Literature Review Sample: Nurse Staffing Levels and Patient Safety Outcomes

Model Answer

Introduction

Registered nurse staffing levels have been debated for over two decades as one of the most consistently modifiable determinants of patient safety in acute hospital care. Following high-profile UK care failures, most notably those examined by the public inquiry into Mid Staffordshire NHS Foundation Trust, safe staffing moved from a largely operational and budgetary concern to a recognised patient safety and regulatory priority (Francis, 2013; National Institute for Health and Care Excellence, 2014). This review critically synthesises empirical evidence on the relationship between registered nurse staffing levels and patient safety outcomes in acute inpatient settings, drawing on international cohort studies, UK-based analyses and systematic reviews published over roughly the past fifteen years.

The scope of the review is deliberately focused on acute adult inpatient wards, since the evidence base for staffing effects is strongest and most consistent in this setting, and because most current UK policy debate centres on general and acute medical or surgical care. This is particularly true given the divergence between Wales, which has legislated minimum registered nurse staffing considerations through the Nurse Staffing Levels (Wales) Act, and England, which continues to rely on locally determined, acuity-based establishments rather than a mandated ratio. Four themes structure the discussion that follows: first, the association between nurse-to-patient ratios and mortality or major adverse events; second, missed or “rationed” nursing care as the mechanism through which understaffing is thought to translate into patient harm; third, the role of skill mix and the proportion of registered relative to unregistered staff; and fourth, the reciprocal relationship between staffing levels, nurse wellbeing and the broader organisational safety climate. The review closes by identifying gaps in the evidence base that currently limit confident UK policymaking.

This is a narrative, thematic review rather than a systematic review; it prioritises depth of critical synthesis within each theme over exhaustive search coverage. The studies discussed span several countries, several operational definitions of “understaffing”, and a range of outcome measures, and this methodological heterogeneity is itself returned to in the discussion of gaps below. Wherever possible, study design, sample size and effect size are reported alongside each citation, so that the reader can weigh the strength of the underlying evidence rather than treating every study as equally authoritative.

Nurse-to-Patient Ratios and Mortality Outcomes

The most extensively researched theme in this literature is the direct association between nurse-to-patient staffing ratios and patient mortality. Aiken et al. (2014), analysing linked survey and discharge data from over 400 hospitals across nine European countries as part of the RN4CAST consortium, found that each additional patient added to an individual nurse’s workload was associated with a 7% increase in the odds of inpatient death within 30 days of admission, after adjusting for patient case-mix and hospital characteristics. This cross-national consistency across health systems with very different funding models lent the finding considerable weight, although the underlying design remains cross-sectional and survey-linked, which limits causal inference.

Needleman et al. (2011), working with shift-level staffing data from a large United States hospital system rather than unit-average ratios, found that exposure to shifts staffed below the target level for registered nurse hours was associated with an increased hazard of inpatient mortality, and that this risk accumulated with each additional understaffed shift a patient was exposed to during their stay. This shift-level design offered a methodological advance over earlier ward-average studies, since it captured within-patient variation in staffing exposure rather than assuming a constant staffing level throughout an admission.

The strongest causal evidence in this theme comes from McHugh et al. (2021), who exploited the phased introduction of mandated minimum nurse-to-patient ratios across Queensland, Australia as a natural experiment. Comparing hospitals before and after ratio implementation, and against hospitals that did not yet have ratios, they found that mandated ratios were associated with statistically significant reductions in mortality, readmissions and length of stay. Because this design compares outcomes before and after a policy change rather than relying solely on cross-sectional association, it provides considerably stronger grounds for a causal interpretation than the correlational designs that dominate most of this literature.

A meta-analysis by Kane et al. (2007), pooling evidence from over 90 observational studies published up to that point, found that each additional registered nurse per patient day was associated with reduced odds of hospital-related mortality, particularly in intensive care and surgical patient populations where the margin for missed early warning signs is narrowest. Taken together, these four studies, spanning cross-sectional survey-linkage, shift-level exposure analysis, natural-experiment policy evaluation and formal meta-analysis, converge on a consistent direction of effect: lower registered nurse staffing is reliably associated with higher mortality risk, even though the individual study designs vary considerably in the strength of causal claim they can support.

