Fall Prevention Strategies and Their Effectiveness in Reducing Inpatient Falls: A Retrospective Analysis

Introduction

Falls are consistently identified as one of the most common adverse events occurring in acute care hospitals, with an estimated 700,000 to 1,000,000 inpatient falls occurring annually in United States hospitals alone (Bouldin et al., 2013). Between 30% and 51% of inpatient falls result in some degree of physical injury, and roughly 6% to 9% result in serious injury such as fracture, intracranial hemorrhage, or laceration requiring surgical repair (Currie, 2008). Beyond the direct physical harm to patients, injurious falls are associated with extended hospital length of stay, increased likelihood of discharge to a skilled nursing facility rather than home, and substantial additional cost, with the Centers for Medicare and Medicaid Services designating certain fall-related injuries as hospital-acquired conditions ineligible for additional reimbursement (Wong et al., 2011). These clinical and financial consequences have positioned fall prevention as a persistent priority within nursing-sensitive quality measurement and hospital accreditation standards.

The etiology of inpatient falls is widely recognized as multifactorial, involving the interaction of intrinsic patient factors, including advanced age, altered mental status, orthostatic hypotension, impaired mobility, and polypharmacy, with extrinsic environmental factors such as unfamiliar surroundings, inadequate lighting, and clutter near the bedside (Oliver et al., 2010). Because no single risk factor accounts for the majority of inpatient falls, prevention strategies have increasingly moved away from single-intervention approaches toward multicomponent programs that combine standardized risk assessment, environmental modification, and behavioral or process-based interventions delivered by nursing and interdisciplinary staff (Hempel et al., 2013). Commonly implemented components include validated fall-risk screening instruments such as the Morse Fall Scale or Hendrich II model, bed and chair exit alarms, structured hourly rounding protocols addressing the “4 Ps” of pain, positioning, personal needs, and placement of items, non-slip footwear, and visual signage identifying high-risk patients to all members of the care team (Morse, 2009).

Despite widespread adoption of multicomponent fall prevention programs, the evidence base evaluating their comparative effectiveness remains mixed. A 2013 systematic review found that while multicomponent interventions were generally associated with reduced fall rates, effect sizes varied considerably across studies, and several individual components, particularly bed alarms and generic patient education materials, showed limited independent effect when isolated from the broader program (Hempel et al., 2013). Similarly, a large multi-site cluster-randomized trial found no significant reduction in fall rates attributable to a structured risk-assessment-driven intervention, raising questions about whether risk stratification alone, absent consistent behavioral follow-through, meaningfully changes fall incidence (Cumming et al., 2008). These inconsistent findings suggest that implementation fidelity, rather than program design alone, may substantially influence whether a given fall prevention strategy translates into measurable reduction in patient harm.

Compounding this uncertainty is the observation that many published evaluations of fall prevention programs report aggregate, hospital-wide fall-rate change without examining whether reductions are distributed evenly across unit types, shifts, and fall circumstances, or are instead concentrated among certain subgroups while others remain largely unaffected (Currie, 2008). A program that appears highly effective when evaluated at the level of an entire facility may nonetheless leave specific, high-risk categories of falls, such as those occurring overnight or during toileting, essentially unchanged, a distinction with direct implications for how hospitals prioritize additional resources once a baseline prevention program has been implemented.

The purpose of this retrospective analysis was to examine fall rates before and after implementation of a multicomponent fall prevention program across a multi-hospital health system and to identify which specific program components, and which patterns of staff compliance, were most strongly associated with reductions in overall and injurious fall incidence. In addition to evaluating aggregate fall-rate change, this analysis examined variation in effect by unit type, nursing shift, and circumstances surrounding the fall, with the goal of identifying persistent gaps that multicomponent programs, as currently structured, may not adequately address.

Methods

This retrospective analysis was conducted across six acute care hospitals within a single regional health system, encompassing a combined 1,842 licensed beds. Data were drawn from three sources: standardized fall incident reports submitted through the health system’s electronic event-reporting platform, structured fall-risk assessment and hourly rounding documentation extracted from the electronic health record, and unit-level nurse staffing ratios obtained from workforce management records. The analysis period spanned five years, including a two-year baseline period preceding system-wide implementation of a multicomponent fall prevention program and a three-year post-implementation period.

