Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 Feb 7;81(10):6560–6568. doi: 10.1111/jan.16803

Understaffing as a Two‐Dimensional Phenomenon: A Cross‐Sectional Study of Hospital Nurses' Manpower and Expertise Understaffing

Shani Pindek 1,✉, Melisa R Hayman 2, David J Howard 3, Maryana L Arvan 4, Paul E Spector 5
PMCID: PMC12460951  PMID: 39918111

ABSTRACT

Aim(s)

To examine whether manpower and expertise understaffing are distinct, and whether they relate similarly to nursing stressors, burnout, job satisfaction and intentions to turnover.

Design

A cross‐sectional survey of hospital nurses nested within units was used.

Methods

The sample included 402 nurses. Nurses provided ratings of the study's variables using validated self‐report measures. The data were analysed both as multilevel and single‐level data.

Results

Manpower and expertise understaffing contributed unique explained variance to all of the examined outcomes. Nurses within the same units experience different understaffing levels. Expertise understaffing emerged as a significantly stronger predictor than manpower understaffing for three of the six of the outcome variables (illegitimate tasks, job satisfaction and turnover intentions).

Conclusion

Manpower and expertise understaffing are distinct, and both are associated with nurse outcomes.

Reporting Method

We have adhered to the STROBE guideline for cross‐sectional studies.

Implications for the Profession and/or Patient Care

Considering both manpower and expertise understaffing to maintain proper staffing levels in nursing units is crucial.

Patient or Public Contribution

A Director of Patient Care Services from the hospital where the study was conducted is a member of the research team. This member contributed to designing and conducting the study as well as interpreting the results.

Keywords: burnout, hospital nurses, illegitimate tasks, job satisfaction, turnover, understaffing, workload

Summary.

  • Impact
    • ○
      The study highlighted the importance of expertise understaffing in nursing.
    • ○
      Expertise understaffing is distinct from manpower understaffing, and a stronger predictor of some outcomes.
    • ○
      This research will have impact on staffing of nursing units, which traditionally relied on nurse‐to‐patient ratio, a crude measure of manpower understaffing that does not account of expertise understaffing.
  • What does this paper contribute to the wider global clinical community?
    • ○
      Nurse‐to‐patient ratios are not nuanced enough to capture understaffing.
    • ○
      Considering both manpower and expertise understaffing to maintain proper staffing levels in nursing units is crucial.
    • ○
      Expertise understaffing can be more important than manpower understaffing.

1. Introduction

Understaffing occurs when human resources are insufficient for a unit to effectively accomplish all its tasks. This has become a pressing problem in many healthcare organisations, as there are worldwide shortages of nurses and other healthcare providers (Metcalf et al. 2018). The consequences of nurse understaffing are of vital concern due to its connection to poor patient care (Twigg et al. 2015), adverse patient outcomes (Aiken et al. 2014) and nurse burnout (Aiken et al. 2002). Although understaffing in nursing units is recognised as an important global problem, there is an incomplete understanding of the nature of this phenomenon because most prior research has focused merely on the number of nurses or nurse‐to‐patient ratios on a unit. What is missing is that understaffing is a more complex phenomenon and how it is perceived by nurses is important, especially when it comes to the link with stress, burnout and retention.

2. Background

2.1. Understaffing Outcomes

Research on the effect of understaffing on the levels of other work stressors has been focused primarily on how understaffing relates to work overload (Hudson and Shen 2018). Nurses who have excessive workloads due to inadequate staffing on their units are likely to experience decrements to well‐being resulting in burnout and impaired performance (Pérez‐Francisco et al. 2020). Understaffing can lead to increases in workload for nurses who have to perform the work of more than one person. However, there may be additional stressors that arise when nursing units are understaffed. In addition to having to do more work, nurses may be asked to perform tasks that go beyond their roles. Specifically, they can be assigned tasks that ideally someone else should be doing or should not be done at all, sometimes referred to as illegitimate tasks (Semmer et al. 2015). Examples of such ‘illegitimate tasks’ within nursing are nursing tasks that should be done by a different nurse (e.g., a nurse with more training), non‐nursing tasks that should be done by someone with a different job (e.g., a nursing aide or a sanitation worker) or tasks that need to be carried out without proper resources such as time and training (Kilponen et al. 2021). Illegitimate tasks are a prevalent problem in nurses (Spector et al. 2024) and lead to negative outcomes such as decreased commitment (Gahrmann and Klumb 2024), burnout (Moncayo‐Rizzo et al. 2024) and increased intentions to leave (Sasso et al. 2019).

Employees are likely to perceive requests to perform tasks they see as illegitimate as unfair, as such requests violate norms about what tasks are considered acceptable for their position (Ding and Kuvaas 2022). Asking nurses to perform such tasks can be seen as a form of disrespect that threatens their professional identities and undermine their sense of self‐esteem (Semmer et al. 2010). Feelings of being disrespected serve as social stressors that can undermine positive well‐being for nurses (Gilin Oore et al. 2010).

Understaffing, particularly when it is chronic, is theorised to lead to burnout and job dissatisfaction (Hudson and Shen 2015). Evidence both outside of healthcare (Hudson and Shen 2018) and within healthcare and nursing in particular (Aiken et al. 2002) supports the effects of understaffing on burnout dimensions such as fatigue and emotional exhaustion. Nurse‐to‐patient ratio, a measure of understaffing, is a strong and significant predictor of burnout and job dissatisfaction (Aiken et al. 2002). Furthermore, job dissatisfaction is a leading cause of nurses intentions to leave their jobs, and the nursing profession in general (Sasso et al. 2019). Therefore, understaffing is likely a direct and indirect predictor of burnout, job dissatisfaction and, importantly, intention to turnover (Wendsche et al. 2017). Intentions to turnover are particularly important within the context of understaffing because understaffing leads to more nurses leaving, and this retention problem in turn increases understaffing.

