysis plan, we selected the rational model approach due to its well-organized framework, supported by research, emphasizing evidence-based analysis and transparent decision-making [ 29 ]. We assumed that the various alternatives would be of interest to achieving the objectives and responding to our policy analysis question. This analysis revised the most encountered alternatives for identifying and selecting potential organizational leaders.
The evaluation criteria for each alternative must account for its capacity to achieve policy objectives and outcomes, reduce costs, be politically and administratively feasible, and be suitable for long-term goals. The criteria will determine how effective and fair the different options are to determine how well the goals and objectives are met.
Some organizations approach internal recruitment more closely, asking managers to nominate high-performing employees for internal positions. When individuals are familiar with employees' work in different departments in smaller organizations, this informal approach can be quite effective. However, this technique may appear to reflect discriminatory practices [ 30 ]. Organizations often have a well-defined process for performance evaluation; however, many managers are confused about evaluating potential, resulting in incorrect potential assessments [ 10 ]. Unconscious biases could also affect how employees are judged if there aren't clear ways to pick team members with high potential.
A 360-degree assessment tool allows you to get input from others on an applicant's potential leadership skills. When used correctly, it will most likely assist you in identifying and selecting your potential leader. A 360-degree assessment tool is a game-changing approach to providing employees with consistent feedback, support, and possibilities for advancement. It is a significant improvement over traditional approaches to learning and development, as well as annual performance reviews [ 31 ].
An interview is a sort of interaction sometimes used to identify the best applicant. This is essentially a conversation between two or more people seeking to know and understand more about one another. The interviewee will usually talk about their opinions, views, and background, while the interviewer will ask about their knowledge, skills, and abilities [ 32 ]. One of the most significant benefits of conducting interviews is gaining knowledge about your organization's potential leaders' skills and abilities. However, there are some disadvantages to this method [ 33 ]. The method also takes a very long time and requires a significant amount of stress response. It's important to remember that interviewing is a two-way street. Suppose you go out of your way to ensure the interviewee has a positive experience that benefits them. This might also assist you in selecting the best people for the position [ 34 ].
The 9-box grid is a typical method of identifying and classifying potential intended to assist organizations in understanding the kinds of potential they have and determining where to focus their future development and budget [ 35 ]. Anyone can utilize the 9-box Grid, but it is mostly used by HR professionals, managers, and other development professionals. The 9-box grid is frequently used in succession planning [ 36 ]. A complete succession plan, on the other hand, extends beyond identifying and selecting talent in an organization.
The results present a comparison between different alternatives. Six criteria make up the evaluation tool. Here, we used a Likert scale of 1 to 5 to rate each criterion. (A score of 1 indicates a very low likelihood; a score of 2 indicates a low likelihood; a score of 3 indicates a moderate likelihood; a score of 4 indicates a high likelihood, and a score of 5 indicates a very high likelihood.) The evaluation carried out by the group in this analysis paper uses a few different ways to determine the best alternative in selecting potential managers, including supervisor nomination, a 360 assessment tool, an interview, and a 9-Box Grid. Each group member was required to rate every alternative separately and submit their results to their team leader. The results show that the 9-box-grid tool receives the highest score when using the scorecard method, as illustrated in Table 1 , which happened after the group's team leader calculated the submitted comparisons by utilizing evaluation criteria shown in Table 2 and came up with the mean scores.
The 9-box grid is a useful performance management tool that managers use for several reasons. It is simple [ 37 ], and the 9-box grid has a straightforward structure, as shown in Table 3 . Also, it can be used in various organizations because it needs little background research or data collection, and it is possible to perform this task based on first-hand observation [ 38 ]. The 9-box grid produces a visual representation of a firm's talent pool. It's a tool that can be used to compare potential leaders and facilitate debate and decision-making [ 39 ]. The benefit of implementing a 9-box grid is that it saves both time and money [ 39 ].
Most other alternatives may initially seem attractive, but they have a high long-term cost [ 40 ]. Here are a few limitations when implementing the 9-box Grid. The distinction between performance and potential can be challenging, especially if neither idea has a clear definition or comprehension [ 41 ]. If you choose to be honest and transparent with your staff and reveal performance measurements, you risk negatively affecting them and decreasing their satisfaction. Having a "poor performance" or "low potential" can have a detrimental impact on lower-level employees, and with good reason [ 42 ].
