Published studies were considered for inclusion here if they reported relevant epidemiological or economic data on adherence to one of the therapies usually used in palliative care. A search on adherence to cancer palliative care was made using Medline (1990 - 2002). Some reviews and reports from international and national organizations were also included. The search retrieved only studies that evaluated adherence to pain relief in palliative care.

Adherence was usually not explicitly defined in the articles retrieved, but referred to generally as "patients following medical recommendations". In operational terms, the variables of adherence were defined as: "not filling a prescription", "not taking medication", "errors in dosage", "reducing medication", "taking extra medication" and "taking additional nonprescribed medication" (6,7,9,10).

The studies reviewed here used several different methods to estimate the adherence of patients to their medication. These methods, which can be used either separately or in combination, include review of medical records, patient self-report, family report, residual pill counting, electronic measurement devices, prescription refill rates, biological markers in serum or urine, assays to quantify medications or their metabolites and therapeutic outcome (6,9).

Few studies have provided data on the level of adherence of oncology patients to their pain relief, and the methods used to calculate adherence rates were not always described. Zeppetella et al., reported that 40% of patients with cancer adhered to pain relief drugs (9). Miaskowski et al. reported adherence rates for opioid analgesics. Cancer patients prescribed relief on an around-the-clock basis took an average of 88.9%, whereas those who were prescribed relief on an as-needed basis had an adherence rate of about 24.7% (6). Du Pen et al. reported that adherence of oncology patients to their prescribed opioid therapy was between 62% and 72% (7) and Ferrell et al. reported a mean adherence rate of 80% (10).