Alcohol-drinking history and fatal injury in older adults

Abstract

Although most clinical guidelines for older adults allow for one drink a day in persons without a history of alcoholism, diabetes, or cardiovascular disease, alcohol may contribute to fatal injury in the elderly. Using two national surveys, this case–control study determined the associations between drinking history and fatal injuries from falls, motor vehicle crashes and suicides. We performed a case–control study using 1,735 cases who died of falls, motor vehicle crashes, or suicides selected from the 1993 National Mortality Follow-Back Survey; controls (n = 13,381) were a representative sample of the U.S. population from the 1992 National Longitudinal Alcohol Epidemiologic Survey. Cases and controls were restricted to ages 55 years and older. Having 12 or more drinks in the year before death or interview for the controls was used to assess alcohol-drinking history. The unadjusted relative odds for drinkers versus nondrinkers for falls, motor vehicle crashes, and suicides were 1.7, 1.7, and 1.6, respectively. Adjustment for age, gender, marital status, education, and working in the last year did not change these effect estimates, which all excluded the null value. Drinking increased the risk of suicide more for women than for men. Drinking history in older adults is associated about equally with an increased risk of fatal injury from falls, motor vehicle crashes, and suicides.

Introduction

Older adults are particularly susceptible to the ill effects of alcohol consumption. Age-related physiological changes, such as decreased gastric alcohol dehydrogenase and reduced total body water content, result in higher blood alcohol levels per amount consumed in older adults compared to younger adults (Vestal et al., 1977). Older adults also have increased central nervous system sensitivity and decreased tolerance to alcohol (Blow, 1977). At the same peak level of blood alcohol concentration (BAC), the postural balance of older adults is twice as impaired as younger adults (Vogel-Spratt & Barrett, 1984). Over time, severe alcoholism can cause degeneration of the cerebellum and can produce marked disturbances in gait (Gilman et al., 1998). Comorbidity and subsequent medication use increase with age, thus increasing the likelihood of drug interactions with alcohol in late life. Although most clinical guidelines for older adults allow for one drink a day in persons without a history of alcoholism, diabetes, or cardiovascular disease, alcohol has both short-term and chronic effects that can contribute to fatal injury by altering the risk of the initial injurious event and the risk of subsequent fatal sequelae (Moore, 2003).

Our earlier work found that drinking history was associated with a significantly increased risk of all types of fatal injury (Chen et al., 2005); however, the relationship between alcohol use and fatal injury among older persons has yet to be explored. Falls, motor vehicle crashes, and suicide were the three leading causes of injury death among persons aged 65 years and older, comprising 64% of the 40,321 injury deaths in 2002 (National Center for Injury Prevention and Control, 2005). Using two national surveys, the aim of this case–control study was to determine the associations between drinking history and fatal injuries in the elderly, mainly from falls, motor vehicle crashes, and suicides.

Alcohol consumption is associated with falls in the total population (Hingson & Howland, 1993), but its role in the elderly is less clear. For instance, there are seven studies of nonfatal falls in the elderly that show either no statistical association or an inverse association between regular alcohol consumption and fall risk (Iliffe et al., 1991, Lord and Ward, 1993, Mukamal et al., 2004, Nelson et al., 1992, O'Loughlin and Robitaille, 1993, Sheahan and Coons, 1995, Tinetti et al., 1988). Among fatal falls, 59% occurred in persons aged 75 years or older, who comprise only 5% of the population (Baker et al., 1992). Potential mechanisms of death include fatal cerebral injuries and hip fractures, which are both associated with chronic alcohol use (Felson et al., 1988, Kirkpatrick and Pearson, 1978).

Studies of motor vehicle deaths in the elderly have shown weak associations with alcohol use. One study reported that 4% of 71 fatally injured drivers had a detectable BAC of <0.05 g/100 ml, which was below the legal limit in Sweden (Sjogren et al., 1993). Another study reported that one of 110 killed or injured elderly drivers had a BAC over 0.08 g/l00 ml (Holubowycz et al., 1994). Risk of a crash for elderly drivers has been associated with diabetes (odds ratio [OR] = 2.6, 95% confidence interval [CI] = 1.4–4.7), but only marginally with alcohol abuse (OR = 2.1, 95% CI = 0.8–6.0) (Koepsell et al., 1994). Of 4,073 older adult cases in a state-wide trauma registry over a 6-year period 1995–2001, 8.5% tested positive for blood alcohol: 5% of persons injured in motor vehicle crashes and 13% of fall victims (Selway, 2004).

Data from the National Mortality Follow-Back Survey indicated that moderate or heavy alcohol use was associated with suicide in the elderly (Grabbe et al., 1997). Compared with deaths from noninjury causes, moderate (one to two drinks/day) and heavy alcohol use (three or more drinks/day) increased the risk of suicide by about nine times. An autopsy study of alcohol use and suicide found alcohol present in 15% of adults aged 64–73 years, and in 6% of adults aged 74 years and older (Conwell et al., 1990). Adults aged 65 years or older who have more than seven drinks per week may be at risk for alcohol-related problems, including suicide (Blow et al., 2004).

Despite the different mechanisms of injury, a drinking history may represent a point at which a clinician can intervene. For this reason, we evaluated the relationship between drinking history and the causes of two-thirds of all fatal injury in older persons.