Format

Send to

Choose Destination
Psychiatr Serv. 2007 Sep;58(9):1187-92.

Relationships among subjective and objective measures of adherence to oral antipsychotic medications.

Author information

1
Department of Psychiatry, University of Texas Health Science Center at San Antonio, San Antonio, TX 78229-3900, USA. velligand@uthscsa.edu

Abstract

OBJECTIVE:

The most common ways of assessing adherence to oral antipsychotic medications in research and in clinical practice are self-report and physician report. This prospective study examined the agreement among measures of adherence to oral antipsychotic medications among 52 outpatients with schizophrenia.

METHODS:

Participants were assessed at baseline during a visit to their outpatient clinic and followed for 12 weeks. Adherence was assessed by using subjective measures (self-report and physician report) and objective measures (pill counts conducted in the home, electronic monitoring, and blood plasma concentrations). Electronic monitoring was used as an imperfect standard against which other methods were judged.

RESULTS:

Data from pill counts and from electronic monitoring were strongly correlated (r(k)=.61). Self-report and physicians' ratings of compliance were weakly correlated with pill count and electronic monitoring when compliance scores were examined with rank-order correlations (r(k)=.18-.32). When the sample was dichotomized into adherent and nonadherent groups on the basis of electronic monitoring or pill count (at least 80% adherent), neither physicians nor patients identified adherent behavior (kappa<or=20). Blood plasma concentrations were not correlated with any other measures of adherence (kappa<or=20). Self-report and physician report were best correlated with clinical state (r(k)=-.27, r(k)=-.25), suggesting that patients and treating professionals may use clinical state to estimate adherence.

CONCLUSIONS:

Patients and physicians were not able to identify adherence. The inability of physicians to accurately identify adherent individuals is likely to have important consequences for prescribing behavior, health care costs, and patient outcomes.

PMID:
17766564
DOI:
10.1176/ps.2007.58.9.1187
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Atypon
Loading ...
Support Center