全球智库信息集成服务系统

The Intervention Did Not Substantially Affect Costs

• No statistically significant effect on costs was detected.
• The 95-percent confidence interval for the intervention's effect on total monthly Medicare costs ranged from a $94-per-month decrease to a $130-per-month increase.
• The 95-percent confidence interval for the intervention's effect on direct injury–related Medicare costs ranged from a $12-per-month decrease to a $48-per-month increase.

Researchers Did Not Detect an Effect on the Frequency of ED Visits for Fall-Related Injuries

• No statistically significant effect on fall-related ED visits was detected.
• The 95-percent confidence interval for the estimated effect of the intervention on the incidence of fall-related injuries leading to ED visits ranged from a 26-percent reduction to a 43-percent increase (i.e., an incidence rate ratio of 0.74 to 1.43). The wide range of this estimate suggests that the study was underpowered for this outcome and that a clinically meaningful effect could have gone undetected.

Researchers Developed and Validated a Model That Can Be Used to Measure the Incidence of Fall-Related ED Visits Without Relying on External Cause-of-Injury Codes

• The model assigns each ED visit a probability weight that represents the likelihood that the visit was precipitated by a fall.
• The weights can be summed to measure the incidence of fall in a given population and time interval. They can also be used to determine the incidence of particular fall-related injuries of interest by summing the weights for ED visits during which particular injuries were diagnosed.

The Office of the Assistant Secretary for Planning and Evaluation (ASPE) in the U.S. Department of Health and Human Services asked the RAND Corporation to evaluate the effectiveness of an intervention designed to prevent falls in the elderly, with a particular focus on fall-related injuries and on health care costs.

Researchers linked data collected during a randomized trial to Medicare enrollment and claims files to compare health care costs and the frequency of fall-related emergency department (ED) visits between treatment and control groups. Using claims from both before and after trial enrollment, they adjusted for baseline differences and used intention-to-treat analyses, thereby overcoming limitations inherent in the outcome data collected during the trial.

The researchers did not find a statistically significant effect of the intervention on costs or on the rate of fall-related ED visits. They estimate that the intervention was associated with an $18-per-month increase (95-percent confidence interval [CI] = –$94 to $130) in total health care spending, an $18-per-month increase (95-percent CI = –$12 to $48) for care directly related to injuries, and a 4-percent increase in the risk of falls (95-percent CI = 26-percent decrease to 43-percent increase).

The researchers concluded that the fall-prevention intervention did not have a substantial effect on health care costs. Although they did not find evidence that the intervention reduced ED visits for fall-related injuries, they cautioned that the study was underpowered for this outcome and that a clinically meaningful effect could have gone undetected.

• Chapter One

  Background

• Chapter Two

  Methods

• Chapter Three

  Populations and Baseline Comparisons

• Chapter Four

  Study Outcomes

• Chapter Five

  Discussion

• Appendix A

  Model for Identifying Emergency Department Visits for Fall-Related Injuries

• Appendix B

  Discussion of Potential Confounders and Remedies

• Appendix C

  Combining Self-Reported Falls with Published Estimates of Cost Savings per Fall Averted