A Descriptive Analysis of Location of Older Adult Falls That Resulted in Emergency Department Visits in the United States, 2015

Methods

Narratives from the 2015 National Electronic Injury Surveillance System–All Injury Program (NEISS-AIP) were reviewed to determine the locations of nonfatal fall injuries among adults age 65 and older. NEISS-AIP, a nationally representative data system operated by the US Consumer Product Safety Commission, captures data on ED visits that result in an injury. NEISS-AIP includes data from a sample of 66 of 100 NEISS-participating hospitals in US states and territories weighted to represent the US population using the inverse probability of hospital selection in each stratum and adjusted for nonresponse. ¹⁷ For each ED visit, the precipitating cause of injury, principal diagnosis, primary body part injured, and location of the fall were abstracted. Only ED visits in which the precipitating cause, as noted in the ED record, was an unintentional fall were included in the analysis. The location of injury indicator, provided in the NEISS-AIP dataset, included the following locations: home, public, street, industry, school, farm, apartment, sports, or mobile home. Additional information regarding whether these falls occurred indoors or outdoors, or at a specific site (such as bathrooms, yards, kitchens, etc) could only be determined from the narratives.

There were 38 654 ED records of unintentional nonfatal fall injuries among older adults. NEISS coordinators abstract injury-related ED records at participating EDs each day providing information for NEISS variables and a narrative of up to 2 lines of text explaining the circumstances of the injury. ¹⁸ For our analysis narratives were divided evenly and coded by 1 of 4 researchers. A codebook was developed (Appendix A, available online) by reading 100 narratives and then updated for every 2000 narratives coded. Ten percent of all narratives were reviewed by a second researcher, and coding discrepancies were discussed among 4 researchers until consensus was reached. Previous narratives were recoded based on changes made to the codebook.

Using the narrative text we coded location of a fall by 3 separate categories: (1) indoor versus outdoor, (2) home versus not-at-home, and (3) sites in and around the home (specific rooms, stairs, driveway, yard, etc; Appendix A). Additionally, we coded residential status of older adults into community-dwelling, residential facility (eg, nursing home, assisted living, or long-term care) or prison (Appendix A). Those living in prison were excluded due to small sample size (n = 9) leaving 38 645 narratives. Out of the remaining narratives, we were able to determine indoor versus outdoor location for 23 198 narratives (Figure 1). Only the narratives for community-dwelling older adults (n = 34 336) were used to identify if the fall occurred at home or not-at-home. The narratives of 15 328 community-dwelling older adults provided sufficient information on whether the fall occurred at home or not-at-home. To address the low sample size for this variable, we also used the location indicator available in the NEISS-AIP dataset. We coded as home when the NEISS-AIP variable indicated home, apartment, farm, or mobile home. Older adults who fell at their own residence or at the residence of another person were both included in the category of “Home.” Not-at-home indicated public, street, industry, school, or sports location. This increased the sample size to 27 686. Out of the 21 707 falls that occurred at home, 10 791 narratives could be used to determine the specific sites of these falls around the home (bathroom, bedroom, yard, patio, etc; Figure 1).

Results

The narratives from 38 654 unintentional fall-related ED visits in older adults were reviewed, representing over 3 million ED visits in 2015. Older women had more fall-related ED visits (estimated 1 997 090 ED visits or 7.5 ED visits per 100 older women) than older men (estimated 1 040 460 ED visits or 4.9 ED visits per 100 older men). Indoor/outdoor location of falls could be determined for 60.8% of estimated ED visits. The majority of these visits were from falls that occurred indoors (71.6%; Table 1). Indoor falls made up the majority of fall-related ED visits for both community-dwelling older adults (68.2%) and older adults living in residential facilities (98.2%; Table 1). An estimated 28.4% of falls resulting in ED visits could be classified as having occurred outdoors. Indoor versus outdoor location of falls for community-dwelling older adults differed by sex and age group, where location could be determined. For women, 71.6% (95% confidence interval [CI]: 70.7, 72.6) of ED visits for fall injuries occurred indoors and 28.4% (95% CI: 27.4, 29.3) occurred outdoors. For men, 61.7% (95% CI: 60.3, 63.1) of fall injuries occurred indoors and 38.3% (95% CI: 36.9, 39.7) occurred outdoors (Table 1). For community-dwelling adults aged 85 and older, 79.4% (95% CI: 78.1, 80.7) of ED visits for fall injuries occurred indoors. In the 65- to 74-year age group, 58.8% (95% CI: 57.4, 60.2) had an ED visit due to a fall that occurred indoors (Table 1).