A recurring critical point across this body of work is the difficulty of fully separating the effect of staffing itself from the effect of the hospital-level factors that tend to accompany chronic understaffing, such as under-resourcing of equipment, weaker nurse-physician communication structures, and higher rates of agency or bank staff usage that can disrupt continuity of care even when the raw headcount ratio looks acceptable on paper. None of the four studies discussed above can fully rule out this possibility, since none used a design capable of isolating staffing as a single manipulated variable independent of these correlated organisational conditions; the Queensland natural experiment reported by McHugh et al. (2021) comes closest, because the policy intervention specifically targeted staffing ratios rather than a bundle of quality improvements, but even here secular trends in hospital care occurring alongside the policy change cannot be entirely excluded as an alternative explanation.

Missed Care as a Mediating Mechanism

A second and closely related theme addresses the mechanism through which understaffing is thought to translate into adverse outcomes, namely missed, delayed or “rationed” nursing care. Griffiths et al. (2018), in a systematic review of studies measuring the association between nurse staffing and care omissions, found a consistent relationship between lower staffing levels and higher rates of missed care activities, including delayed medication administration, missed patient mobilisation, and reduced frequency of surveillance activities such as vital sign observation rounds. This body of work reframes understaffing not simply as a resourcing issue but as a direct driver of specific, nameable lapses in fundamental nursing care.

Ball et al. (2018), using RN4CAST data from nine European countries, found that missed care partially mediated the relationship between nurse staffing and post-operative mortality, meaning that lower staffing appeared to increase mortality risk in part because it increased the frequency of specific missed care activities such as delayed pain assessment and infrequent post-operative observation, rather than through some more diffuse or unmeasured pathway. This mediation finding is important because it begins to open what critics have called the “black box” linking aggregate staffing numbers to downstream patient harm.

Direct observational evidence supports this account. Bridges et al. (2019), conducting structured observation of nurse-patient interactions on understaffed and adequately staffed shifts, found that nurses working understaffed shifts spent significantly less time in direct patient interaction and that the interactions that did occur were more likely to be brief, task-focused exchanges rather than the longer, relationship-focused interactions associated with holistic assessment and early deterioration recognition. Dall’Ora et al. (2016), reviewing evidence on shift characteristics more broadly, including twelve-hour shifts and unplanned overtime, found that both were associated with increased missed care and reduced quality of interaction, suggesting that the staffing-missed care relationship is shaped not only by headcount but also by how nursing hours are scheduled and distributed across a shift pattern.

Collectively, this theme suggests that missed care functions as a measurable, intermediate outcome that helps explain why lower nurse staffing translates into worse patient safety outcomes, rather than staffing numbers acting on mortality or adverse events through some unspecified or purely statistical pathway. A persistent methodological limitation, however, is that most missed care measures rely on nurse self-report rather than independent observation, which raises the possibility of recall or social-desirability bias and complicates comparison across studies that use different missed care instruments.

The clinical significance of individual missed-care items also varies considerably, and this heterogeneity is not always reflected in how the construct is aggregated and reported. A missed hourly comfort round is unlikely to carry the same safety consequence as a missed set of post-operative observations or a delayed administration of a time-critical medication such as an antibiotic or anticoagulant, yet many missed-care instruments combine such items into a single composite score. Future primary research disaggregating missed-care items by their plausible clinical consequence, rather than treating all omissions as equivalent, would strengthen the case that missed care is not merely correlated with poor outcomes but is a genuinely explanatory mechanism linking staffing levels to specific, preventable harms.

Skill Mix and Registered Nurse Proportion

A third theme examines whether the proportion of registered nurses within the total nursing workforce, commonly termed skill mix, matters independently of total nursing hours per patient. Blegen et al. (2011), comparing safety-net and non-safety-net United States hospitals, found that a higher proportion of registered nurse hours relative to total nursing hours was associated with fewer patient safety events, and that this relationship was particularly pronounced in non-safety-net hospitals, suggesting that the benefit of a richer skill mix may interact with broader organisational resourcing.

Twigg, Gelder and Myers (2015), studying Australian acute hospitals, found that shifts falling below target registered nurse hours were associated with increases in nurse-sensitive outcomes including failure to rescue, even after accounting for total nursing hours delivered by the ward, indicating that registered nurse hours specifically, rather than total nursing hours of any grade, carried the strongest protective association. This distinction matters directly for UK workforce policy, where nursing associate roles have been introduced partly as a response to registered nurse shortages, and where the safety implications of substituting registered nurse hours with support-worker hours remain only partially evidenced.

