Data analysis. To determine whether there were statistically significant associations between the outcome variable and the potential predictive variables, bivariate analyses (χ 2 tests) were used; t tests could not be used to examine the continuous variable (age) between patients with and without fall-related injury because the sample was not random—univariate logistic regression was used instead. The categorical variables (sex, fall history, use of diuretics, use of CNS medications, cognitive impairment, primary discharge diagnoses, abnormal laboratory values, impaired mobility, and BMI) were compared using the χ 2 test of association. Analyses were conducted using SPSS statistical software, version 22. Statistical significance was set at P ≤ 0.05. However, previous studies have considered P ≤ 0.2 to be sufficiently significant to warrant further analysis using multivariate modeling.20, 24 Thus, variables meeting this criterion were further explored using multivariate logistic regression.
A total of 1,369 falls in 1,369 patients were included in this study. Of these, 381 falls (27.8%) resulted in some form of injury. The study sample comprised 747 men (54.6%) and 622 women (45.4%). The mean age of all subjects was 55.1 years; the mean age of subjects who experienced injurious falls was 55.5 years. Data for all variables except BMI were available for all subjects; BMI data were available for 1,228 subjects.
Results of the univariate logistic regression and bivariate analyses are presented in Table 3. The primary discharge diagnosis of “symptoms, signs, and ill-defined conditions” was the only factor to have a statistically significant association with injurious falls (P = 0.019). Of the 64 subjects with that discharge code, 40.6% were injured. Of the 1,305 subjects with a different primary discharge diagnosis, only 27.2% suffered injury. No statistically significant associations with injurious falls were found between other variables examined.
In this study, six variables had a significance level of P ≤ 0.2, including abnormal laboratory values; primary discharge diagnoses including injury and poisoning; impaired mobility; the use of CNS medications; primary discharge diagnoses of diseases of the respiratory system; and primary discharge diagnoses including endocrine, nutritional, metabolic, and immunity disorders. These variables were further examined using multivariate logistic regression. To ensure that there was no collinearity between these variables and the variables of age, sex, and fall history, we included all of them in the multivariate analysis. Accordingly, patients with a primary discharge diagnosis of symptoms, signs, and ill-defined conditions had significantly higher odds of experiencing an injurious fall (P = 0.037). This association remained significant after adjusting for other predictors in the logistic regression model. See Table 4 for detailed results of the multivariate analyses.
This study examined falls resulting in all levels of injury. Although it seems high, the percentage of patients who sustained injurious falls (27.8%) is consistent with previous estimates indicating that between 6% and 44% of falls in the acute care setting result in injury.1 This study found that patients with the primary discharge diagnosis of symptoms, signs, and ill-defined conditions were significantly more likely to be injured if they fell. This category includes signs and symptoms that suggest the involvement of more than one disease or system, or that have an unknown etiology.27 For patients with this primary discharge diagnosis, either a more precise diagnosis could not be made or the patient was admitted for the sole purpose of treating the presenting problem, without treatment or further evaluation of the underlying disease. This diagnostic category can include nausea and vomiting, alterations in consciousness, convulsions, dizziness, fatigue or malaise, sleep disturbances, lack of coordination, paresthesia, abnormal weight loss, urinary incontinence, and abnormal blood chemistries.28 For example, a patient with cancer who is receiving outpatient chemotherapy might be admitted for nausea and vomiting. Treatment would address the nausea and vomiting, but not the underlying causes (cancer and chemotherapy).
The higher risk of injurious falls in patients with a primary discharge diagnosis of symptoms, signs, and ill-defined conditions might be related to the consequences of such conditions. For instance, patients with urinary incontinence might visit the bathroom more often, where hard surfaces and objects (such as sinks and toilets) are present and likely to be struck during a fall. Patients with poor coordination, paresthesia, or extreme fatigue might have difficulty protecting themselves during a fall. Either scenario could increase the likelihood of injury.
This study was the first to evaluate diuretics use, impaired mobility, and BMI as fall risk factors among hospitalized patients. Although studies conducted in other settings have found associations between diuretics use and injurious falls,22, 24, 29, 30 this study found no such association. This may be because hospitalized patients are likely to have better access to assistive devices, such as bedside commodes, and to receive toileting assistance from nursing staff. Safety features such as grab bars might also be more available to these patients.
To our knowledge, only one other study has considered a possible association between injurious falls and CNS medications in hospitalized patients. Pierce and colleagues found a statistically significant association between the administration of narcotics and injurious falls,15 whereas our study did not. The difference in findings may stem from differences in the CNS medications selected: the study by Pierce and colleagues looked at narcotics, benzodiazepines, antihistamines, and zolpidem; our study looked at narcotics, benzodiazepines, barbiturates, neuroleptics, and antidepressants.
Although our study was similar to two studies that reported an association between cognitive impairment and injurious falls,12, 15 we did not find such an association. The difference in findings may stem from differences in how cognitive impairment was defined. Fischer and colleagues used variables such as residence on a geriatric psychiatry floor and level of confusion based on subjective observation as indicators,12 while Pierce and colleagues used prefall confusion.15 Our study defined cognitive impairment more broadly, using the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) coding.28 In the study by Fischer and colleagues, patients on the geriatric psychiatric floor had the highest fall rate.12 As a group, they may have been considerably older and have had more comorbidities than the patients with cognitive impairment in our study, making injurious falls more likely.