2.2. Understaffing as a Multidimensional Construct

Taking a closer look at the understaffing construct itself, Hudson and Shen (2018) distinguish manpower (too few people) from expertise (unit lacks needed skills) understaffing. However, there is little known about the distinction between these two dimensions of understaffing in nursing, and if they relate differently to nurse outcomes. Most studies on understaffing in nursing units used some variation of the nurse‐to‐patient ratio as a measure of understaffing (Aiken et al. 2014). Examples include whether staff hours worked on a shift were less than 8–10 h below the associated unit's mean (Twigg et al. 2015), the patient load on a shift, regardless of which shift (day, evening, night) or specialty (Aiken et al. 2002), or sometimes a nurse‐to‐patient ratio adjusted for the level of care that is needed depending on patient criticality (Kiekkas et al. 2019). These measures are often too crude because the optimal nurse‐to‐patient ratio may vary across units and shifts or patient acuity and complexity of care (this is accounted for in some cases but not always). Another limitation of using nurse‐to‐patient ratio as a measure of understaffing is that it cannot distinguish manpower from expertise understaffing. This distinction has practical implications for hospitals that are trying to manage their staffing levels. For example, understaffed units sometimes employ temporary nurses to relieve their manpower understaffing. However, these temporary nurses are not as effective at resolving expertise understaffing, as they require onboarding and supervision by the permanent nurses. These extra demands may be perceived as illegitimate by the permanent nurses who are already facing high workload levels (Gahrmann and Klumb 2024).

Finally, nurse‐to‐patient ratio is calculated at the unit or hospital level, but there may be individual differences in how the level of understaffing is experienced among nurses within each unit. It is possible, for example, that nursing managers assign more tasks and tasks requiring greater expertise to experienced nurses than inexperienced nurses. On the other hand, new nurses in need of support from experienced colleagues might find those colleagues overloaded and unavailable on understaffed units. Therefore, in this study, we use a more nuanced approach by asking nurses themselves to estimate understaffing levels from their personal experience and asking them to assess both manpower and expertise understaffing. We also used a study design that will allow us to separate unit‐level understaffing estimates from individual‐level understaffing perceptions.

3. The Study

The main aim of the current study was to acquire a more nuanced understanding of the understaffing phenomenon within nursing units that can inform how nursing units can most effectively be managed. We used a quantitative cross‐sectional survey of nurses nested within nursing units in one large American non‐profit academic medical centre and level one trauma centre to assess all of the study's variables. First, we included separate measures of manpower and expertise understaffing. Second, we consider that within each unit, different nurses may experience the unit's understaffing differently. Last, we examine an array of potential outcomes including additional stressors (workload and illegitimate tasks), nurse burnout (cognitive fatigue and emotional exhaustion) and job satisfaction and intentions to turnover. We propose that both forms of understaffing will lead to increases in nurse stressors, burnout and intentions to turnover and a decrease in job satisfaction. We will examine whether the two understaffing dimensions have different associations with these outcomes, and we will demonstrate their incremental validity when predicting these outcomes.

4. Method

4.1. Sample

A survey invitation was distributed through five internal email lists of nurses in a non‐profit academic medical centre in the southeastern United States in December 2023, with three reminders spaced at 8, 14 and 16 days from survey launch. We sent out 1432 email invitations and received 402 responses, but some nurses were included in multiple lists; therefore, the calculated response rate of 28% is likely an underestimate. The survey was presented as a research project, and participation was voluntary and anonymous. Six hundred and one nurses clicked the survey link during the survey period (December 3–20), with 402 indicating their unit (first question in the survey). Our final sample for the main analyses included all 402 nurses who participated. However, data for the multilevel analyses were included only if at least three nurses within the same unit completed the survey. Our final sample for the multilevel analyses therefore includes 356 hospital nurses, working in 46 different hospital units. We did not collect data on participants' gender or age because this information would have made some of the participants identifiable. The average tenure as a nurse was 11 years (SD = 10.6), ranging from less than a year to 46 years, with 19.7% having 2 years or less of nursing experience. The highest education level of the participating nurses varied with 15% having an associate degree, 75% a bachelors' degree, 9% a master's degree and 1% a doctoral degree.

4.2. Measures

All variables were assessed using established measures. Reliability was assessed in the current study with coefficient alphas, presented on the main diagonal of Table 1.

TABLE 1.

Descriptive statistics and correlations among the study variables.

Mean SD 1 2 3 4 5 6 7
1 Manpower understaffing 4.77 1.68 0.91
2 Expertise understaffing 3.86 1.66 0.45 0.85
3 Workload 4.14 0.84 0.42 0.38 0.87
4 Illegitimate tasks 4.45 1.8 0.39 0.56 0.41 0.80
5 Cognitive fatigue 4.18 1.51 0.25 0.31 0.41 0.43 0.93
6 Emotional exhaustion 3.36 1.29 0.27 0.33 0.40 0.39 0.75 0.77
7 Job satisfaction 5.49 1.33 −0.24 −0.33 −0.30 −0.38 −0.53 −0.43 0.89
8 Turnover intentions 3.01 1.47 0.31 0.44 0.33 0.48 0.63 0.52 −0.69
Open in a new tab

Note: All correlations are significantly different from zero at p < 0.001. N = 392–400. Reliabilities are in bold on the diagonal.