There are three phases to constructing a 9-box grid for your organization's succession planning [ 37 ]:
The first phase is evaluating each employee's level of performance. The actual criteria for evaluating performance vary depending on an organization's needs. However, each employee must be assigned to one of three groups: Low performance: The individual does not meet the work criteria and does not meet the company's targets and objectives. They lack motivation and alignment with the company's vision [ 37 ]. Moderate performance: the employee's job requirements, personal targets, and ambitions are only partially met [ 37 ]. High performance: The employee meets all their job description and personal goals and performs consistently throughout all tasks [ 37 ].
The next step is to evaluate each employee's potential. Employees' potential is based on how much they are expected to grow in the future, how willing they are to learn new skills, and how well they can use what they know in everyday situations. One of the problems is that organizations aren't as good at judging employees' potential as they are at judging their performance. Performance refers to previous conduct, while potential refers to expected future behavior to simplify the distinction between the two concepts.
The next stage is plotting all employees on a 3 × 3 grid, resulting in your 9-box grid, after assessing them as low, moderate, or high on both performance and potential. Managers can also see where each employee ranks according to the matrix.
Those in leadership positions are responsible for challenging the status quo, innovating, and rallying their people to propel enterprises to new heights of success [ 43 ]; in contrast, in leadership positions, selection failures can cost an organization a large amount of money and disrupt its potential growth for several years. Therefore, organizations must carefully consider their leadership choices [ 44 ]. It is necessary to have a leadership selection procedure that identifies the abilities, qualities, and competencies of potential candidates and decides whether they are appropriate for these positions [ 45 ].
One of the greatest practices for filling leadership positions is to create an internal pool of potential leaders who can be accessed at any time. Our goals and objectives could include analyzing productivity, turnover, days to fill, cost per hire, and quality of hire, as well as measuring productivity, turnover, days to fill, recruitment costs, and quality of talent management. Various indicators in which management is polled regarding the recruitment and selection process, their staff, or patient experience. Collaboration and organizational success within their departments and throughout the organization.
The 9 Box Grid is a tool that provides a comprehensive view of an organization's intellectual equity. The research reveals that utilizing the 9-box grid is a practicable tool for succession planning and talent management. This tool helps organizations assess and categorize their workforce according to their performance and potential, enabling the identification of potential leaders and the development of strategies for developing an effective leadership succession plan [ 37 , 42 , 46 , 47 , 48 ]. However, despite having higher job performance assessments than males, one study indicated that women receive significantly lower "potential" ratings after reviewing 9-box grid data for close to 30,000 employees. The implementation of the 9-box tool has the potential to jeopardize an organization's efforts toward fostering diversity, equity, and inclusion by unintentionally impeding the progression of underrepresented individuals into leadership positions. As a result of these individuals' potential demotivation and departure from the organization in search of better possibilities, diversity levels may be reduced overall. Therefore, the subjectivity of the 9-box Grid can affect individuals from historically marginalized backgrounds [ 48 ].
This article can inspire further studies in succession planning since it relates to healthcare and nursing contexts. Nurses and healthcare organizations can utilize the findings of this study to optimize leadership development programs, therefore adapting strategies to address the needs of the healthcare professions. Nurses have the potential to enhance their comprehension of the avenues leading to leadership positions within healthcare organizations, influencing their professional paths and goals. Implementing a well-designed leadership succession plan can significantly impact the entire quality of patient care, offering advantages for both nurses and their patients. Leadership may acknowledge the significance of fostering a talent pipeline inside their organizations, ensuring smooth succession when leadership roles become vacant.
Healthcare education programs have the potential to incorporate succession planning aspects into their curriculum to adequately equip aspiring healthcare employees for positions of leadership in the future. Policymakers may consider the implications of the findings on healthcare policy, which could potentially impact policies related to succession planning within healthcare organizations. Also, policymakers may see a potential connection between implementing effective leadership succession planning and enhancing healthcare quality, which encourages them to allocate resources towards efforts to develop competence in leadership. A research study on succession planning in healthcare organizations can have significant consequences in various areas, including research, nursing practice, leadership development, education, and policymaking.
Incorporating that knowledge can effectively inform appropriate approaches, facilitate improvements in the development of leadership qualities, and ultimately result in enhanced patient care and improved healthcare outcomes.