In the study by Pierce and colleagues, patients with prefall confusion were found to be more likely to suffer an injurious fall than patients without such confusion.15 Patients who had received narcotics within the 24 hours preceding a fall were also more likely to suffer injury; and of patients who had injurious falls, 79% were under the age of 65 years. Although Pierce and colleagues don't discuss whether administration of narcotics could have caused some of the prefall confusion, this seems possible.
Although two studies in community settings found a significant association between impaired mobility and injurious falls,18, 21 our study did not. One reason for this difference in findings may be that hospitalized adults with mobility issues have better access to assistive devices and receive more assistance during ambulation than their community-dwelling counterparts. And hospitalized adults with mobility issues are often considered to be at risk for falling. For such patients, active fall prevention efforts by nursing staff may reduce the frequency of falls and therefore the likelihood of fall-related injury.
Abnormal laboratory values indicative of altered coagulation, such as low platelet counts and elevated prothrombin times, warrant particular attention, as intracranial injuries (including intracranial hemorrhage) are more often fatal than other types of fall-related injuries.31 That said, our study found no association between either abnormal prothrombin times or low platelet counts and injurious falls. This finding is supported by similar findings in the study by Pierce and colleagues, who also reported no such associations.15 But a study by Bond and colleagues, using international normalized ratio (INR) values, found that abnormal INR values did contribute to the risk of hemorrhagic injury during a fall.10 The difference in findings may have to do with which laboratory values were examined. And in hospitals that flag abnormal laboratory values as a falls risk factor, nursing staff may be making an extra effort to prevent these patients from falling.
No associations between BMI and injurious falls were found. A large proportion of the sample subjects were underweight (78.7%), while fewer than 1% of patients were overweight or obese. This relative lack of variability may have contributed to the insignificant finding. The large proportion of underweight patients in the sample may reflect the patient populations at the study site. For example, the study site has a large population of oncology patients, for whom weight loss is often a side effect of treatment.
Limitations. This study examined injurious falls occurring at a single institution. Thus, these results may not be generalizable to other settings. Variables reported in the literature as associated with injurious falls were the only variables included in this study. There may be other variables associated with injurious falls. To capture falls for study, we relied solely on the study site's adverse event reporting system, which may not have captured all falls that occurred during the study period. Also, several variables were identified through EMR coding, and EMRs aren't always complete or comprehensive. We did not consider possible environmental factors, such as fall location. Lastly, it's important to note that although some associations were found between certain variables and injurious falls, associations do not infer causation.
IMPLICATIONS AND CONCLUSION
The results of this study have several implications for clinicians. We found a significant association between a primary discharge diagnosis of symptoms, signs, and ill-defined conditions and injurious falls; indeed, 40.6% of patients in this category suffered injurious falls. This suggests that clinicians should pay particular attention to patients admitted primarily for treatment of presenting signs or symptoms without further evaluation, or for whom a more precise diagnosis can't be made. Examples include patients admitted with conditions such as nausea and vomiting, alterations in consciousness, and seizures, or for signs such as electrolyte imbalances (for example, hyponatremia, elevated ammonia levels, and hyperkalemia). Decisions about care and treatment with regard to preventing falls and injury should take these diagnoses into consideration.
It's also important not to dismiss clinically significant findings that may lack statistical significance. As Yet and colleagues have cautioned, relying purely on data-driven approaches to prediction “may not provide either accurate predictions or the insights required for improved decision making.”32 In our study, a large proportion of the total sample (N = 1,369) who took CNS medications, had abnormal laboratory values (elevated prothrombin times or low platelet counts), or were underweight (BMI below 18.5) suffered injurious falls (13.7%, 11%, and 21.9%, respectively), although the associations lacked statistical significance. Hospital fall prevention programs often don't consider these variables. It's imperative that nurses be aware of patients for whom these factors are relevant, so that appropriate interventions can be used.
Patients at risk for injurious falls may benefit from interventions designed to offer protection from hard surfaces during a fall, such as floor mats, low beds, hip protectors, and protective helmets or caps. An awareness of which patient factors are associated with injurious falls will help clinicians provide the most appropriate interventions to the patients who most need them. Given that preventive resources are often limited, it's critical that such resources be used with the right patients at the right time.
Further research. As our study appears to be the first hospital-based study examining impaired mobility and BMI and their relationship to injurious falls, further study of these variables in hospitalized adults is recommended. Samples showing greater variability in BMI should be studied. Many studies in hospital settings, including this one, rely on retrospective analysis of existing data. Prospective studies examining more diverse variables might yield new information and could reveal additional predictors of injurious falls. Further research should also include patients who experience multiple falls. A better understanding of the characteristics of patients who suffer injurious falls is essential to improving prediction of which patients are at higher risk and to developing interventions that minimize the effects of falls.