Manpower and expertise understaffing were measured each with a 3‐item scale (Hudson and Shen 2018), adapted to nursing units by replacing the word ‘employees’ with ‘nurses’. Items were rated on a 7‐point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree). Sample items for manpower and expertise understaffing are ‘There are not enough nurses in our work unit to complete all required job tasks’ and ‘Our work unit is missing nurses with key knowledge and skills’, respectively.

Workload was assessed with the 5‐item Quantitative Workload Inventory (Spector and Jex 1998). Items were rated on a 1 (less than once per month or never) to 5 (several times per day) frequency scale. An example item is ‘How often does your job leave you with little time to get things done?’

Illegitimate tasks were measured with a 2‐item scale (Matthews et al. 2022). The items were as follows: ‘I had to carry out tasks at work that were unnecessary, or could be done easier if things were better organized’ and ‘I had to do unreasonable things at work that fell outside of my job responsibilities and should be done by someone else’. These items were rated on a 7‐point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree).

The burnout dimensions of cognitive fatigue and emotional exhaustion were assessed with five items for cognitive fatigue and three items for emotional exhaustion (Shirom et al. 2006). Items were rated on a 7‐point frequency scale ranging from 1 (almost never) to 7 (almost always). An example item is ‘feeling difficulty in concentrating’ for cognitive fatigue, and ‘feeling emotionally fatigued’ for emotional exhaustion.

Job satisfaction were measured with the 3‐item Cammann et al. (1983) scale from the Michigan Organizational Assessment Questionnaire. An example item is ‘All in all, I am satisfied with my job’. Items were rated on a 7‐point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree).

Intentions to turnover was measured with a single item, ‘How often have you seriously considered leaving your current job?’ (Spector et al. 1988). This item was rated on a 6‐point scale ranging from 1 (Never) to 6 (Extremely often).

4.3. Design and Statistical Approach

Hospital nurses were nested in their units for the more restricted sample (N = 356). Therefore, as a first step, interclass correlations (ICC(1)s) were estimated for the study's variables. ICC(1)s are estimates of the proportion of total variance among responses by the nurses that can be partitioned into two components: the between‐unit level and the within‐unit level. Between‐unit variance indicates how different the unit means are on a variable. Within‐unit variance indicates how much nurses vary from one another within their own unit. If all the units have the same mean score, but nurses within the units have different scores, all the variance is at the within‐unit level (an ICC(1) of 0), and if each unit has a different mean score, but all nurses within each unit have the same score, then all the variance is at the unit level (an ICC(1) of 1). For psychological variables such as the variables in the current study, there is usually some variance at each of the levels (an ICC(1) between 0 and 1). With the exception of manpower understaffing, which had an ICC(1) of 0.24, indicating that 24% of the variance resides at the between‐unit level, all other ICC levels were below 0.10, ranging from 0.09 for expertise understaffing, to 0.02 for fatigue. These ICC(1) results indicate that almost all of the variance resides at the within‐unit level, and there is hardly any variance among units (< 10%). We therefore proceeded with an analysis of the bivariate correlations at the individual nurse level only, assuming the variance among units is of no practical significance in this case. To verify this assumption, we also compared these correlations to the within‐group level correlations obtained from a multilevel analysis conducted using the Mplus software version 8.8. These correlations control for between‐unit differences. The mean difference in correlation magnitude between the two sets of correlations (controlling and not controlling for unit mean differences) was 0.01. The within‐group correlations alongside the bivariate calculated on the comparable sample (N = 356) can be found in the Appendix A.

We used a cross‐sectional design (all data were collected in a single survey), which is a commonly used design in survey studies of nurses. Data from a cross‐sectional survey can be in line with or contradict a theoretical process model, but it cannot directly test the causal relationships within the model. Our goal was to determine if the pattern of correlations was consistent with understaffing resulting in increased stressors (workload and illegitimate tasks), increased burnout, decreased job satisfaction and increased intentions to turnover. We also use statistical comparisons between dependent correlations (Lenhard and Lenhard 2014). Finally, we used regression analyses to examine the incremental validity of the two understaffing dimensions. A separate regression analysis was used for each of the six outcome variables: workload, illegitimate tasks, the two burnout dimensions, job satisfaction and intentions to turnover. Each regression included both understaffing dimensions in order to examine whether or not both dimensions have a unique contribution to the prediction of the outcomes.

4.4. Ethical Considerations

This project received approval (exempt status) from the University of South Florida Institutional Review Board, Study number 004360, and from the Tampa General Hospital Office of Clinical Research. Participants were presented with an informed consent sheet that explained the purpose of the study and that participation is voluntary and anonymous.

5. Results

Table 1 contains the means, standard deviations and internal consistency reliability levels of the study's variables as well as the zero‐order correlations among them. Both manpower and expertise understaffing were correlated with workload (r = 0.41 and r = 0.39, respectively, p < 0.001 for both). Furthermore, manpower and expertise understaffing were also correlated with illegitimate tasks (r = 0.39 and r = 0.56, respectively, p < 0.001 for both), and these correlations significantly differed from one another (z = 3.85, p < 0.01; note all correlation comparisons were one‐tailed) such that the association between expertise understaffing and illegitimate tasks is stronger than the association between manpower understaffing and illegitimate tasks.

