Elsevier

Schizophrenia Research

Volume 63, Issues 1–2, 1 September 2003, Pages 49-58
Schizophrenia Research

Cognitive predictors of medication adherence among middle-aged and older outpatients with schizophrenia

https://doi.org/10.1016/S0920-9964(02)00314-6Get rights and content

Abstract

Objective: Medication nonadherence presents a considerable problem in patients with schizophrenia. There are limited and conflicting data on the association of cognitive impairment with antipsychotic nonadherence. In this study, we evaluated the correlation of patients' scores on Mattis' Dementia Rating Scale (DRS; total and subscale scores) with scores on the Medication Management Ability Assessment (MMAA), a performance-based measure of medication management. Methods: Participants included 110 outpatients with schizophrenia or schizoaffective disorder. Each was evaluated using the MMAA role-play tasks and the DRS. Patients also completed the Drug Attitude Inventory (DAI), and the PANSS (Positive And Negative Syndrome Scale). Results: Age, DAI score, and DRS scores were all correlated with MMAA performance. In a stepwise regression analysis, only DRS scores were predictive of MMAA performance. Among the DRS subscales, conceptualization and memory were the best statistical predictors of MMAA performance. Conclusion: Cognitive functions, especially conceptualization and memory, were the strongest patient-related predictors of his or her ability to manage medications, over and above the effects of age, gender, education level, symptom severity, and attitudes toward medications. These results suggest a need for intervention studies focused on improving, or at least compensating for, specific cognitive deficits such as those in memory and conceptualization among patients with schizophrenia in order to improve their ability to manage medications.

Introduction

Effective pharmacotherapy requires a partnership between physicians and their patients. It can succeed only if both parties contribute to its implementation. This holds especially true for the problem of nonadherence to medications among psychiatric patients. Studies have reported nonadherence rates ranging from 26% (Drake et al., 1989) to as high as 73% (Razali and Yahya, 1995), depending upon patient characteristics and specific measures of adherence used. Multiple investigations have demonstrated that poor medication adherence among patients with schizophrenia is associated with worse health outcomes, such as increased rehospitalization, repeated emergency room visits, worsening of symptoms, and even homelessness Olfson et al., 2000, Weiden and Olfson, 1995, Moore et al., 2000, Marder, 1998. Clearly, nonadherence is a major public health problem, but it can potentially be improved. It becomes critical for clinicians and researchers to evaluate specific factors undermining adherence. By defining such risk factors, we may be able to appropriately target interventions.

Factors involved in nonadherence have been divided into those related to environment, medications, and patient (Fenton et al., 1997). Examples of environmental variables include family support, physician–patient relationship, and access to medications or to health care facilities Owen et al., 1996, Gaebel, 1997, Frank and Gunderson, 1990, Nageotte et al., 1997, while medication-related factors include efficacy, side effects, dosing pattern, and route of administration Fenton et al., 1997, Weiden et al., 1995. Most of the patient-related factors reported in previous studies have included the patients' subjective interpretations of their illness or treatment such as insight into illness, subjective well-being, the perceived benefits (both direct and indirect) of medication, and negative attitudes towards medications Ruscher et al., 1997, Moore et al., 2000, Kemp et al., 1996, Budd et al., 1996, Lysaker et al., 1998. Fenton et al. (1997) characterized these findings as fitting into a “health belief model” of medication adherence, in which behavior is a product of a subjective assessment of the costs and benefits of compliance in relation to personal goals and limitations of everyday life. In their conclusion, however, the authors emphasized that cognitive capacity should be further evaluated as influencing adherence. In schizophrenia, a major barrier to adherence is related to a lack of insight or inability to understand one's disorder and the need for treatment (Lacro et al., in press). Diminished insight, coupled with the cognitive deficits seen in schizophrenia Green, 1996, Green et al., 2000, may decrease patients' ability to adhere to their treatment regimens. People with schizophrenia may have difficulty determining the benefits and subsequent need for treatment as a result of possible difficulties evaluating such issues. In addition, deficits in information processing of complex medication regimens can lead to problems adhering to medication instruction (Corrigan et al., 1990).

