Elsevier

Psychiatry Research

Volume 133, Issues 2–3, 28 February 2005, Pages 129-133
Psychiatry Research

A comparison of electronic monitoring vs. clinician rating of antipsychotic adherence in outpatients with schizophrenia

https://doi.org/10.1016/j.psychres.2004.11.002Get rights and content

Abstract

Antipsychotic non-adherence rates of outpatients with schizophrenia or schizoaffective disorder was assessed by electronic monitoring and clinician rating. Antipsychotic adherence was determined monthly over 3 consecutive months with (1) the Medication Event Monitoring System (MEMS®) cap and (2) the Clinician Rating Scale. Non-adherence was defined as daily adherence of <70% during any one of three monthly evaluations for MEMS® and ratings of ≤4 (scale of 1–7) on the Clinician Rating Scale. Non-adherence was detected in 12 of 25 patients (48%) by MEMS® and 0% by the Clinician Rating Scale. Clinician assessment dramatically underestimated antipsychotic non-adherence.

Introduction

Medication non-adherence may lead to serious consequences including exacerbation of psychotic symptoms (Ayuso-Gutierrez and del Rio Vega, 1997), increased aggression toward self and others (Steadman et al., 1998), worsened prognosis (Lieberman et al., 1998, Wyatt, 1991), increased hospital and emergency room use (Olfson et al., 2000), and high societal costs (Weiden and Olfson, 1995). Antipsychotic medication non-adherence is frequent in persons with schizophrenia, with estimates suggesting rates of approximately 50% over 1 year (Fenton et al., 1997, Young et al., 1986).

Most of what is known about the antipsychotic non-adherence of persons with schizophrenia, however, comes from studies that have employed assessment methods of questionable validity. For example, compared with objective data (e.g., determination of antipsychotic in urine samples), clinician assessment, the most commonly used research tool, has limited ability to detect non-adherence (McClellan and Cowan, 1970, Norell, 1981). Other assessment methods (e.g., self-report, prescription renewal, pill counts, urine/blood levels) are either easily manipulated by false reporting or by taking medications immediately before biological testing, or may be less acceptable to patients due to their intrusive nature (Farmer, 1999).

Because of these limitations, electronic monitoring has become increasingly used to assess adherence in general medical populations. (Farmer, 1999). Electronic monitoring offers increased precision and sophistication compared with earlier methods; however, despite these advantages, important limitations exist. In particular, the documentation produced by electronic monitoring (e.g., date/time of bottle cap opening) does not ensure that patients have ingested the prescribed medication. However, the lack of an electronic device signal (e.g., absence of bottle cap opening) is effective at identifying instances in which patients did not access medication bottles, and therefore likely to establish periods of non-adherence. Thus, electronic devices may underestimate non-adherence, but are unlikely to overestimate non-adherence. A composite approach to assessment may resolve this potential underestimation of non-adherence, as indicated in a recent publication by Liu et al. (2001). This published trial of patients with HIV infection suggests that combining adherence assessment methods (MEMS, pill count, and patient interview) into a composite assessment may better predict clinical response than individual measures alone (Liu et al., 2001).

Limited research currently exists on electronic monitoring of medication adherence. One published trial in schizophrenia found electronic monitoring of adherence feasible: however, an evaluation of its value compared with other adherence methods was not reported (Diaz et al., 2001). Thus, although increasingly sophisticated approaches are being used to assess adherence in clinical research trials, these approaches have been relatively understudied in mental health populations. In this preliminary study, we sought to compare estimates of antipsychotic adherence using electronic monitoring and clinician rating among public sector outpatients with schizophrenia and schizoaffective disorder.

Section snippets

Methods

Approval was obtained from the university Internal Review Board, and written informed consent was obtained from all subjects. After participating in a brief survey study (with eligibility criteria consistent with the current study), patients were approached about participation in the proposed trial. Subjects were usually self-referred from flyers where there was no mention of medication adherence status or the investigational focus of this study. Study subjects were 25 adult outpatients

Results

Of the 25 study participants (18 men and 7 women), 18 met criteria of schizophrenia and 7 met schizoaffective disorder criteria. The mean age of the patients was 39.2 years (±8.6). Twenty-two patients were receiving antipsychotic once daily and three, twice daily. During the course of the study, four patients lost a total of five MEMS® caps.

Table 2 describes the proportion of patients that met various thresholds of MEMS® and Clinician Rating Scale-detected antipsychotic adherence thresholds

Discussion

MEMS® cap monitoring detected clinically meaningful antipsychotic non-adherence (taking of <70% of prescribed antipsychotic during any one of three monthly assessments) in nearly 50% of 25 public sector outpatients with schizophrenia or schizoaffective disorder. The mean antipsychotic adherence during the 3-month study period was 78.2% as assessed by the MEMS®. Using the MEMS® in six monthly assessments, a prior study of 14 outpatients with schizophrenia or schizoaffective found mean monthly

Acknowledgments

Portions of this report were presented at the American Psychiatric Association Annual Meeting, San Francisco, CA, May 19, 2003.

References (21)

There are more references available in the full text version of this article.

Cited by (165)

View all citing articles on Scopus
View full text