Predictive validity of a questionnaire that identifies older persons at risk for hospital admission

Abstract

Objective: To determine the predictive validity of a mailed questionnaire designed to measure older adults' risk of repeated hospitalization.

Design: Prospective cohort study.

Setting: Ramsey County, Minnesota.

Participants: Medicaid recipients aged 70 and older who completed the questionnaire.

Measurements: Responses were used to calculate the subjects' probability of repeated admission (Pra) to hospitals within 4 years. Subjects were classified as low-risk (Pra < 0.5) or high risk (Pra > or = 0.5). One year later, Medicaid claims data were analyzed to determine the subjects' actual use of hospitals.

Results: One-fifth of the respondents (20.6%) were classified as high-risk at baseline. During the following year, the high-risk subjects used hospitals at approximately twice the rate of the low-risk subjects (4.5 vs 2.4 days/person-year, P = .009).

Conclusions: The instrument, which was previously found to be valid in a national sample of Medicare beneficiaries, appears to be valid also in a local sample of Medicaid beneficiaries. Older adults at risk of heavy hospital use can be identified prospectively through their responses to this brief, mailed, self-administered questionnaire. The instrument may be useful in targeting older persons for interventions designed to prevent the need for hospital care.