The fall prevention program under evaluation incorporated five core components consistent with those described in prior implementation literature (Spoelstra et al., 2012): universal fall-risk screening using the Morse Fall Scale at admission, each shift, and following any change in patient condition; a standardized hourly rounding protocol addressing pain, positioning, personal needs, and proximity of the call light and personal items; bed and chair exit alarms activated for all patients scoring in the moderate- or high-risk range; universal non-slip footwear issued at admission; and standardized visual signage, including door and wristband indicators, identifying patients at elevated fall risk to all staff and visiting personnel.

Fall events were classified by injury severity using the National Database of Nursing Quality Indicators (NDNQI) injury classification, ranging from no injury through major injury or death, and were further characterized by unit type, time of day and shift, and documented circumstances surrounding the fall, including whether the fall occurred during toileting, transfer, or ambulation, or was unwitnessed (Bouldin et al., 2013). Fall rates were calculated as the number of falls per 1,000 patient days, consistent with standard nursing-sensitive quality indicator methodology, and injurious fall rates were calculated using the same denominator restricted to falls resulting in at least minor injury. Unit-level compliance with hourly rounding and fall-risk reassessment documentation was calculated as the proportion of expected rounding or reassessment entries actually completed, and units were subsequently stratified into high-compliance (≥85%) and lower-compliance (<85%) groups for comparative analysis. Given the retrospective, non-randomized design of this analysis, statistical comparisons were limited to descriptive rate comparisons and chi-square testing for categorical differences between the pre- and post-implementation periods; no adjustment for potential confounding secular trends was performed.

Results

A total of 4,157 fall incident reports were identified across the five-year study period, comprising 2,486 falls during the two-year baseline period and 1,671 falls during the three-year post-implementation period. The overall fall rate declined from 4.62 falls per 1,000 patient days during the baseline period to 2.85 falls per 1,000 patient days following program implementation, a relative reduction of 38.4%. The injurious fall rate declined from 1.41 to 0.76 falls per 1,000 patient days over the same period, a relative reduction of 46.1%, and the rate of falls resulting in moderate or major injury declined from 0.29 to 0.14 falls per 1,000 patient days, consistent with the magnitude of reduction reported in comparable multicomponent program evaluations (Dykes et al., 2010).

Reduction in fall incidence was not uniform across unit types. Medical-surgical units, which accounted for the largest proportion of total patient days, showed a 41.2% relative reduction in overall fall rate, while neurology and rehabilitation units, which historically carry disproportionately high baseline fall risk due to impaired mobility and cognitive status, showed a comparatively smaller reduction of 27.6%, a pattern broadly consistent with prior observations that units with the highest baseline acuity and risk burden tend to show more modest relative improvement even when absolute fall counts decline substantially (Oliver et al., 2010). Critical care units, which maintain closer nurse-to-patient ratios and continuous monitoring at baseline, showed the smallest relative reduction of 18.3%, likely reflecting a ceiling effect given already low baseline fall rates in this setting.

Analysis stratified by hourly rounding and reassessment documentation compliance revealed a strong association between implementation fidelity and outcome magnitude. Units achieving 85% or greater compliance with hourly rounding documentation during the post-implementation period showed a 51.7% relative reduction in overall fall rate, compared to a 24.9% reduction on units with compliance below this threshold, a difference of more than twofold that was statistically significant (χ² = 38.6, p < .001). This pattern is consistent with prior implementation research suggesting that the behavioral and process components of fall prevention programs, rather than screening or signage alone, account for a disproportionate share of the achievable reduction in fall incidence (Krauss et al., 2008).

Examination of the circumstances surrounding individual fall events indicated that toileting-related falls, including falls occurring during ambulation to the bathroom, transfer on or off the toilet, and unassisted attempts to reach the bathroom, accounted for 34.7% of all falls in the baseline period and remained the single largest circumstantial category following program implementation, declining only modestly to 31.2% of all post-implementation falls, a smaller relative reduction than was observed for falls associated with other circumstances such as transfer or ambulation more broadly (Tzeng, 2010). Falls occurring during night shift, defined as 2300 to 0700, similarly showed a smaller relative reduction than falls occurring during day or evening shifts, and continued to account for a disproportionate 42.3% of all post-implementation falls despite representing a substantially smaller proportion of total patient care hours, a finding consistent with prior reports linking reduced staff visibility, patient disorientation upon waking, and lower nurse-to-patient ratios overnight to elevated night-shift fall risk (Krauss et al., 2008; Tzeng, 2010).