2. Wu S, et al. A cost-effectiveness analysis of a proposed national falls prevention program Clin Geriatr Med 2010 26 4 751–66
3. American Nurses Credentialing Center. 2014 Magnet application manual. Silver Spring, MD: American Nurses Association 2013.
4. American Nurses Credentialing Center. Addendum: changes to the 2014 Magnet application manual. Silver Spring, MD: American Nurses Association 2014.
5. Department of Health and Human Services, Centers for Medicare and Medicaid Services. 42 CFR Parts 411, 412, 413, and 489. Medicare program; proposed changes to the hospital inpatient prospective payment systems and fiscal year 2008 rates. Washington, DC: Federal Register 2007 24679-5135.
6. Tinetti ME Factors associated with serious injury during falls by ambulatory nursing home residents J Am Geriatr Soc 1987 35 7 644–8
7. Hester AL, Davis DM Validation of the Hester Davis Scale for fall risk assessment in a neurosciences population J Neurosci Nurs 2013 45 5 298–305
8. Nelson R Englebardt SP, Nelson R Major theories supporting health care informatics editors. Health care informatics: an interdisciplinary approach 2002 St. Louis Mosby 3–27
9. American Nurses Association. Nursing informatics: scope and standards of practice
. 2nd ed. Silver Spring, MD; 2015.
10. Bond AJ, et al. The risk of hemorrhagic complications in hospital in-patients who fall while receiving antithrombotic therapy Thromb J 2005 3 1 1
11. Brand CA, Sundararajan V A 10-year cohort study of the burden and risk of in-hospital falls and fractures using routinely collected hospital data Qual Saf Health Care 2010 19 6 e51
12. Fischer ID, et al. Patterns and predictors of inpatient falls and fall-related injuries in a large academic hospital Infect Control Hosp Epidemiol 2005 26 10 822–7
13. Krauss MJ, et al. Circumstances of patient falls and injuries in 9 hospitals in a midwestern healthcare system Infect Control Hosp Epidemiol 2007 28 5 544–50
14. Nettleman MD, et al. Epidemiology of bone fractures occurring during hospitalization Clin Perform Qual Health Care 1993 1 2 81–3
15. Pierce JR Jr., et al. Narcotic administration and fall-related injury in the hospital: implications for patient safety programs and providers Int J Risk Saf Med 2013 25 4 229–34
16. Tzeng HM Inpatient falls in adult acute care settings: influence of patients’ mental status J Adv Nurs 2010 66 8 1741–6
17. Greenspan SL, et al. Fall direction, bone mineral density, and function: risk factors for hip fracture in frail nursing home elderly Am J Med 1998 104 6 539–45
18. Grisso JA, et al. Risk factors for falls as a cause of hip fracture in women. The Northeast Hip Fracture Study Group N Engl J Med 1991 324 19 1326–31
19. Spector W, et al. Risk factors associated with the occurrence of fractures in U.S. nursing homes: resident and facility characteristics and prescription medications J Am Geriatr Soc 2007 55 3 327–33
20. Thapa PB, et al. Injurious falls in nonambulatory nursing home residents: a comparative study of circumstances, incidence, and risk factors J Am Geriatr Soc 1996 44 3 273–8
21. Tinetti ME, et al. Risk factors for serious injury during falls by older persons in the community J Am Geriatr Soc 1995 43 11 1214–21
22. Vassallo M, et al. The relationship of falls to injury among hospital in-patients Int J Clin Pract 2005 59 1 17–20
23. Colon-Emeric CS, et al. Can historical and functional risk factors be used to predict fractures in community-dwelling older adults? Development and validation of a clinical tool Osteoporos Int 2002 13 12 955–61
24. Coutinho ES, et al. Risk factors for falls with severe fracture in elderly people living in a middle-income country: a case control study BMC Geriatr 2008 8 21
25. Tinetti ME, et al. The contribution of predisposing and situational risk factors to serious fall injuries J Am Geriatr Soc 1995 43 11 1207–13
27. American Hospital Association. Symptoms, signs, and ill-defined conditions. In: International Classification of Diseases, Ninth Revision, Clinical Modification
. Oak Terrace, IL; 2008.
28. Buck CJ 2013 ICD-9-CM for hospitals, volumes 1, 2, and 3
. St. Louis: Elsevier; 2013.
29. Formiga F, et al. Characteristics of falls producing hip fractures in nonagenarians J Nutr Health Aging 2008 12 9 664–7
30. Myers AH, et al. Risk factors associated with falls and injuries among elderly institutionalized persons Am J Epidemiol 1991 133 11 1179–90
31. Stevens JA, et al. The costs of fatal and non-fatal falls among older adults Inj Prev 2006 12 5 290–5
32. Yet B, et al. Not just data: a method for improving prediction with knowledge J Biomed Inform 2014 48 28–37
For eight additional continuing nursing education activities on the topic of falls, and more than 100 activities on research, go to www.nursingcenter.com/ce.
Keywords:Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
falls; fall-related injury; hospital-acquired condition; injurious falls; inpatient