An explanation for the recent increase in the fall death rate among older Americans: a subgroup analysis

Objective

We sought to explain the recent increase in the death rate from falls among Americans aged 65 years and older.

Methods

Using the CDC WONDER online database, a longitudinal analysis of subgroups of fall mortality based on the International Classification of Diseases, 10th Revision (ICD-10) was conducted in older adults and in younger people. We used linear regression to examine the statistical significance of trends in mortality rates during 1999–2007.

Results

The overall mortality rate from falls increased by 55% among older Americans (≥65 years of age) during 1999–2007, from 29 per 100,000 population to 45 per 100,000 population. For those aged ≥65 years, the largest increase by far (698%) occurred in the subgroup “other falls on the same level,” followed by a moderate increase in falls involving wheelchairs or furniture (48%). The steepest increases at all ages occurred from 1999 to 2000, after ICD-10 took effect. State-level analysis confirmed the findings for the entire United States. From 1999 to 2007, total mortality from falls decreased by 5% in people younger than 45 years of age and increased by 44% for those aged 45–64 years; mortality from “other falls on the same level” increased by 202% and 431%, respectively, in these age groups.

Conclusions

Because the reported minor increases in emergency department and hospitalization rates for falls were insignificant, the almost sevenfold increase in death rates from “other falls on the same level” strongly suggests an effect of improved reporting quality.Falls among older adults are a serious public health problem. Each year, one in three Americans aged 65 years and older falls.^1,2 The annual direct and indirect cost of fall injuries is expected to reach $55 billion (in 2007 U.S. dollars) by 2020.³ A substantial increase in the death rate from unintentional falls has recently been reported among older Americans.^4–9 This increase is perplexing because neither the nonfatal fall rate of 2001–2007 based on Web-based Injury Statistics Query and Reporting System data for patients treated in emergency departments (EDs), nor the fall hospitalization rate of 1997–2005 based on the National Hospital Discharge Survey increased significantly.^9,10 In older people, the rate of ED visits for falls per 100,000 population rose from 4,631 in 2001 to 4,944 in 2007—a 6% increase—compared with a 55% increase in the death rate. This contrasting evidence has raised a critical question: Does the recent increase in fall-related mortality among older Americans reflect an actual increase in fatal falls?There are several possible explanations for the recent increase in deaths coded with a fall as the underlying cause: (1) an increase in fatal injuries from falls; however, the lack of a corresponding increase in ED cases and in hospitalizations negates this possibility;^9,10 (2) an increase in the case fatality rate, which also appears to be equally unlikely given the general improvement in trauma care and treatment of complications in the United States; and (3) a change in reporting the underlying cause of death,⁹ which could have led to the identification of more deaths as having been due to falls.To provide an explanation of the recent increase in fall mortality, we analyzed changes in mortality during 1999–2007 in the many subgroups of falls. Subgroup analysis provided the opportunity to determine whether the increase was associated with certain types of falls. We studied the fall mortality rates of older Americans, as well as younger age groups, at the national and state level.