Bed and chair alarm activation was documented in 76.4% of post-implementation falls among patients identified as moderate or high risk, indicating that alarm activation alone did not prevent the majority of falls among appropriately identified at-risk patients, a finding consistent with prior evidence suggesting that exit alarms function primarily as a notification mechanism rather than a physical barrier to falling, and are effective only when paired with a sufficiently rapid staff response (Shorr et al., 2012). Among the subset of falls for which post-fall huddle documentation was available, staff-reported contributing factors most frequently included patient attempts to mobilize independently despite instructions to call for assistance, delayed staff response following alarm activation, and, less frequently, incomplete or outdated risk-level signage following a change in patient condition.

Discussion

The findings of this retrospective analysis indicate that implementation of a multicomponent fall prevention program was associated with substantial reductions in both overall and injurious inpatient fall rates, with effect sizes broadly consistent with those reported in prior multi-site evaluations of similar programs (Dykes et al., 2010; Spoelstra et al., 2012). The magnitude of reduction observed here, particularly the disproportionately larger decline in injurious relative to overall falls, suggests that the program may have been especially effective in mitigating fall severity even in instances where a fall was not fully prevented, potentially reflecting the combined influence of non-slip footwear, environmental decluttering, and more rapid staff response associated with structured rounding.

The strong association observed between hourly rounding compliance and outcome magnitude reinforces a recurring theme in the fall prevention literature: that the behavioral and process fidelity with which a program is delivered at the bedside exerts a greater influence on outcomes than the presence of any individual program component in isolation (Hempel et al., 2013; Krauss et al., 2008). Units achieving high rounding compliance realized more than twice the relative reduction observed on lower-compliance units, a gap substantially larger than would be expected from documentation differences alone, suggesting that rounding compliance may serve as a proxy for broader unit-level safety culture and staff engagement with the prevention program rather than functioning as an isolated causal mechanism. This interpretation is consistent with prior work proposing that structured rounding functions less as a stand-alone intervention and more as an organizing framework that increases the frequency and consistency with which other program elements, including toileting assistance and environmental checks, are actually delivered (Krauss et al., 2008).

The persistence of toileting-related and night-shift falls despite overall program effectiveness identifies a clear priority area for future intervention. Toileting-related falls have been consistently identified in prior literature as difficult to prevent through standard rounding alone, given that the urge to void frequently arises between scheduled rounds and that patients, particularly those with cognitive impairment, may attempt to reach the bathroom independently despite prior instruction (Tzeng, 2010). Interventions specifically targeting this circumstance, such as scheduled toileting protocols independent of the standard rounding interval, bedside commode placement for patients with mobility limitations, and improved call-light response time tracking, may be necessary to achieve further reduction beyond what a general rounding-based program can offer. Similarly, the disproportionate representation of night-shift falls suggests that current staffing patterns and environmental design, including lighting adjustments and nurse station sightlines, may warrant reassessment specifically for the overnight period rather than assuming that daytime-optimized interventions translate equally to night-shift conditions (Krauss et al., 2008).

The finding that bed and chair alarms were activated in the majority, but not all, of falls among identified at-risk patients, and that alarm activation alone did not prevent most of these events, is consistent with prior evidence characterizing exit alarms as a notification rather than a preventive mechanism in the strictest sense (Shorr et al., 2012). This finding suggests that continued investment in alarm technology alone, absent corresponding attention to staff response time and unit-level staffing adequacy, is unlikely to yield substantial further reduction in fall incidence, and that resource allocation may be better directed toward interventions addressing response time or, for the highest-risk patients, direct observation.

