Medication inappropriateness for older adults receiving hospice care: a pilot survey

To test the feasibility and reliability of a tool and methodology for evaluating expert clinicians' perceptions about the application of the Beers criteria in hospice.

A pilot survey.

A national medication therapy management provider specializing in hospice care.

Thirty-five participants from a multidisciplinary panel were invited to complete the survey. They were selected to represent acute, long-term care, and community practice settings with various levels of experience and judgment.

Respondents were asked to complete the survey by rating their agreement or disagreement with the inappropriateness of the medications or medication classes for hospice patients, using a five-point Likert scale from strongly agree (1) to strongly disagree (5), with the midpoint (3) expressing equivocation.

Feasibility as measured by the percentage of returned and completed surveys. A secondary aim was to measure inter-rater reliability and response.

Twenty-four clinicians (69%) completed the survey, including 13 clinical pharmacists, 6 nurses, and 5 physicians. Twenty-nine responses (2%) were furnished by imputation methods. The intraclass correlation for medication inappropriateness for hospice patients was 0.89 (0.81-0.95), indicating "good" inter-rater reliability. Short-acting benzodiazepines, gastrointestinal antispasmodics, anticholinergics, and antihistamines were considered appropriate for use in older hospice patients, but they are considered inappropriate according to the Beers criteria.

We established a viable methodology for evaluating clinician judgment about medication inappropriateness in older hospice patients. Some medications routinely considered to be inappropriate may be appropriate at end of life; different criteria may be needed to determine medication inappropriateness in hospice care.