Manpower and expertise understaffing were correlated with the two burnout dimensions (correlations ranged from r = 0.26 to r = 0.31), and workload and illegitimate tasks were also correlated with the two burnout dimensions (correlations ranged from r = 0.39 to r = 0.44). The z‐tests comparing correlation pairs were nonsignificant.

Finally, manpower and expertise understaffing were correlated with job satisfaction (r = −0.24 and r = −0.33, respectively, p < 0.001 for both), and these correlations significantly differed from one another (z = 1.81, p < 0.05) such that the association between expertise understaffing and job satisfaction is stronger than the association between manpower understaffing and job satisfaction. Manpower and expertise understaffing were also correlated with intentions to turnover (r = 0.31 and r = 0.44, respectively, p < 0.001 for both), and these correlations also significantly differed from one another (z = 2.73, p < 0.01) such that the association between expertise understaffing and intentions to turnover is stronger than the association between manpower understaffing and intentions to turnover.

Next, we conducted regression analyses for each of the six outcome variables: workload, illegitimate tasks, the two burnout dimensions, job satisfaction and intentions to turnover. The results, presented in Table 2, show that both understaffing dimensions have a unique contribution for predicting all six outcome variables. In all cases, both understaffing variables have statistically significant regression coefficients. We ran the same set of regressions a second time, including education level and tenure on the job as control variables. The standardised estimates remained similar (changes ranged from 0 to 0.02 with a mean of 0.01). Both understaffing variables remain statistically significant predictors with one exception: In predicting job satisfaction, the estimates for manpower understaffing loses significance (β = 0.10, p = 0.056).

TABLE 2.

Regression results showing standardised regression coefficients and overall R 2.

Workload Illegitimate tasks Cognitive fatigue Emotional exhaustion Job satisfaction Intent to turnover
Manpower understaffing 0.32** 0.15** 0.13** 0.15** −0.12* 0.14**
Expertise understaffing 0.23** 0.48** 0.24** 0.25** −0.27** 0.35**
R 2 0.22** 0.32** 0.10** 0.12** 0.12** 0.18**
Open in a new tab
*

p < 0.05.

**

p < 0.01.

6. Discussion

This study provides a nuanced understanding of understaffing in nursing units. Whereas most understaffing studies use unit‐level nurse‐to‐patient ratios, our study assessed the experiences and perceptions of the nurses themselves. A cross‐sectional survey was used to provide evidence for the distinction between manpower and expertise understaffing as seen by nurses. We examined potential outcomes for both manpower and expertise understaffing and found significant associations between both of these dimensions and stressors (workload and illegitimate tasks), burnout (cognitive fatigue and emotional exhaustion), job satisfaction and intention to turnover. The two dimensions were distinct, each contributing unique explained variance in all of the examined outcomes. For three of the six outcomes, expertise understaffing was a significantly stronger predictor than manpower understaffing. Simply put, when there are not enough nurses, particularly nurses with the right mix of skills, the nurses on the unit experience additional stressors, increased burnout levels, reduced job satisfaction and increased turnover intentions.

6.1. Understaffing at the Individual Level

Most of the variance when estimating the levels of understaffing was found between individuals, and not between nursing units. This was particularly striking when estimating expertise understaffing, where more than 90% of the variance was at the individual level. This pattern emerged in a hospital that, as the data shows, has fairly equal understaffing levels across its nursing units, at least as viewed by the nurses themselves. This finding shows that an objective indicator of understaffing, such as the nurse‐to‐patient ratio, or an indicator based on unfilled nurse position within each unit, can only tell part of the story, because some nurses (regardless of their unit) experienced higher levels of understaffing than other nurses on their unit. In this sense, efforts by the hospital to combat understaffing by changing the units' nurse‐to‐patient ratio are unlikely to resolve the problem completely because the experience of understaffing can be different for nurses within the same unit, and because nurse‐to‐patient ratio only deals with manpower understaffing (even that in a crude way) and not with expertise understaffing which is no less important, as shown in this study.

There are a number of reasons that can explain why nurses in the same unit might experience understaffing differently. Years of nursing experience, expertise, level of personal and professional resilience, adaptability and skill are all likely to affect the extent to which understaffing has an impact. Some nurses may be better equipped to adapt to varying patient ratios and complexity because of their years or diversity of clinical experience or personality characteristics (e.g., Louch et al. 2016). Another phenomenon that may impact how an individual nurse experiences understaffing is the ‘experience‐complexity gap’ defined as the decline in the overall experience of the nursing workforce (Virkstis et al. 2019) characterised by new graduate nurses replacing experienced retiring baby boomer nurses and early career nurses exiting bedside care in pursuit of advanced degrees. The experience‐complexity gap was exacerbated by the COVID pandemic as new nursing graduates entered the workforce with little to no clinical experience due to COVID restrictions preventing hospital‐based clinical rotations (Diaz et al. 2021). The experience‐complexity gap may exist on an individual level indicated by the disparity between an individual nurse's level of experience compared the complexity of care required by the patient(s) they are assigned. This may be evident in an organisation with a lower mean years of nursing experience or when a nurse floats (or is assigned patients) to a clinical unit with a nursing specialty different from their home unit. A third reason nurses may experience understaffing differently is related to the resources available to them during their shift. For example, nurses who have resources like unlicensed personnel (nurse techs or aides), injury prevention teams to assist with repositioning or ambulating patients, and resource nurses like ‘helping hands’ and admit or discharge nurses are able to load shed some nursing and illegitimate tasks to other employees leaving them more time to provide direct care to their assigned patients. Finally, a nurse manager facing a shortage is likely to look to the most experienced and skilled nurses to take up the slack, particularly with tasks requiring a high level of expertise. These reasons are relevant at the individual level (Virkstis et al. 2022) and may contribute to different experiences of understaffing within the same unit.

