0
Get Alert
Please Wait... Processing your request... Please Wait.
You must sign in to sign-up for alerts.

Please confirm that your email address is correct, so you can successfully receive this alert.

1
Brief Report   |    
Using Pharmacy Data on Partial Adherence to Inform Clinical Care of Patients With Serious Mental Illness
Emily M. Woltmann, M.S.W.; Marcia Valenstein, M.D., M.S.; Deborah E. Welsh, M.S.; Todd A. Lee, Pharm.D., Ph.D.; Patricia A. Wolschon, M.S.W.; John Grabowski, M.D.; Peter A. Reilly, Pharm.D.
Psychiatric Services 2007; doi: 10.1176/appi.ps.58.6.864

Objective: Low adherence to antipsychotic medications is a risk factor for poor outcomes for people with serious mental illness. Pharmacy data might be used by health systems to identify partially adherent patients for interventions. This study assessed whether using pharmacy data is an accurate screening method for identifying at-risk patients. Methods: Administrative data were used to identify 1,712 veterans as having schizophrenia or a schizoaffective or bipolar disorder and who had 12-month antipsychotic medication possession ratios (MPRs) of less than .80. Patients' charts were reviewed for alternative explanations for low rates of filling prescriptions for antipsychotic medication. Results: Of 1,712 patients whose pharmacy data indicated partial adherence (MPRs less than .80), 17% (N=297) may have been adherent. Patients with bipolar disorder had higher odds of receiving a false-positive designation (adjusted odds ratio of 1.8, 95% confidence interval of 1.31—2.39). Conclusions: MPRs constructed from pharmacy data can be a useful first screen for identifying patients who need assistance with medication adherence. (Psychiatric Services 58:864—867, 2007)

Abstract Teaser
Figures in this Article

Ms. Woltmann is affiliated with the Dartmouth Psychiatric Research Center, Lebanon, New Hampshire, and the Center for Evaluative Clinical Sciences, Dartmouth Medical School, Dartmouth College, Hanover. Dr. Valenstein and Ms. Welsh are with the Serious Mental Illness Treatment, Research, and Evaluation Center, Department of Veterans Affairs, Ann Arbor, Michigan, and Dr. Valenstein is also with the Department of Psychiatry, University of Michigan. Dr. Lee is with the Midwest Center for Health Services and Policy Research, Hines Veterans Affairs Hospital, Hines, Illinois, and with the Institute for Healthcare Studies and Division of General Internal Medicine, Northwestern University Feinberg School of Medicine, Chicago. Ms. Wolschon, Dr. Grabowski, and Dr. Reilly are with the John D. Dingell Medical Center, Department of Veterans Affairs, Detroit, Michigan. An earlier version of this article was presented in the poster sessions of the annual research meeting of Academy Health, June 6—8, 2004, and the annual meeting of the American Psychiatric Association, May 1—6, 2004. Send correspondence to Ms. Woltmann, Center for Evaluative Clinical Sciences, Dartmouth Medical School, Dartmouth College, 7252 MML Building, Hanover, NH 03755 (e-mail: emily.woltmann@dartmouth.edu).

Antipsychotic medications are a first-line treatment for people with schizophrenia and schizoaffective disorder and may be a useful treatment for some patients with bipolar disorder (1,2). Partial adherence to antipsychotic medications is a significant risk factor for rehospitalization, and an estimated 50% to 75% of patients with schizophrenia will relapse within a year after cessation of antipsychotic medications (3). As many as 40% of patients are at least partially nonadherent with their antipsychotic medications (4). Despite this well-documented challenge, clinicians continue to have difficulty identifying patients who may need help with adherence (5).

Pharmacy data have been used to quantify adherence to a wide range of medications for research purposes. Several studies of either Department of Veterans Affairs (VA) or Medicaid patient populations have demonstrated that poor antipsychotic medication adherence as identified through pharmacy data is related to important clinical and health system outcomes, such as inpatient hospital utilization and costs (6,7,8,9,10). Some researchers have suggested that the increasing availability of pharmacy refill data may allow health care systems to identify patients for targeted interventions to improve adherence (10). Although pharmacy data may be a valid and reliable means to estimate average adherence rates in populations of patients, if a health care system were to use these data to inform clinical care for individuals, the measure would need reasonable predictive value.

In this study, we examined the potential clinical use of one measure of medication adherence constructed from pharmacy data and commonly used in research, the medication possession ratio (MPR). We chose this measure on the basis of our experience in using it to identify patients with severe mental illness for an intervention to improve medication adherence. We also investigated by chart review whether patients with a bipolar diagnosis had a different false-positive rate of nonadherence than those with schizophrenia.

This retrospective study was conducted with administrative and clinical chart data from four VA hospitals and their associated community-based outpatient clinics. The local institutional review board at each site approved a clinical trial of which this report is a component. The review boards approved a waiver of consent for access to personal health information for screening purposes for the study. Screening data for patients who did not enroll in the clinical trial were masked from the records.

Demographic, 12-month pharmacy, and inpatient hospitalization data were obtained from facility-level administrative data for all patients who received a diagnosis of schizophrenia, schizoaffective disorder, or bipolar disorder between August 2002 and September 2004. Patients had to have filled at least two prescriptions for antipsychotic medication within a 12-month period at one of the study sites to be included in the study. Data were extracted from each VA facility's Veterans Health Information Systems and Technology Architecture, the information infrastructure that supports the administrative operations' electronic medical record in the VA, and physician medication order entry. Recent reports have indicated that VA administrative data have high levels of concordance between administrative and medical record data (11). Diagnoses of schizophrenia recorded in VA inpatient administrative data and in Medicaid claims data closely reflect clinical diagnoses of schizophrenia (12). Further, VA mental health services users have been shown to receive low proportions of their mental health care outside the VA system (13).

