Developing indicators of appropriate and inappropriate end-of-life care in people with Alzheimer’s disease,cancer or chronic obstructive pulmonary disease for population-level administrative databases: A RAND/UCLA appropriateness study

Abstract

Background:

A substantial amount of aggressive life-prolonging treatments in the final stages of life has been reported for people with progressive life-shortening conditions. Monitoring appropriate and inappropriate end-of-life care is an important public health challenge and requires validated quality indicators.

Aim:

To develop indicators of appropriate and inappropriate end-of-life care for people with cancer, chronic obstructive pulmonary disease or Alzheimer’s disease, measurable with population-level administrative data.

Design:

modified RAND/UCLA appropriateness method.

Setting/participants:

Potential indicators were identified by literature review and expert interviews and scored in a survey among three panels of experts (one for each disease group). Indicators for which no consensus was reached were taken into group discussions. Indicators with consensus among the experts were retained for the final quality indicator sets.

Results:

The final sets consist of 28 quality indicators for Alzheimer’s disease, 26 quality indicators for cancer and 27 quality indicators for chronic obstructive pulmonary disease. The indicator sets measure aspects of aggressiveness of care, pain and symptom treatment, specialist palliative care, place of care and place of death and coordination and continuity of care.

Conclusion:

We developed a comprehensive set of quality indicators of appropriate and inappropriate end-of-life care in people with Alzheimer’s disease, cancer or chronic obstructive pulmonary disease, to be used in population-level research. Our focus on administrative healthcare databases limits us to treatment and medication, excluding other important quality aspects such as communication, which can be monitored using complementary approaches. Nevertheless, our sets will enable an efficient comparison of healthcare providers, regions and countries in terms of their performance on appropriateness of end-of-life care.