6.2. Understaffing Outcomes

We examined two stressors as potential understaffing outcomes. The first is an increase in workload, which is considered an immediate consequence of understaffing (Hudson and Shen 2015). If a unit is understaffed, then the same amount of work that needs to be done within the unit has to be done by fewer individuals, increasing the load on each one. In the current study, there were differences in perceptions of understaffing within the units, and nurses who perceived greater understaffing levels (manpower or expertise) also experienced higher workloads.

The second stressor is illegitimate tasks. Workload and illegitimate tasks often go hand in hand, and these variables were correlated in the current study. Nevertheless, illegitimate tasks go beyond an increase in workload. They are perceived as unfair and are more likely to be assigned when the unit suffers from expertise understaffing. When a nursing unit is understaffed, nurses not only have to do more work, but also they often have to do more work that they do not consider to be a part of their jobs, such as mentoring less experience nurses (Gahrmann and Klumb 2024). This type of task can be perceived as being someone else's job, or as being the result of poor management that led to the understaffing in the first place. If nurses are being asked to do patient care tasks for which they do not have the training and/or time resources needed (Kilponen et al. 2021), this would likely be viewed as an illegitimate task that can be easily attributed to poor management. Both workload and illegitimate tasks are also known predictors of burnout in healthcare (e.g., Moncayo‐Rizzo et al. 2024), and future studies can examine the potential process by which understaffing can result in burnout and other outcomes indirectly, via an increase in these stressors.

Beyond the increased stressor levels, both forms of understaffing were also associated with burnout, job satisfaction and turnover intentions. This is in line with prior studies (Aiken et al. 2002; Sasso et al. 2019; Wendsche et al. 2017). Interestingly, expertise understaffing was a stronger predictor of job satisfaction and turnover intentions compared with manpower understaffing. It is possible that expertise understaffing, which results in nurses being assigned more illegitimate tasks, is consequently more harmful to nurses' satisfaction with the job, because of the perceived unfairness of their day‐to‐day tasks. Dissatisfied nurses are more likely to intend to leave their jobs (Sasso et al. 2019). This highlights the importance of distinguishing between manpower and expertise understaffing. While both are associated with an array of negative outcomes, some outcomes may be more associated with expertise understaffing.

6.3. Implications for Nurse Leaders

When planning each unit's target staffing levels, it is important to remember that expertise understaffing and not just manpower understaffing are important. Relying only on nurse‐to‐patient ratio to assess the level of understaffing is not likely to resolve the problem entirely because this ratio only deals with manpower understaffing and not with expertise understaffing. Current trends necessitate replacing experienced nurses who quit with newly graduated and inexperienced nurses, which contributes to the experience‐complexity gap. The hiring of inexperienced nurses addresses the manpower issue but not necessarily the expertise issue. Moreover, the different perceptions within each unit are an indicator of the difficulty in estimating the actual understaffing levels. This was especially true with expertise understaffing and is a challenge that nurse leaders have to face as part of their role.

Furthermore, when a unit is understaffed, nurse leaders must pay special attention to the increase in workload, and the type of tasks that are being assigned to the nurses, because increased workload and illegitimate tasks are additional stressors that can result from understaffing, and these additional stressors can also lead to negative consequences for the nurses. Ultimately, when a unit is understaffed, there may be no choice but to increase the workload, but special effort should be invested in reducing tasks that are unnecessary or unfair. Importantly, what tasks are considered illegitimate can vary among nurses, both across and within countries. In the current study, nurses rated the overall level of illegitimate tasks they were assigned, but nurse leaders assign specific tasks. They should aim to minimise assigning tasks deemed illegitimate, keeping in mind that different nurses may have differing judgements on the legitimacy of each task.

Given the aging nursing workforce and the increasing proportion of nurses planning to retire within the next 5 years (Smiley et al. 2021), understaffing and its associated challenges are expected to intensify. At the same time, newly graduated nurses are taking longer to develop the clinical competence required in an increasingly complex care environment (Herleth et al. 2020). To address these challenges, the healthcare sector must adopt strategic and innovative approaches. This includes prioritising professional development for new graduates, hiring from the pool of temporary travel nurses so they transition to permanent roles (Virkstis et al. 2022), implementing innovative upskilling programs (Rascón et al. 2024) and creating integrative succession planning initiatives that focus on the recruitment and development of future nurse leaders at all management levels (Griffith 2012; Tucker 2020). Nurse leaders must explore innovative nursing models of care like team‐based nursing or ‘buddy’ nursing where a more experienced nurse is paired with a less experienced nurse to share a patient assignment. This supports the ‘right nurse—right patient’ approach by allowing the more experienced nurse to manage complex care like critical intravenous medications while the less experienced nurse manages system assessments. This model allows for a safe learning environment for the newer nurse and an opportunity for the experienced nurse to mentor them in skill development and critical thinking. Other resources, such as regularly scheduled breaks away from the unit that allow nurses to rest and recuperate are also helpful when units are understaffed (Wendsche et al. 2017). These breaks are effective only if there is someone responsible for the assignments of the nurse that is taking a break. Other creative staffing models can include ‘mid‐shift’ nurses who come in for a few hours to relieve team members for breaks and meal periods, a role that can be attractive to nurses who wish to work part‐time.