This analysis is subject to several limitations inherent to its retrospective, non-randomized design. Fall incident reporting is known to be subject to underreporting, and the extent of underreporting may have differed between the baseline and post-implementation periods if the introduction of the prevention program was accompanied by heightened reporting awareness, potentially causing the true magnitude of fall-rate reduction to be underestimated. The absence of a concurrent, non-implementing control group limits the ability to fully exclude secular trends, including concurrent staffing changes, unrelated quality initiatives, or gradual shifts in patient acuity, as partial explanations for the observed reduction. Additionally, compliance stratification relied on documentation completion rather than direct observation of rounding quality, and it is possible that documentation compliance imperfectly reflects the true consistency or quality of rounding interactions at the bedside. Finally, because the health system implemented all five program components simultaneously rather than in a staggered rollout, this analysis cannot isolate the independent contribution of any single component, and the observed reduction should be interpreted as the effect of the bundle as a whole rather than as evidence for or against any individual element in isolation.

Future research would benefit from prospective, multi-site designs capable of isolating the independent contribution of specific program components, direct observational assessment of rounding quality rather than reliance on documentation alone, and evaluation of targeted interventions for toileting-related and night-shift falls specifically, given that these circumstances appear to represent the residual burden least responsive to standard multicomponent programs as currently designed. Taken together, these findings support the continued use of multicomponent fall prevention programs as a core patient safety strategy while highlighting implementation fidelity, toileting-related risk, and night-shift conditions as priority areas warranting focused, additional intervention.

References

Bouldin, E. L. D., Andresen, E. M., Dunton, N. E., Simon, M., Waters, T. M., Liu, M., Daniels, M. J., Mion, L. C., & Shorr, R. I. (2013). Falls among adult patients hospitalized in the United States: Prevalence and trends. Journal of Patient Safety, 9(1), 13–17.

Cumming, R. G., Sherrington, C., Lord, S. R., Simpson, J. M., Vogler, C., Cameron, I. D., & Naganathan, V. (2008). Cluster randomised trial of a targeted multifactorial intervention to prevent falls among older people in hospital. BMJ, 336(7647), 758–760.

Currie, L. (2008). Fall and injury prevention. In R. G. Hughes (Ed.), Patient safety and quality: An evidence-based handbook for nurses. Agency for Healthcare Research and Quality.

Dykes, P. C., Carroll, D. L., Hurley, A., Lipsitz, S., Benoit, A., Chang, F., Meltzer, S., Tsurikova, R., Zuyov, L., & Middleton, B. (2010). Fall prevention in acute care hospitals: A randomized trial. JAMA, 304(17), 1912–1918.

Hempel, S., Newberry, S., Wang, Z., Booth, M., Shanman, R., Johnsen, B., Shier, V., Saliba, D., Spector, W. D., & Ganz, D. A. (2013). Hospital fall prevention: A systematic review of implementation, components, adherence, and effectiveness. Journal of the American Geriatrics Society, 61(4), 483–494.

Krauss, M. J., Nguyen, S. L., Dunagan, W. C., Birge, S., Costantinou, E., Johnson, S., Caleca, B., & Fraser, V. J. (2008). Circumstances of patient falls and injuries in nine hospitals in a midwestern healthcare system. Infection Control & Hospital Epidemiology, 28(5), 544–550.

Morse, J. M. (2009). Preventing patient falls: Establishing a fall intervention program (2nd ed.). Springer Publishing Company.

Oliver, D., Healey, F., & Haines, T. P. (2010). Preventing falls and fall-related injuries in hospitals. Clinics in Geriatric Medicine, 26(4), 645–692.

Shorr, R. I., Chandler, A. M., Mion, L. C., Waters, T. M., Liu, M., Daniels, M. J., Kessler, L. A., & Miller, S. T. (2012). Effects of an intervention to increase bed alarm use to prevent falls in hospitalized patients. Annals of Internal Medicine, 157(10), 692–699.

Spoelstra, S. L., Given, B. A., & Given, C. W. (2012). Fall prevention in hospitals: An integrative review. Clinical Nursing Research, 21(1), 92–112.

Tzeng, H. M. (2010). Understanding the prevalence of inpatient falls associated with toileting in adult acute care settings. Journal of Nursing Care Quality, 25(1), 22–30.

◆