The present policy analysis study has the potential to provide policymakers and stakeholders with carefully analyzed information and evidence, enabling them to make sensible and informed decisions. In addition to providing realistic policy recommendations or alternatives based on research findings, these conclusions can serve as valuable guidance for policymakers in formulating effective policies. As the availability of precise and Comprehensive data can be a limitation, especially when studying the implications of policy, there are many limitations in this policy analysis paper. Another limitation is our comprehensive evaluation to identify and select potential organizational leaders. We made an effort to consider a variety of alternatives, but we may have missed certain alternatives. Furthermore, it is worth mentioning that our study did not proceed to the implementation phase. Therefore, the practical implications and effectiveness of the selected approach have yet to be tested.
Identifying potential by comparing performance to potential. The outcomes of this activity can successfully reveal the most appropriate succession, thereby identifying employees with high potential and performance. This term is used to describe employees who are most prepared to lead.
During the talent identification and selection phase, the 9-Box Grid is the primary instrument for assessing the potential and establishing organizational workforce strategic plans. Each department employee is carefully examined and assigned to the appropriate quadrant (box) in this task. By completing the 9 Box Grid tasks, accurate plans can be made for the development and retention of each staff member. To evaluate the successful operation for selecting potential leaders to achieve the aims of implementing it, we must determine which indications best represent the method defined and established within the organization.
Not applicable.
World Health Organization. Global strategy on human resources for health: workforce 2030.
Lockhart L. Strategies to reduce nursing turnover. Nurs Made Incred Easy. 2020;18(2):56.
Article Google Scholar
DeCenzo DA, Robbins SP, Verhulst SL. Fundamentals of human resource management. John Wiley & Sons; 2016.
National Academies of Sciences, Engineering, and Medicine. Assessing progress on the Institute of Medicine report. The Future of Nursing; 2016.
Wei H, et al. Nurse leaders’ strategies to foster nurse resilience. J Nurs Manag. 2019;27(4):681–7.
Article PubMed Google Scholar
Smith A. Eur J Public Health. 2019; 29 (Supplement_4):p. ckz185.
Al Suwaidi M, et al. Determinants linked to executive succession planning in public sector organizations. Vision. 2020;24(3):284–99.
Griffin RW, Phillips JM, Gully SM. Organizational behavior: Managing people and organizations. CENGAGE learning; 2016.
Stone RJ, Cox A, Gavin M, Carpini J. Human resource management. John Wiley & Sons; 2024.
Church AH, et al. Is there potential in assessing for high-potential? Evaluating the relationships between performance ratings, leadership assessment data, designated high-potential status and promotion outcomes in a global organization. Leadersh Q. 2021;32(5): 101516.
Hile K, Agnor M, Roesgen L, Lamb C. A Comprehensive Tool for Principal Succession Planning (Doctoral dissertation, Saint Louis University).
Smith AD. Corporate succession planning: accident or part of the strategic benchmarking process? Int J Process Manag Benchmark. 2022;12(6):695–729.
Behie SW, Pasman HJ, Khan FI, Shell K, Alarfaj A, El-Kady AH, Hernandez M. Leadership 4.0: The changing landscape of industry management in the smart digital era. Process Saf Environ Protect. 2023;172:317-28.
Obianuju AA, Ibrahim UA, Zubairu UM. Succession planning as a critical management imperative: A systematic review. Modern Manag Rev. 2021;26(4):69–92.
Pavlov OV, Katsamakas E. Will colleges survive the storm of declining enrollments? A computational model. PLoS ONE. 2020;15(8): e0236872.
Article CAS PubMed PubMed Central Google Scholar
Masenya TM. Integrating talent and knowledge management practices in the new normal business environment: developing future leaders in public sector organizations. In Navigating the new normal of business with enhanced human resource management strategies 2022 (p. 113–144). IGI global.
Acheampong J, Baidoo P, Omari C. Archives of the social sciences: a journal of collaborative memory. Arch Soc Sci A J Collab Mem. 2023;1(1):67–77.
Google Scholar
Okwakpam JA. Effective succession planning: A roadmap to employee retention. Kuwait Chapter Arab J Bus Manag Rev. 2019;8(2):1–10.
Yawson RM. Human resource development and executive leadership succession planning in nonprofits. In Proceedings of the 56th Annual Eastern Academy of Management Conference. 2019 (p. 1–16).