Nurse managers should also be aware of the experiences and capabilities of individual nurses. Not everyone will respond the same to a given situation, and while understaffing is a root issue that needs addressing, a unit that is understaffed might be experienced as overwhelming to one nurse and as a growth opportunity to another. Nurses differ in their skills and abilities. Some are better able to handle stressful workloads than others, while some are more efficient at doing their jobs than others. Nursing managers should make an effort to learn the capabilities and reactions of their nurses so they can identify those individuals who might be more prone to experience, and who would need more support in handling increased workloads or illegitimate tasks. They should also be willing to listen to concerns by nurses who feel that being given a particular task is inappropriate and be transparent in explaining why particular work assignments were made, as this can reduce the perception that the assigned tasks are unfair. Addressing the understaffing strategically (see for example Virkstis et al. 2022) may help reduce its negative consequences, including reduced job satisfaction, and increased burnout and intentions to turnover.

6.4. Limitations

The biggest limitation of this study is that it demonstrated how two dimensions of understaffing related to outcomes, but it could not indicate whether understaffing was the cause or the effect. Although in theory understaffing would be expected to lead to stressors, burnout, dissatisfaction, and ultimately turnover, we cannot be certain from the results of this study that this is the case. Nurses who are burned out or dissatisfied, for example, might attribute those feelings to unreasonably heavy workloads or being asked to perform illegitimate tasks even though those feelings are due to something else.

As with all self‐report studies, there is the possibility that observed relationships were due to biases inherent in the survey method, that is, the monomethod bias issue (Spector 2006). To address the limitations of cross‐sectional survey research designs, studies are needed to evaluate the impact of staffing interventions to determine if they are effective in reducing perceived understaffing as well improving subsequent outcomes.

Furthermore, like most prior understaffing studies, this study does not account for the potential variation in perceived understaffing from shift to shift within a single unit. This variation is based on a combination of how sick and complex the patients are, their care needs across shifts and which specific nurses, with their varying levels of skills and experience, are scheduled to work during that shift. Future studies could examine this variation by using study designs with repeated measures, such as experience sampling designs whereby nurses report on their work environment and experiences every day or shift (e.g., Louch et al. 2016).

This study relied on a sample of nursing units from a single hospital, limiting the generalisability of the results. Moreover, in this study, there was hardly any variance between units, indicating that in the hospital where the study was conducted, staffing levels, as assessed in the current study, are perceived as fairly similar across units. This feature of the hospital may or may not be typical of hospitals in general. Nevertheless, it prevented us from conducting a full multilevel analysis and gaining a better understanding of understaffing at the unit level which has obvious important implications for understanding the phenomenon. Perhaps, a better approach would be to collect data on a sample of different hospitals that varied in objective staffing levels where we would expect to find more between‐unit (in this case hospital) variance.

7. Conclusion

Maintaining proper staffing levels in nursing units, considering both manpower and expertise understaffing, is a key antecedent that can lead to lower levels of nurse stress, burnout, dissatisfaction and ultimately turnover. The results of the current study showed that expertise understaffing and not just manpower understaffing are important predictors. Furthermore, it is important for nurse leaders to consider the variability in the experienced understaffing levels within their units and whether it is due to uneven, and perhaps unfair, assignments. For example, are some nurses within the unit receiving more illegitimate tasks than others? Nurses who are experiencing higher understaffing levels are likely to experience additional stressors and suffer from increased burnout, reduced satisfaction and increased intentions to turnover. Therefore, a more nuanced understanding of understaffing is imperative for improved management of nursing units.

Disclosure

The authors have checked to make sure that our submission conforms as applicable to the Journal's statistical guidelines. There is a statistician on the author team, Paul Spector. The authors affirm that the methods used in the data analyses are suitably applied to their data within their study design and context, and the statistical findings have been implemented and interpreted correctly. The author(s) agrees to take responsibility for ensuring that the choice of statistical approach is appropriate and is conducted and interpreted correctly as a condition to submit to the Journal.

Ethics Statement

This project received approval (exempt status) from the University of South Florida Institutional Review Board, Study number 004360, and from the Tampa General Hospital Office of Clinical Research.

Conflicts of Interest

The authors declare no conflicts of interest.

Peer Review

The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer‐review/10.1111/jan.16803.

Appendix A.

Within‐Unit Correlations Among Study variables

1 2 3 4 5 6 7
1 Manpower understaffing
2 Expertise understaffing 0.37
3 Workload 0.36 0.35
4 Illegitimate tasks 0.32 0.54 0.41
5 Cognitive fatigue 0.26 0.30 0.39 0.44
6 Emotional exhaustion 0.23 0.28 0.40 0.39 0.74
7 Job satisfaction −0.25 −0.29 −0.29 −0.37 −0.52 −0.44
8 Turnover intentions 0.32 0.41 0.33 0.48 0.63 0.52 −0.69
Open in a new tab

Note:  N = 356. All correlations are significantly different from zero at p < 0.001.

Bivariate Correlations Among Study Variables

1 2 3 4 5 6 7
1 Manpower understaffing
2 Expertise understaffing 0.43
3 Workload 0.41 0.39
4 Illegitimate tasks 0.36 0.56 0.41
5 Cognitive fatigue 0.26 0.31 0.40 0.44
6 Emotional exhaustion 0.28 0.30 0.41 0.39 0.74
7 Job satisfaction −0.23 −0.32 −0.30 −0.37 −0.52 −0.42
8 Turnover intentions 0.31 0.42 0.33 0.49 0.63 0.50 −0.69
Open in a new tab

Note: N = 356. All correlations are significantly different from zero at p < 0.001.