Rafferty et al. (2007), analysing survey and discharge data from English acute hospitals, found that hospitals with higher patient-to-nurse ratios had higher risk-adjusted mortality and that nurses in these hospitals reported lower confidence in the quality of care delivered on their ward, providing an early UK-specific parallel to the largely international evidence reviewed above. Cho et al. (2015), studying South Korean hospitals, extended this picture by showing that nurse staffing, practice environment quality and the proportion of nurses holding a bachelor’s-level qualification jointly predicted patient mortality, with a richer educational skill mix appearing to moderate, though not eliminate, the mortality risk associated with lower staffing.

Read together, this theme indicates that simple headcount ratios are an incomplete proxy for safety-relevant staffing adequacy; the grade, qualification level and specifically the registered nurse proportion of the nursing workforce appear to carry independent safety significance, a finding with direct relevance to current UK debates about the pace and safety implications of nursing associate role expansion as a substitute for registered nurse recruitment.

It is worth noting that skill mix and total staffing are not independent variables in most of the studies reviewed here, since trusts facing registered nurse shortages often respond by increasing support-worker numbers to maintain an acceptable total headcount, meaning that a favourable skill-mix effect can be partly confounded with an underlying registered nurse shortfall that the skill-mix ratio is itself a symptom of. Disentangling these two effects statistically requires either a study design with independent variation in total staffing and skill mix, which is rare in observational workforce data, or a natural experiment of the kind McHugh et al. (2021) were able to exploit, underscoring again how much of the strongest available evidence in this field depends on unusual policy or administrative circumstances rather than being routinely producible through conventional cross-sectional survey research.

Nurse Staffing, Workforce Wellbeing and the Safety Climate

A fourth theme concerns the reciprocal relationship between nurse staffing, nurse wellbeing and the broader organisational safety climate. Buchan, Catton and Shaffer (2022), reporting on the global nursing workforce in the aftermath of the COVID-19 pandemic, documented widespread staffing shortages, elevated burnout and rising intention-to-leave among registered nurses internationally, and argued that chronic understaffing and workforce attrition had become mutually reinforcing rather than separate problems. The Royal College of Nursing (2021) reached a similar conclusion in a UK-specific report, describing a self-perpetuating cycle in which persistent understaffing increases the workload and moral distress of remaining staff, which in turn increases sickness absence and staff turnover, further eroding effective staffing levels on the wards most affected.

Griffiths et al. (2019), in a large retrospective longitudinal cohort study of English hospitals, examined the more specific question of how nursing assistant staffing interacts with registered nurse staffing in relation to mortality, finding a complex pattern in which higher nursing assistant staffing was associated with increased mortality risk specifically on wards where registered nurse staffing was already low, consistent with an interpretation that support-worker hours cannot straightforwardly substitute for registered nurse hours when overall registered nurse capacity is already constrained. Dall’Ora et al. (2016), discussed above in relation to missed care, also found that demanding shift patterns, including consecutive long shifts and short turnaround times between shifts, were independently associated with nurse burnout, adding a further layer to the staffing-wellbeing-safety relationship.

The National Institute for Health and Care Excellence (2014) safe staffing guidance for adult inpatient wards explicitly recognised this feedback dynamic, recommending that staffing establishments be reviewed using evidence-based acuity tools rather than fixed historical ratios, and that persistent gaps between planned and actual staffing be treated as a recurring safety indicator requiring escalation rather than a routine operational variance to be tolerated. Considered together, this theme suggests that nurse staffing, wellbeing and the wider organisational safety climate should be understood as interdependent rather than as staffing being simply an upstream input to an otherwise separate safety outcome.

UK Policy Responses and the Case of Mandated Ratios

A fifth and more policy-focused theme concerns how health systems have translated the staffing-safety evidence reviewed above into regulation, and what can be learned from the resulting natural experiments. Wales enacted the Nurse Staffing Levels (Wales) Act, which places a legal duty on health boards to have regard to nationally agreed workforce planning triangulation methods, including professional judgement and evidence-based tools, when setting registered nurse establishments on specified adult acute medical and surgical wards. This places Wales alongside jurisdictions such as California and, more recently, Queensland, evaluated by McHugh et al. (2021), as one of a small number of health systems to move from voluntary guidance to a statutory staffing duty.

England, by contrast, has followed the approach set out by the National Institute for Health and Care Excellence (2014), which recommends locally determined establishments informed by evidence-based acuity and dependency tools rather than a single mandated ratio, on the grounds that a fixed national ratio risks being too blunt an instrument across wards with very different patient acuity profiles. The Royal College of Nursing (2021) has been broadly critical of this more flexible approach, arguing that without a statutory duty, safe staffing recommendations are too easily deprioritised when trusts face simultaneous financial and recruitment pressures, a concern amplified by the workforce shortages documented internationally by Buchan, Catton and Shaffer (2022) in the years following the COVID-19 pandemic.