The majority of fall-related ED visits among community-dwelling older adults occurred at home (79.2%, 95% CI: 78.6, 79.8; Table 2). When home versus not-at-home location could be determined, falls that occurred at home versus not-at-home did not vary by sex but did vary by age group. For adults 65 to 74 years old, 25.4% (95% CI: 24.3, 26.5) of falls did not occur at home but only 16.2% (95% CI: 15.2, 17.2) of adults 85 and older who went to the ED because of a fall, did not fall at home (Table 2). Among those aged 65 to 74 years, 74.6% (95% CI: 73.5, 75.7) of falls occurred at home, followed by 79.9% (95%CI: 78.9, 80.9) among those aged 75 to 84, and 83.8% (95% CI: 82.8, 84.8) for those 85 and older.

The bedroom (25.0%, 95% CI: 24.0, 26.0), followed by stairs (22.9%, 95% CI: 21.9, 23.9) and bathrooms (22.7%, 95% CI: 21.7, 23.7) were the most common places for falls to have occurred in or around the home (Table 3). Among falls that occurred at home, 5.6% (95% CI: 4.7, 6.6) occurred in the kitchen for men and 8.1% for women (95% CI; 7.3, 8.9). For men, 7.8% of falls occurred in the driveway/garage (95% CI: 6.7, 8.9). The driveway/garage only accounted for 5.0% (95% CI: 4.4, 5.7) of women’s falls at home (Table 3). For adults 85 and older, a high percentage of falls occurred in the bedroom (31.6%, 95% CI: 29.7, 33.6) with only 19.8% (95% CI: 18.2, 21.4) in those aged 65 to 74 years and 24.2% (95% CI: 22.5, 25.9) in those aged 75 to 84 years. For adults in the 65 to 74 age group, a higher percentage of ED visits were due to a fall on the stairs (30.0%, 95% CI: 28.1, 31.8) than any other place at home (Table 3).

Discussion

This study found that the majority of fall-related ED visits among older adults were due to falls that occurred indoors. This was true for both sexes and all age groups. Where location could be determined, indoor falls accounted for 71.6% of women’s fall-related ED visits and 61.7% of men’s. Our findings are similar to a study in the Netherlands, which found that older adults’ fall-related ED visits were more likely due to falls indoors than outdoors and a higher percentage of older men’s fall-related ED visits were from outdoor falls. ⁶ In the MOBILIZE Boston study, Duckham et al explored self-reported falls and fall injuries in a cohort of 765 older adults and found women had a higher rate of fall injuries overall and a higher rate of indoor fall injuries per person year than men. ⁷ However, the rate of outdoor fall injuries per person year did not differ between the sexes. ⁷ A possible explanation for the gender differences is differences in exposure to specific environments. Older men reported spending more time engaged in recreational activities than older women and older women reported spending more time doing light housework compared to older men, which could explain some of the variation in fall location observed between men and women. ⁷

Two studies used data from the National Health Interview Survey to assess location of medically treated fall injuries.^10,11 Both studies found that fall injuries for which older adults sought medical treatment were more likely to occur in their homes compared to public locations. This is similar to our findings which show that the home was the most commonly reported location of fall-related ED visits among community-dwelling older adults, particularly among those 85 and older. This could be due, in part, to older adults spending more time at home than at public places. However, this could not be determined from the study data.

Among fall-related ED visits that occurred at home, most occurred in the bedroom, stairs, or bathroom. This was most evident among adults 85 and older. An Australian study found, among older adults discharged from the hospital in the past 6 months, the most commonly listed location of falls and fall injuries was the bedroom. ¹⁶ This was potentially related to older adults using fewer fall prevention strategies in their bedrooms, such as limited use of assistive devices around their bedrooms. ¹⁶ It is also possible differences in locations around the home are related to the time spent in specific rooms. For example, older adults may spend more time in their bedrooms than other locations. Specifically, the Hill et al study focused on patients recently discharged from the hospital. Older adults may experience functional decline as a result of low mobility while hospitalized, ²⁰ which could result in them spending more time in their bedrooms postdischarge. Our study is not limited to those recently discharged from the hospital, which may explain the high percentages of fall injuries in other places such as bathrooms or on stairs.