Only a few studies have assessed the relationship between cognitive function and adherence. The three such published reports used different measures of cognitive function and other putative predictors of adherence, and obtained divergent results. Buchanan et al. (1992) studied 61 outpatients, most of them on depot neuroleptics, and assessed individuals with the Mini-Mental State Examination (MMSE) (Folstein et al., 1975). Their investigation revealed no correlation between MMSE score and adherence. Cuffel et al. (1996) studied 89 patients, aged 18–55 years, using the Brief Psychiatric Rating Scale (BPRS) (Overall and Gorham, 1962), the Neurobehavioral Cognitive Status Examination (Kiernan et al., 1987), and the Awareness of Illness Interview. Cuffel et al. (1996) found a correlation between patient awareness and adherence. The authors also reported that neurocognitive impairment was associated with a more positive self-report of adherence. Finally, Smith et al. (1999) evaluated 46 middle-aged outpatients with a neurocognitive battery as well as the Scale for Assessment of Positive Symptoms (SAPS) (Andreasen, 1984), the Scale for Assessment of Negative Symptoms (SANS) (Andreasen, 1982), and the BPRS, and reported no association between neurocognitive performance and adherence. The divergent findings among these studies are likely the result of several factors including examination of cognitive functioning with different measures and the use of self-report adherence data. While commonly employed, self-reports often lead to an overestimation of adherence (Vitolins et al., 2000) and may not be reliable in patients with psychotic disorders (Williams, 1994). Overestimations of medication adherence may lead to erroneous conclusions of little or no relationship between cognition and adherence. Self-report has also been found to be a somewhat inaccurate representation of actual adherence and, in fact, is itself influenced to an extent by cognitive status (Cuffel et al., 1996). With respect to the evaluation of cognitive performance, the MMSE is a rather crude measure, while an extensive neuropsychological battery may not be practical for use in nonacademic outpatient settings. Other differences in study design such as patient sample and trial length may have also contributed to differences in results.

In order to address the problem of quantifying adherence in patients with schizophrenia, our group (Patterson et al., 2002) recently developed a performance-based measure of adherence modeled after the Medication Management Test used for HIV-positive patients (Albert et al., 1999). Entitled the Medication Management Ability Assessment (MMAA), it uses role-play to determine a person's capacity to comply with medication regimens. Briefly, participants are orally presented a scripted description of a four-medication regimen. Subjects are provided four pill bottles with printed labels and dried beans of different colors to represent medications. After a 1-h delay, participants are asked to demonstrate how they would take the medication regimen and are able to use the labeled medication bottles. Advantages of the MMAA over other measures of adherence are the direct observation of hypothetical medication management, quantitative results, and ease of test administration. In the first published report with the MMAA, it was found that patients with schizophrenia or schizoaffective disorder took a significantly smaller number of correct pills in a day than did normal comparison subjects. Furthermore, the best predictor of a patient's ability to manage medications was the total score on the MMSE (Patterson et al., 2002). Given the paucity and inconsistency of published data on cognitive function as a predictor of medication adherence, and the limitations of the MMSE, our goal in the present study was to evaluate this association using a more comprehensive measure of evaluating cognitive function.

We chose Mattis' Dementia Rating Scale (DRS) (Mattis, 1973) to quantify cognitive function. We previously reported that the MMSE total score was a significant predictor of patient performance on the MMAA; however, the MMSE is a general screening tool for cognitive dysfunction. We wished to more thoroughly examine the relationship between cognitive dysfunction and medication adherence. It has been suggested that specific domains of cognitive dysfunction in schizophrenia are related to functional outcomes Green, 1996, Green et al., 2000. A more comprehensive measure of cognition (DRS) was employed in the hopes of gaining a better understanding of the specific areas of cognitive dysfunction related to medication nonadherence. The DRS yields scores on subscales for attention, initiation/perseveration, construction, conceptualization, and memory. By further evaluating the correlation between MMAA performance and scores on specific subscales of the DRS, we hoped to gain more detailed information about the specific areas of cognitive impairment related to nonadherence. We also assessed correlations of MMAA performance with scores on the Positive and Negative Syndromes Scale (PANSS) (Kay et al., 1987), and the Drug Attitude Inventory (DAI) (Hogan et al., 1983), a questionnaire designed to obtain patients' subjective views about their medications. The DAI was developed as a measure to predict antipsychotic adherence based on patient attitudes toward their medications. This measure is congruent with the HBM (previously described), which involves subjective interpretation of the benefits and costs of taking medication. The DAI was included in the present study to examine the relationship between attitudes toward treatment and medication management.

We chose to study middle-aged and older adults for several reasons. First, the MMAA was designed to simulate medication regimens commonly encountered by middle-aged and older patients. Secondly, there is a dearth of studies on medication adherence in older persons with psychotic disorders. Thirdly, studies of cognition are important in older people in view of the suggestion from other performance-based measures of activities of daily living in older adults that cognitive performance influences results on such tasks as medication management Fulmer and Gurland, 1997, Diehl et al., 1995. In addition, medication adherence in older persons presents unique issues as older adults are more likely to be taking multiple medications, are generally more susceptible to adverse side effects, may have sensory deficits that impede adherence, and may experience greater difficulty comprehending and self-managing their medications Salzman, 1995, Yamada et al., 2001.