The comparative value of the Welsh case for this review lies less in demonstrating that mandated ratios work in principle, since McHugh et al.’s (2021) Queensland findings already provide reasonably strong causal evidence for that, and more in the fact that Wales offers a UK-specific test case using an NHS-comparable funding and workforce model, rather than the US or Australian systems from which most of the strongest causal evidence to date has been drawn. A rigorous, published evaluation of patient safety outcomes before and after the Welsh Act’s phased implementation, ideally using a similar shift-level or natural-experiment design to McHugh et al. (2021), would therefore make a disproportionately valuable contribution to UK-specific staffing policy, and its comparative absence from the published literature at the time of writing is one of the more significant gaps identified in this review.

Gaps in the Evidence Base

• The strongest causal evidence, drawn from natural-experiment ratio legislation such as that evaluated by McHugh et al. (2021), comes almost entirely from outside the UK; within the UK, only Wales has equivalent mandated staffing legislation, and robust, published post-implementation evaluation of its patient safety impact remains limited.
• Missed care measures are heterogeneous and predominantly self-reported rather than independently observed, which limits comparability across studies and may under- or over-state the true strength of the staffing-missed care relationship.
• Non-acute settings, including community nursing, mental health inpatient care and primary care, are comparatively under-researched relative to the volume of evidence available for acute adult inpatient wards.
• Long-term and patient-reported outcomes, such as functional recovery and patient-reported experience, are underexamined relative to the mortality- and adverse-event-focused designs that dominate this literature.
• The comparative cost-effectiveness of mandated minimum ratios versus flexible, acuity-based staffing tools such as the Safer Nursing Care Tool used across much of the English NHS has not been well established.
• Skill-mix substitution involving the expanding nursing associate role is a rapidly evolving UK workforce policy area with almost no dedicated, UK-specific patient safety outcome evidence published to date.

Conclusion

This review has synthesised a substantial and methodologically diverse body of evidence indicating that lower registered nurse staffing is consistently associated with worse patient safety outcomes in acute adult inpatient care, with the strongest causal evidence coming from natural-experiment ratio legislation. Missed or rationed nursing care and cumulative nurse burnout appear to function as key intermediate mechanisms linking staffing numbers to downstream patient harm, and skill mix, specifically the proportion of registered nurse hours rather than total nursing hours of any grade, matters independently of raw headcount ratios.

For UK policy and practice, the Welsh legislative approach offers a natural comparator that English policymakers have arguably not yet fully exploited in evaluating the case for a more prescriptive staffing model. The evidence gaps identified above, particularly around non-acute settings, the cost-effectiveness of alternative staffing models, and the safety implications of expanding nursing associate roles, should guide the next generation of UK-focused nurse staffing research, and any future primary study in this area should aim explicitly to strengthen causal inference through longitudinal, natural-experiment or shift-level exposure designs rather than relying further on cross-sectional survey association alone.

For nursing and healthcare management students engaging with this evidence base directly, the practical takeaway is twofold. First, the case for treating registered nurse staffing as a first-order patient safety indicator, rather than a purely financial or operational variable, is now well supported across multiple study designs and national contexts, and should be reflected accordingly in local risk registers and safety governance structures. Second, any locally commissioned staffing evaluation should where possible attempt to measure missed care and workforce wellbeing alongside mortality and adverse-event outcomes, since the theme-by-theme synthesis above indicates that these intermediate variables, rather than staffing numbers alone, are where much of the explanatory and actionable detail in this literature actually resides.

It is also worth restating clearly what this review does not claim. The evidence synthesised above supports registered nurse staffing as one significant, evidenced and modifiable contributor to patient safety among several, rather than as the single determining factor in every adverse outcome; ward-level culture, individual clinical judgement, medical staffing and wider organisational safety systems all continue to matter alongside nursing establishment levels, and none of the studies discussed in this review claim, nor should this review be read as claiming, that adequate staffing alone is sufficient on its own to ensure safe care in the absence of these other conditions.