There are effective, evidence-based fall prevention strategies that could be improved upon by using these data on fall location. For example, home modifications conducted by a health care professional, such as an occupational therapist, have been shown to reduce the risk of falls in older adults. ⁵ Recommendations from occupational therapists including improving lighting, reducing clutter, painting the edge of steps, installing grab bars and stair rails, and removing or changing throw rugs have been shown to reduce the number of falls and the risk of falls in older adults.^21-23 Knowledge about where injurious falls are most likely to occur in the home and how they differ by both sex and age group may help older adults prioritize the numerous home modifications that might be needed and are costly to implement. For example, a higher percentage of adults aged 65 to 74 fell on the stairs than any other location in or around the home. Making sure stairs have rails and appropriate lighting may be especially important for this age group compared to modifications in other areas of the home.

It is important to note that home hazards are not the only modifiable risk factors associated with falls among older adults. Other modifiable risk factors include poor strength, gait, and balance, use of certain medications, postural hypotension, impaired vision, and improper footwear. ²⁴ The greater the number of fall risk factors older adults have, the higher their risk of falling. ²⁵ Fall prevention programs should include multiple components to address not only home hazards, but other modifiable risk factors. ²⁶ For example, bedrooms were a common location for injurious falls, especially among adults aged 85 and older. Removing tripping hazards in the bedroom may prevent an older adult from falling; however, addressing postural hypotension or reducing medications that increase frequent nighttime urination are also important in reducing these falls. The Center for Disease Control and Prevention’s (CDC) Stopping Elderly Accidents, Deaths, and Injuries (STEADI) initiative guides providers in recommending fall prevention activities based on their patient’s individual modifiable risk factors. STEADI includes recommendations for home safety checks conducted by a health care professional such as an occupational therapist and consultation with a pharmacist to review and make recommendations on how to optimize medication use for improved health and reduced fall risk (www.cdc.gov/steadi).

This study has several limitations. First, approximately 40% of fall-related ED visits could not be attributed to an indoor or outdoor location and approximately 50% of falls that occurred at home could not be linked to a specific location in or around the home. This was mostly due to short narratives only capturing 1 to 2 sentences describing the circumstances of the injury. Results of this study should be interpreted cautiously due to the large percentages of unknown locations. Second, it is possible that the way location variables were coded influenced some of the differences found in this study. It was easier to identify some specific locations around the home than others. For example, it was simple to determine a fall happened in a bedroom if bed was mentioned in the narrative; however, for living room we relied on clues such as couch or recliner, which could be found in other locations but were coded as living room. We did not divide locations of falls that did not occur at home into smaller categories due to the difficulty in determining the specific location of falls not-at-home (eg, the difference between falling on a sidewalk vs falling on the street). Furthermore, residential facilities were only captured if specifically mentioned in the narrative, it is possible that some fall injuries occurred among those living in residential facilities but were coded as community dwelling. Additionally, it is possible that the location of older adults’ fall injuries differs by variables not included in NEISS-AIP data, such as socioeconomic status and comorbidities. Future studies about location of older adult falls may benefit from additional information about demographics. Time spent at different locations could not be determined from the NEISS-AIP dataset. It is possible that a majority of older adults reported falling at home and in their bedroom, stairs, or bathroom because older adults spend more time in these locations. Finally, fall injuries seen in EDs are a subset of injuries and falls overall, information about the location of falls or injurious falls not seen in the ED cannot be determined from this study.

Conclusion

Among older adults who had a fall-related ED visit in 2015, indoor falls were more common than outdoor falls. When the location of the fall could be determined, older adults more commonly fell inside the home and more specifically in a bedroom, on stairs, or in a bathroom. These findings suggest that older adults should be aware of hazards in their homes. If conducted at the population level, home modifications could avert more than 120 000 falls and $442 million in health care expenditures each year. ²⁷ In addition to home modifications, other modifiable fall risk factors such as gait impairment, postural hypotension, and vision impairment should be considered. Health care professionals can screen, assess, and intervene to reduce their older patient’s modifiable fall risk factors and offer solutions to help them remain healthy and independent longer.

Supplemental Material