Funding: The authors received no specific funding for this work.

Data Availability Statement

Data available on request from the authors.

References

  1. Aiken, L. H. , Clarke S. P., Sloane D. M., Sochalski J. A., and Silber J. H.. 2002. “Hospital Nurse Staffing and Patient Mortality, Nurse Burnout, and Job Dissatisfaction.” JAMA 288, no. 16: 1987–1993. 10.1001/jama.288.16.1987. [DOI] [PubMed] [Google Scholar]
  2. Aiken, L. H. , Sloane D. M., Bruyneel L., et al. 2014. “Nurse Staffing and Education and Hospital Mortality in Nine European Countries: A Retrospective Observational Study.” Lancet 383, no. 9931: 1824–1830. 10.1016/S0140-6736(13)62631-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Cammann, C. , Fichman M., Jenkins D., and Klesh J. R.. 1983. “Assessing the Attitudes and Perceptions of Organizational Members.” In Assessing Organizational Change: A Guide to Methods, Measures, and Practices, edited by Seashore S., Lawler E., Mirvis P., and Cammann C., 71–138. John Wiley. [Google Scholar]
  4. Diaz, K. , Staffileno B. A., and Hamilton R.. 2021. “Nursing Student Experiences in Turmoil: A Year of the Pandemic and Social Strife During Final Clinical Rotations.” Journal of Professional Nursing 37, no. 5: 978–984. 10.1016/j.profnurs.2021.07.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Ding, H. , and Kuvaas B.. 2022. “Illegitimate Tasks: A Systematic Literature Review and Agenda for Future Research.” Work & Stress 37, no. 3: 1–24. 10.1080/02678373.2022.2148308. [DOI] [Google Scholar]
  6. Gahrmann, C. , and Klumb P. L.. 2024. “Investigating the Impact of Temporary Nurses on Permanent Nurses' Commitment via Perceptions of Illegitimate Tasks: A Cross‐Sectional Study.” Journal of Advanced Nursing 80, no. 7: 2835–2846. 10.1111/jan.16021. [DOI] [PubMed] [Google Scholar]
  7. Gilin Oore, D. , Leblanc D., Day A., et al. 2010. “When Respect Deteriorates: Incivility as a Moderator of the Stressor‐Strain Relationship Among Hospital Workers.” Journal of Nursing Management 18, no. 8: 878–888. 10.1111/j.1365-2834.2010.01139.x. [DOI] [PubMed] [Google Scholar]
  8. Griffith, M. B. 2012. “Effective Succession Planning in Nursing: A Review of the Literature.” Journal of Nursing Management 20, no. 7: 900–911. 10.1111/j.1365-2834.2012.01418.x. [DOI] [PubMed] [Google Scholar]
  9. Herleth, A. , Virkstis K., Renfroe J., and Rewers L.. 2020. “The Challenging Road to Clinical Competence for New Graduate RNs.” Journal of Nursing Administration 50, no. 4: 185–186. 10.1097/NNA.0000000000000864. [DOI] [PubMed] [Google Scholar]
  10. Hudson, C. K. , and Shen W.. 2015. “Understaffing: An Under‐Researched Phenomenon.” Organizational Psychology Review 5, no. 3: 244–263. 10.1177/2041386615576016. [DOI] [Google Scholar]
  11. Hudson, C. K. , and Shen W.. 2018. “Consequences of Work Group Manpower and Expertise Understaffing: A Multilevel Approach.” Journal of Occupational Health Psychology 23, no. 1: 85–98. 10.1037/ocp0000052. [DOI] [PubMed] [Google Scholar]
  12. Kiekkas, P. , Tsekoura V., Aretha D., et al. 2019. “Nurse Understaffing Is Associated With Adverse Events in Postanaesthesia Care Unit Patients.” Journal of Clinical Nursing 28, no. 11–12: 2245–2252. 10.1111/jocn.14819. [DOI] [PubMed] [Google Scholar]
  13. Kilponen, K. , Huhtala M., Kinnunen U., Mauno S., and Feldt T.. 2021. “Illegitimate Tasks in Health Care: Illegitimate Task Types and Associations With Occupational Well‐Being.” Journal of Clinical Nursing 30, no. 13–14: 2093–2106. 10.1111/jocn.15767. [DOI] [PubMed] [Google Scholar]
  14. Lenhard, W. , and Lenhard A.. 2014. “Hypothesis Tests for Comparing Correlations.” https://www.psychometrica.de/correlation.html.
  15. Louch, G. , O'Hara J., Gardner P., and O'Connor D. B.. 2016. “The Daily Relationships Between Staffing, Safety Perceptions and Personality in Hospital Nursing: A Longitudinal On‐Line Diary Study.” International Journal of Nursing Studies 59: 27–37. 10.1016/j.ijnurstu.2016.02.010. [DOI] [PubMed] [Google Scholar]
  16. Matthews, R. A. , Pineault L., and Hong Y.‐H.. 2022. “Normalizing the Use of Single‐Item Measures: Validation of the Single‐Item Compendium for Organizational Psychology.” Journal of Business and Psychology 37, no. 4: 639–673. 10.1007/s10869-022-09813-3. [DOI] [Google Scholar]
  17. Metcalf, A. Y. , Wang Y., and Habermann M.. 2018. “Hospital Unit Understaffing and Missed Treatments: Primary Evidence.” Management Decision 56, no. 10: 2273–2286. 10.1108/MD-09-2017-0908. [DOI] [Google Scholar]