Based on our earlier findings with the MMSE, we hypothesized that cognitive impairment, represented by lower scores on the DRS, would predict poor medication management even after controlling for the previously reported factors of demographic characteristics, positive and negative symptoms, and attitude toward medications (DAI).

Section snippets

Participants

The patient sample was comprised of 110 middle-aged and elderly men and women with a DSM-IV (American Psychiatric Association, 1994) diagnosis of schizophrenia or schizoaffective disorder who were participants in the Intervention Research Center (IRC) for Psychosis in Older Adults at the University of California, San Diego (UCSD). Middle-aged was defined as at least 45 years of age and “elderly” patients were defined as those at least 65 years of age. The exclusion criteria for participants

Correlations with MMAA

MMAA performance was quantified as total correct score, total errors, number of pills under the correct amount, number of correct attempts with food, and number of total attempts. Correlations between the MMAA scores and demographic data, DRS scores, DAI performance, and PANSS subscales are shown in Table 1. The number of pills over the correct amount is not included because only a small number of patients (n=35) took an excess number of pills, and none of the correlations were statistically

Discussion

As hypothesized, cognitive function was a significant predictor of MMAA performance, over and above the contributions of demographic characteristics, severity of psychopathology, and attitudes toward medications. Among the specific subscales of cognitive function, memory impairment and difficulties with conceptualization were most highly correlated with MMAA performance. On the other hand, the MMAA did not correlate with the construction subscale of the DRS and the MMAA total score correlated

Acknowledgments

This work was supported, in part, by the National Institute of Mental Health grants MH49671, MH43693, MH59101, MH19934 and by the Department of Veterans Affairs.

References (54)

  • A. Buchanan

    A two-year prospective study of treatment compliance in patients with schizophrenia

    Psychol. Med.

    (1992)
  • R.J. Budd et al.

    Health beliefs and compliance with antipsychotic medication

    Br. J. Clin. Psychol.

    (1996)
  • L.J. Chapman et al.

    Problems in the measurement of cognitive deficits

    Psychol. Bull.

    (1973)
  • P.W. Corrigan et al.

    From noncompliance to collaboration in the treatment of schizophrenia

    Hosp. Community Psychiatry

    (1990)
  • B.J. Cuffel et al.

    Awareness of illness in schizophrenia and outpatient treatment adherence

    J. Nerv. Ment. Dis.

    (1996)
  • M. Diehl et al.

    Everyday problem solving in older adults: observational assessment and cognitive correlates

    Psychol. Aging

    (1995)
  • C.R. Dolder et al.

    Antipsychotic medication adherence: is there a difference between typical and atypical agents?

    Am. J. Psychiatry

    (2002)
  • R.E. Drake et al.

    Alcohol use and abuse in schizophrenia: a prospective community study

    J. Nerv. Ment. Dis.

    (1989)
  • T.A. Eckman et al.

    Teaching medication management skills to schizophrenic patients

    J. Clin. Psychopharmacol.

    (1990)
  • W.S. Fenton et al.

    Determinants of medication compliance in schizophrenia: empirical and clinical findings

    Schizophr. Bull.

    (1997)
  • A.F. Frank et al.

    The role of the therapeutic alliance in the treatment of schizophrenia: relationship to course and outcome

    Arch. Gen. Psychiatry

    (1990)
  • T. Fulmer et al.

    Evaluating the caregiver's intervention in the elder's task performance: capacity versus actual behavior

    Int. J. Geriatr. Psychiatry

    (1997)
  • W. Gaebel

    Towards the improvement of compliance: the significance of psycho-education and new antipsychotic drugs

    Int. Clin. Psychopharmacol.

    (1997)
  • M.F. Green

    What are the functional consequences of neurocognitive deficits in schizophrenia?

    Am. J. Psychiatry

    (1996)
  • M.F. Green et al.

    Neurocognitive deficits and functional outcome in schizophrenia: are we measuring the “right stuff”?

    Schizophr. Bull.

    (2000)
  • R.K. Heaton et al.

    Clinical neuropsychological findings in schizophrenia and aging

  • R. Heaton et al.

    Neuropsychological deficits in schizophrenia: relationship to age, chronicity and dementia

    Arch. Gen. Psychiatry

    (1994)
  • Cited by (154)

    View all citing articles on Scopus
    View full text