References

Aiken, L.H., Sloane, D.M., Bruyneel, L., Van den Heede, K., Griffiths, P., Busse, R., Diomidous, M., Kinnunen, J., Kózka, M., Lesaffre, E., McHugh, M.D., Moreno-Casbas, M.T., Rafferty, A.M., Schwendimann, R., Scott, P.A., Tishelman, C., van Achterberg, T. and Sermeus, W. (2014) ‘Nurse staffing and education and hospital mortality in nine European countries: a retrospective observational study’, The Lancet, 383(9931), pp. 1824–1830.

Ball, J.E., Bruyneel, L., Aiken, L.H., Sermeus, W., Sloane, D.M., Rafferty, A.M., Lindqvist, R., Tishelman, C. and Griffiths, P. (2018) ‘Post-operative mortality, missed care and nurse staffing in nine countries: a cross-sectional study’, International Journal of Nursing Studies, 78, pp. 10–15.

Blegen, M.A., Goode, C.J., Spetz, J., Vaughn, T. and Park, S.H. (2011) ‘Nurse staffing effects on patient outcomes: safety-net and non-safety-net hospitals’, Medical Care, 49(4), pp. 406–414.

Bridges, J., Griffiths, P., Oliver, E. and Pickering, R.M. (2019) ‘Hospital nurse staffing and staff-patient interactions: an observational study’, BMJ Quality & Safety, 28(9), pp. 706–713.

Buchan, J., Catton, H. and Shaffer, F.A. (2022) Sustain and Retain in 2022 and Beyond: The Global Nursing Workforce and the COVID-19 Pandemic. Geneva: International Council of Nurses.

Cho, E., Sloane, D.M., Kim, E.Y., Kim, S., Choi, M., Yoo, I.Y., Lee, H.S. and Aiken, L.H. (2015) ‘Effects of nurse staffing, work environments, and education on patient mortality: an observational study’, International Journal of Nursing Studies, 52(2), pp. 535–542.

Dall’Ora, C., Ball, J., Recio-Saucedo, A. and Griffiths, P. (2016) ‘Characteristics of shift work and their impact on employee performance and wellbeing: a literature review’, International Journal of Nursing Studies, 57, pp. 12–27.

Francis, R. (2013) Report of the Mid Staffordshire NHS Foundation Trust Public Inquiry. London: The Stationery Office.

Griffiths, P., Recio-Saucedo, A., Dall’Ora, C., Briggs, J., Maruotti, A., Meredith, P., Smith, G.B. and Ball, J. (2018) ‘The association between nurse staffing and omissions in nursing care: a systematic review’, Journal of Advanced Nursing, 74(7), pp. 1474–1487.

Griffiths, P., Maruotti, A., Recio Saucedo, A., Redfern, O.C., Ball, J.E., Briggs, J., Dall’Ora, C., Schmidt, P.E. and Smith, G.B. (2019) ‘Nurse staffing, nursing assistants and hospital mortality: retrospective longitudinal cohort study’, BMJ Quality & Safety, 28(8), pp. 609–617.

Kane, R.L., Shamliyan, T.A., Mueller, C., Duval, S. and Wilt, T.J. (2007) ‘The association of registered nurse staffing levels and patient outcomes: systematic review and meta-analysis’, Medical Care, 45(12), pp. 1195–1204.

McHugh, M.D., Aiken, L.H., Sloane, D.M., Windsor, C., Douglas, C. and Yates, P. (2021) ‘Effects of nurse-to-patient ratio legislation on nurse staffing and patient mortality, readmissions, and length of stay: a prospective study’, The Lancet, 397(10288), pp. 1905–1913.

National Institute for Health and Care Excellence (2014) Safe Staffing for Nursing in Adult Inpatient Wards in Acute Hospitals. London: NICE.

Needleman, J., Buerhaus, P., Pankratz, V.S., Leibson, C.L., Stevens, S.R. and Harris, M. (2011) ‘Nurse staffing and inpatient hospital mortality’, New England Journal of Medicine, 364(11), pp. 1037–1045.

Rafferty, A.M., Clarke, S.P., Coles, J., Ball, J., James, P., McKee, M. and Aiken, L.H. (2007) ‘Outcomes of variation in hospital nurse staffing in English hospitals: cross-sectional analysis of survey data and discharge records’, International Journal of Nursing Studies, 44(2), pp. 175–182.

Royal College of Nursing (2021) Safe and Effective Staffing: Nursing Against the Odds. London: RCN.

Twigg, D.E., Gelder, L. and Myers, H. (2015) ‘The impact of understaffed shifts on nurse-sensitive outcomes’, Journal of Advanced Nursing, 71(7), pp. 1564–1572.