  18. Moncayo‐Rizzo, J. , Alvarado‐Villa G., and Cossio‐Uribe C.. 2024. “The Impact of Illegitimate Tasks on Burnout Syndrome in a Healthcare System: A Cross‐Sectional Study.” International Journal of Nursing Studies Advances 6: 100185. 10.1016/j.ijnsa.2024.100185. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Pérez‐Francisco, D. H. , Duarte‐Clíments G., del Rosario‐Melián J. M., Gómez‐Salgado J., Romero‐Martín M., and Sánchez‐Gómez M. B.. 2020. “Influence of Workload on Primary Care Nurses' Health and Burnout, Patients' Safety, and Quality of Care: Integrative Review.” Health 8, no. 1: 12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Rascón, A. M. , Nelson A., and Richardson D.. 2024. “Nurse Upskilling During Crisis: Collaborating for Continuing Education.” Journal of Continuing Education in Nursing 55, no. 5: 257–260. 10.3928/00220124-20240201-02. [DOI] [PubMed] [Google Scholar]
  21. Sasso, L. , Bagnasco A., Catania G., et al. 2019. “Push and Pull Factors of Nurses' Intention to Leave.” Journal of Nursing Management 27, no. 5: 946–954. 10.1111/jonm.12745. [DOI] [PubMed] [Google Scholar]
  22. Semmer, N. K. , Jacobshagen N., Meier L. L., et al. 2015. “Illegitimate Tasks as a Source of Work Stress.” Work & Stress 29, no. 1: 32–56. 10.1080/02678373.2014.1003996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Semmer, N. K. , Tschan F., Meier L. L., Facchin S., and Jacobshagen N.. 2010. “Illegitimate Tasks and Counterproductive Work Behavior.” Applied Psychology 59, no. 1: 70–96. 10.1111/j.1464-0597.2009.00416.x. [DOI] [Google Scholar]
  24. Shirom, A. , Nirel N., and Vinokur A. D.. 2006. “Overload, Autonomy, and Burnout as Predictors of Physicians' Quality of Care.” Journal of Occupational Health Psychology 11, no. 4: 328–342 0.1037/1076‐8998.11.4.328. [DOI] [PubMed] [Google Scholar]
  25. Smiley, R. A. , Ruttinger C., Oliveira C. M., et al. 2021. “The 2020 National Nursing Workforce Survey.” Journal of Nursing Regulation 12, no. 1, Supplement: S1–S96. 10.1016/S2155-8256(21)00027-2. [DOI] [Google Scholar]
  26. Spector, P. E. 2006. “Method Variance in Organizational Research Truth or Urban Legend?” Organizational Research Methods 9, no. 2: 221–232. 10.1177/1094428105284955. [DOI] [Google Scholar]
  27. Spector, P. E. , Dwyer D. J., and Jex S. M.. 1988. “Relation of Job Stressors to Affective, Health, and Performance Outcomes: A Comparison of Multiple Data Sources.” Journal of Applied Psychology 73, no. 1: 11–19. 10.1037/0021-9010.73.1.11. [DOI] [PubMed] [Google Scholar]
  28. Spector, P. E. , and Jex S. M.. 1998. “Development of Four Self‐Report Measures of Job Stressors and Strain: Interpersonal Conflict at Work Scale, Organizational Constraints Scale, Quantitative Workload Inventory, and Physical Symptoms Inventory.” Journal of Occupational Health Psychology 3, no. 4: 356–367. 10.1037/1076-8998.3.4.356. [DOI] [PubMed] [Google Scholar]
  29. Spector, P. E. , Pindek S., Hayman M. R., Howard D. J., and Arvan M. L.. 2024. “Travel Nurse Work Experiences: A Comparison of Staff and Travel Nurses' Burnout and Job Attitudes.” Nursing Management 55, no. 5: 12–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Tucker, C. A. 2020. “Succession Planning for Academic Nursing.” Journal of Professional Nursing 36, no. 5: 334–342. 10.1016/j.profnurs.2020.02.002. [DOI] [PubMed] [Google Scholar]
  31. Twigg, D. E. , Gelder L., and Myers H.. 2015. “The Impact of Understaffed Shifts on Nurse‐Sensitive Outcomes.” Journal of Advanced Nursing 71, no. 7: 1564–1572. 10.1111/jan.12616. [DOI] [PubMed] [Google Scholar]
  32. Virkstis, K. , Boston‐Fleischhauer C., Rewers L., et al. 2022. “7 Executive Strategies to Stabilize the Nursing Workforce.” Journal of Nursing Administration 52, no. 4: 194–196. 10.1097/NNA.0000000000001129. [DOI] [PubMed] [Google Scholar]
  33. Virkstis, K. , Herleth A., and Rewers L.. 2019. “Closing Nursing's Experience‐Complexity Gap.” Journal of Nursing Administration 49, no. 12: 580–582. 10.1097/NNA.0000000000000818. [DOI] [PubMed] [Google Scholar]
  34. Wendsche, J. , Hacker W., and Wegge J.. 2017. “Understaffing and Registered Nurses' Turnover: The Moderating Role of Regular Rest Breaks.” German Journal of Human Resource Management 31, no. 3: 238–259. 10.1177/2397002216683880. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Data available on request from the authors.

Articles from Journal of Advanced Nursing are provided here courtesy of Wiley

RESOURCES