Chang CY, Besel K. Cultivating next generation of healthcare leaders in Havana: Barriers and recommendation for succession planning. Int J Healthc Manag. 2021;14(4):1062–70.
Rimita KN. Leader readiness in a volatile, uncertain, complex, and ambiguous (VUCA) business environment (Doctoral dissertation, Walden University).
Tucker CA. Succession planning for academic nursing. J Prof Nurs. 2020;36(5):334–42.
Beitz JM. The perioperative succession crisis: A cross-sectional study of clinical realities and strategies for academic nursing. Nurs Econ. 2019;37(4):179–97.
Knoll C, Sternad D. Identifying global leadership potential. J Manag Dev. 2021;40(4):253–72.
Norman J. Leadership potential: Measurement beyond psychometrics . The British Psychological Society. 2020;12(2).
Panait C. Developing leadership skills. Identifying leadership qualities and attributes. Rev Air Force Acad. 2017(1):167.
Knaub AV, Henderson C, Fisher KQ. Finding the leaders: an examination of social network analysis and leadership identification in STEM education change. Int J STEM Educ. 2018;5(1):26.
Article PubMed PubMed Central Google Scholar
Gray D, De Haan E, Bonneywell S. Coaching the ‘ideal worker’: female leaders and the gendered self in a global corporation. Eur J Train Dev. 2019;43(7/8):661–81.
Jann W, Wegrich K. Theories of the policy cycle. In: Handbook of public policy analysis. Routledge; 2017. p. 69–88.
González-Gil M, et al. 360-Degree evaluation: Towards a comprehensive, integrated assessment of performance on clinical placement in nursing degrees: A descriptive observational study. Nurse Educ Today. 2020;95: 104594.
Solarino AM, Aguinis H. Challenges and best-practice recommendations for designing and conducting interviews with elite informants. J Manag Stud. 2021;58(3):649–72.
Team IE. Advantages and disadvantages of interviews (and useful tips). 2023; Available from: https://in.indeed.com/career-advice/interviewing/advantages-and-disadvantages-of-interviews .
Maurer R. The pros and cons of virtual and in-person interviews. 2016; Available from: https://www.shrm.org/resourcesandtools/hr-topics/talent-acquisition/pages/pros-and-cons-virtual-in-person-interviews.aspx .
Park HM. The context matters to the exclusive talent management: how to measure and pay in South Korea. Glob Bus Rev. 2023; 09721509231157014.
Boštjančič E, Slana Z. The role of talent management comparing medium-sized and large companies–major challenges in attracting and retaining talented employees. Front Psychol. 2018;9:1750.
Traynor S, Wellens MA, Krishnamoorthy V. Succession management intro, 9-box, and talent review forms. In: SAP success factors talent: volume 2: a complete guide to configuration, administration, and best practices: succession and development. Springer; 2021. p. 289–328.
Chapter Google Scholar
Garbellano S, Ughetto E, Esfahani HD. Talent acquisition and development in global companies (based on a study on nexans Group). 2020.
Gaonkar S, Khan D, Manisha AS. Impact of gamification on learning and development. J Adv Educ Philos. 2022;6:63–70.
Sparrow P, Hird M, Cooper CL, Sparrow P, Hird M, Cooper CL. Strategic talent management. Palgrave Macmillan UK; 2015.
Mistry R. Dilemmas in the realm of leadership talent development. NHRD Netw J. 2017;10(2):45–52.
Munira S. Importance of skills inventory and training in succession planning-how does it help in employee performance: The case of Heidelberg Cement Bangladesh Limited (HCBL), Chittagong Plant. 2022.
Rivera MJ. Leveraging innovation and intrapreneurship as a source for organizational growth. Int J Innov Sci. 2017;9(2):137–52.
O’Reilly III CA, Tushman ML. Lead and disrupt: How to solve the innovator's dilemma. Stanford University Press. 2021.
Lynn DB. Succession management strategies in public sector organizations. In: Administrative Leadership in the Public Sector. Routledge; 2016. p. 375–89.
Newhall S. Aligning the talent development and succession planning processes: Don’t allow critical leadership talent to fall by the wayside. Dev Learn Organ Int J. 2015;29(5):3–6.
Jones P, Rivers R. Development of a staff nurse to chief nursing officer succession planning program. Nurse Lead. 2021;19(6):646–54.
