OBJECTIVES. Dehydration has been underappreciated as a cause of hospitalization and increased hospital-associated mortality in older people. This study used national data to analyze the burden and outcomes following hospitalizations with dehydration in the elderly. METHODS. Data from 1991 Medicare files were used to calculate rates of hospitalization with dehydration, to examine demographic characteristics and concomitant diagnoses associated with dehydration, and to analyze the contribution of dehydration to mortality. RESULTS. In 1991, 6.7% (731,695) of Medicare hospitalizations had dehydration listed as one of the five reported diagnoses, a rate of 236.2/10,000 elderly Medicare beneficiaries. In 1991, Medicare reimbursed over $446 million for hospitalizations with dehydration as the principal diagnosis. Older people, men, and Blacks had elevated risks for hospitalization with dehydration. Acute infections, such as pneumonia and urinary tract infections, were frequent concomitant diagnoses. About 50% of elderly Medicare beneficiaries hospitalized with dehydration died within a year of admission. CONCLUSIONS. Hospitalization of elderly people with dehydration is a serious and costly medical problem. Attention should be focused on understanding predisposing factors and devising strategies for prevention.

A comparison of hip fracture rates among nine countries (Canada, Chile, Finland, Hong Kong, Scotland, Sweden, Switzerland, the United States and Venezuela) was made using national hospital discharge data for the same time interval. The rates increased by age and were higher for females than males in all nine countries. When based on overall discharge rates, the incidence of hip fracture appeared high in three European countries (Finland, Scotland and Sweden) relative to the other countries. However, when transfer cases were removed and adjustments made for differences in case definition, the risk of hip fracture for both men and women was much similar among the four European and two North American countries, but higher than in Hong Kong. Rates of fracture were lowest in Venezuela and Chile, varying from three to 11 times less than for residents of the other seven countries. Although there are limitations in using hospital discharge data as a measure of incidence, the wide variation in the risk of hip fracture across the nine countries appears real but differences between North American and north European countries may not be as great as previously reported. Such cross-national comparisons may help clarify different etiologic hypotheses.

The results indicate that hip fracture incidence varies as a function of the income level of the ZIP Code area where the population resides. Implications for targeting prevention programs within local areas with low median income are discussed.

No abstract