Martin A. Talent management: Preparing a “Ready” agile workforce. Int J Pediatr Adolesc Med. 2015;2(3–4):112–6.
Benson A, Li D, Shue K. Potential and the gender promotion gap. Unpublished Working Paper. 2021.
Download references
The Qatar National Library funded the publication of this article.
Authors and affiliations.
Nursing Department, Hamad Medical Corporation, Doha, Qatar
Mutaz I. Othman, Abdulqadir J. Nashwan, Hothaifah Hijazi & Ahmed Abu jaber
Faculty of Nursing, Zarqa University, P.O. Box 3050, Doha, Qatar
Islam Oweidat
You can also search for this author in PubMed Google Scholar
MIO: Conceptualization. MIO, IO, AJN, HH, AA: Methods, Literature search, Manuscript preparation. All authors have accepted responsibility for the entire content of this manuscript and approved its submission.
Correspondence to Abdulqadir J. Nashwan .
Ethics approval and consent to participate, consent for publication, competing interest.
The authors declare no competing interests.
Publisher's note.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .
Reprints and permissions
Othman, M.I., Oweidat, I., Nashwan, A.J. et al. Identifying and selecting the next generation of nursing leaders through effective succession planning: a policy analysis. Discov Health Systems 3 , 73 (2024). https://doi.org/10.1007/s44250-024-00130-5
Download citation
Received : 25 September 2023
Accepted : 05 August 2024
Published : 04 September 2024
DOI : https://doi.org/10.1007/s44250-024-00130-5
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative
IMAGES
VIDEO
COMMENTS
Once you have read and re-read your articles and organized your findings, you are ready to begin the process of writing the literature review. 2. Synthesize. (see handout below) Include a synthesis of the articles you have chosen for your literature review. A literature review is NOT a list or a summary of what has been written on a particular ...
Furthermore, novice nursing researchers may receive little formal training to develop the skills required to generate a comprehensive integrative review (Boote and Beile, 2005). Aveyard and Bradbury-Jones (2019) also emphasised the limited literature providing guidance surrounding integrative reviews. Therefore, novice nursing researchers need ...
in undertaking a traditional or narrative review of the Table 2. The literature review process • Selecting a review topic • Searching the literature • Gathering, reading and analysing the literature • Writing the review • References literature {Table 2). The first step involves identifying the subject ofthe literature review.
A literature review is a comprehensive and up-to-date overview of published information on a subject area. Conducting a literature review demands a careful examination of a body of literature that has been published that helps answer your research question (See PICO). Literature reviewed includes scholarly journals, scholarly books ...
Run a few sample database searches to make sure your research question is not too broad or too narrow. If possible, discuss your topic with your professor. 2. Determine the scope of your review. The scope of your review will be determined by your professor during your program. Check your assignment requirements for parameters for the Literature ...
This broader and more varied literature often leads to a better understanding of the topic. Whittemore and Knafl 7 developed a framework for conducting an integrative review, commonly used in nursing. This framework has five stages: (1) problem identification, (2) literature search, (3) data evaluation, (4) data analysis, and (5) presentation ...
A literature review is an essay that surveys, summarizes, links together, and assesses research in a given field. It surveys the literature by reviewing a large body of work on a subject; it summarizes by noting the main conclusions and findings of the research; it links together works in the literature by showing how the information fits into the overall academic discussion and how the ...
Doing a Literature Review in Nursing, Health and Social Care by Michael Coughlan; Patricia Cronin. Call Number: RT 81.5 .C68 2021. ISBN: 9781526497512. ... The Conducting a Literature Review Guide gives you links to key resources to help you get started finding and organizing your resources.
fi. taken is in uenced by the purpose of the review and. fl. resources available. However, the stages or methods used to undertake a review are similar across approaches and include: Formulating clear inclusion and exclusion criteria, for example, patient groups, ages, conditions/treat-ments, sources of evidence/research designs;
Abstract. Systematic reviews provide a synthesis of evidence for a specific topic of interest, summarising the results of multiple studies to aid in clinical decisions and resource allocation. They remain among the best forms of evidence, and reduce the bias inherent in other methods. A solid understanding of the systematic review process can ...
There are several steps in developing a literature review. These include: Step 1 Define Your Goal. Define your paper's goal. Literature review will match paper's goal. Step 2 Do Your Research. Review articles related to your paper's topic. Articles are written by scholars. Identify top scholars in the field about your topic.
These steps for conducting a systematic literature review are listed below. Also see subpages for more information about: What are Literature Reviews? ... Asking the Clinical Question, AJN The American Journal of Nursing: March 2010 - Volume 110 - Issue 3 - p 58-61 doi: 10.1097/01.NAJ.0000368959.11129.79 ...
Scoping Review: A preliminary assessment of the size and scope of available published literature. A scoping review is intended to identify current research and the extent of such research, and determine if a more comprehensive review is viable. Can include research in progress, and the completeness of searching is determined by time/scope.
In this quick 11 minute video, Dr Zina O'Leary explains the misconceptions and struggles students often have with writing a literature review. She also provides step-by-step guidance on writing a persuasive literature review. This open textbook is designed for students in graduate-level nursing and education programs.
Essential First Step in Research, Quality Improvement, and ...
A literature review can be a short introductory section of a research article or a report or policy paper that focuses on recent research. Or, in the case of dissertations, theses, and review articles, it can be an extensive review of all relevant research. The format is usually a bibliographic essay; sources are briefly cited within the body ...
Assesses the potential scope of the research literature on a particular topic. Helps determine gaps in the research. 2-8 weeks: 1-2: Traditional (narrative) literature review: A generic review which identifies and reviews published literature on a topic, which may be broad. Typically employs a narrative approach to reporting the review findings.
Ten Simple Rules for Writing a Literature Review - PMC
A literature review is a summary and analysis of research published on a specific topic. Literature reviews give a "snapshot" of individual articles and explain how each work has contributed to the field's understanding of the topic. The purpose of a literature review is to trace the history of research on a particular subject, evaluate that ...
Systematically Reviewing the Literature: Building ...
This article provides a practical overview of the process of undertaking systematic reviews, explaining the rationale for each stage. It provides guidance on the standard methods applicable to every systematic review: writing and registering a protocol; planning a review; searching and selecting studies; data collection; assessing the risk of ...
Doing a Literature Review in Nursing, Health and Social Care by Michael Coughlan; Patricia Cronin. Call Number: RT 81.5 .C68 2021. ISBN: 9781526497512. ... The Conducting a Literature Review Guide gives you links to key resources to help you get started finding and organizing your resources.
Writing and research can be challenging for nurses at undergraduate and postgraduate level; however, understanding the process and developing the skills to conduct a literature review with a staged strategy will positively affect care delivery. Nurses have a responsibility to deliver care based on the best evidence available. Therefore, developing the necessary skills to conduct a literature ...
A literature review is exploring research that has been done directly on the topic you have chosen. Conducting a literature review will give you the big picture of what is already known about your topic and allow you to see where there may be gaps in the knowledge. <<
Aim: This systematic review identifies the factors and effective strategies related to nursing students' readiness for practice. Method: A search was conducted from 2012 to 2022 in PubMed, CINAHL ...
This scoping review was conducted under the Joanna Briggs Institute (JBI) framework. It included primary studies published until 30 April 2023, obtained through a systematic search across PubMed, Web of Science, CINAHL, and MEDLINE databases. The review focused on primary studies that used the Delphi technique in nursing competence research, especially those related to defining core competency ...
Background Successful deployment of clinical prediction models for clinical deterioration relates not only to predictive performance but to integration into the decision making process. Models may demonstrate good discrimination and calibration, but fail to match the needs of practising acute care clinicians who receive, interpret, and act upon model outputs or alerts. We sought to understand ...
In healthcare settings worldwide, workplace violence (WPV) has been extensively studied. However, significantly less is known about gender-based WPV and the characteristics of perpetrators. We conducted a comprehensive scoping review on Type II (directed by consumers) and Type III (perpetuated by healthcare workers) gender based-WPV among nurses and physicians globally. For the review, we ...
This paper aims to conduct a bibliometric analysis and traditional literature review concerning collaborative project delivery (CPD) methods, with an emphasis on design-build (DB), construction management at risk (CMAR), and integrated project delivery (PD) Methods. This article seeks to identify the most influential publications, reveal the advantages and disadvantages of CPD, and determine ...
Organizations usually struggle to identify potential leaders for further development. As healthcare is a dynamic work environment, it is crucial to provide a solid leadership pool for the future [].Identifying and selecting the most successful potential leaders are crucial strategic objectives for maintaining an organization's viability and competitiveness [].