The MPR was calculated as the days of medication supply that patients received in the one-year period divided by the number of days that patients were outpatients during the year. In the event of concurrent prescriptions for multiple antipsychotic medications, the weighted average of the MPRs was computed. Patients were given credit for "days adherent" for days spent in inpatient hospitalization. Patients who received less than 80% of antipsychotic medication were considered to have partial adherence. Although published reports do not yet suggest a specific threshold at which partial adherence with antipsychotics becomes problematic, taking 80% or more of one's prescribed medications has often been used as a traditional cutoff point for "good adherence" (9,14).

A list of patients with MPRs less than .80 was distributed to research associates at each of the four sites. Medical records for each patient for the prior 12 months were abstracted to determine whether there were explanations for patients' low MPRs other than partial adherence. After an initial iterative coding process, four possible alternative explanations were considered: transferring mental health services to a different VA facility or non-VA treatment setting, receiving depot antipsychotic medications during at least part of the year, moving to a different place of residence or spending significant portions of the year out of state, or having antipsychotic medications discontinued or decreased on the basis of clinical reasons documented in the medical record but not having orders changed in the prescription order.

Frequencies were computed dichotomously for false positives (yes or no), meaning the low MPR reflected nonadherence but the chart indicated at least one alternative explanation for the low rating. Frequencies of false positives were then broken down according to the categories previously described. Multivariate logistic regression analysis was used to evaluate the probability of obtaining a false-positive designation by diagnosis, with controls for demographic, site, and clinical characteristics. Statistical analyses were completed with the SAS System for Windows, release 9.1.3.

A total of 4,394 patients were identified as having a diagnosis for inclusion in the study, and 39% of these patients (N=1,712) had an MPR of less than .80. Among these patients with a low MPR, 1,577 (92%) were male and the mean±SD patient age was 52±11.31 years. Of the veterans in the low MPR sample, 997 (58%) were white, 573 (33%) were African American, 44 (3%) were Hispanic, and 98 (approximately 6%) were of unknown race or ethnicity. Approximately 64% (or 1,099) had a diagnosis of schizophrenia or schizoaffective disorder, and 613 (36%) had a diagnosis of bipolar disorder. Thirteen percent of patients (N=218) had some indication of problematic substance use on chart review. The MPR for the 1,712 patients with low MPRs was .40±.21. Upon chart review, we found that 297 patients (about 17%) with MPRs less than .80 had explanations for low MPRs other than or in addition to partial adherence. These patients' MPRs may have represented false positives for partial adherence.

Table 1 presents the percentages of patients who had one of four possible explanations for low MPRs other than partial adherence, resulting in false-positive screens when looking at MPRs. Forty-seven percent (N=141) of the patients with false-positive results appeared to have low MPRs because their physicians intentionally discontinued or decreased their antipsychotic medication sometime in the year but did not change the pharmacy order. In most cases, the antipsychotic medication was discontinued, but if it was not, the MPR was recalculated on the basis of the chart notes to verify that the MPR was above .80. Clinical notes indicated that 58 (20%) of patients with false positives transferred to another VA facility or outside of the VA health care system during the year, and another 59 (20%) were on depot medications, resulting in an artificially low oral antipsychotic MPR. Finally, 39 (13%) of these patients appeared to have moved or to have been spending part of the year out of state during the calculation period.

Because continuous antipsychotic therapy is strongly recommended for patients with schizophrenia but many patients with bipolar disorder may receive antipsychotic medications on a more intermittent basis, we examined the odds of receiving false positives for poor antipsychotic adherence by diagnosis, controlling for potentially confounding covariates of race, age, site location, and clinical characteristics using multivariate logistic regression. After the analyses controlled for covariates, the odds of having a false-positive screen were 80% higher for patients with bipolar disorder than for patients with schizophrenia or schizoaffective disorder (odds ratio [OR]=1.8, 95% confidence interval [CI]=1.3—2.4). This result was virtually identical to the unadjusted OR for false positives among people with bipolar disorder compared with those with schizophrenia (OR=1.8, CI=1.3—2.7).

This was the first study to examine the feasibility of using the MPR as a screening tool to identify patients in health care organizations with partial adherence for systematic interventions to improve adherence. Our data suggest that a low MPR is a reasonable first-screen indicator of partial adherence but that only a relatively small percentage of patients will have clear alternative explanations in their medical charts for inconsistent filling of prescriptions. Our study data indicated that antipsychotic MPRs may be a less useful screen for patients with bipolar disorder than for patients with schizophrenia. For a health system, the MPR might be used to efficiently identify people who may benefit from additional support with adherence. The MPR could also be used as input for decision support tools for shared decision making between clinician and patient regarding medications.

This study has several limitations. The study used VA pharmacy and clinical data, which were captured in a mandated electronic record and were of high quality. Other organizations, however, may have less complete or less reliable data. Findings may not generalize to other patient populations. The VA patient population is generally older, is more predominantly male, and has lower socioeconomic status than the larger U.S. population (15).

Perhaps the most important caution regarding inferences made from these data is that this study cannot provide a rigorous, formal evaluation of the MPR as a screening tool for clinically significant partial adherence to antipsychotic medication. Although we believe that ascertaining the proportion of patients that most obviously had false-positive ratings on chart review is an important step in evaluating the clinical usefulness of a measure of adherence that could be systematically and inexpensively used to improve clinical care, our chart abstraction method is conditional on patients' having a positive screen for partial adherence. As such, our study does not provide detailed information about the test operating characteristics of the MPR. Future research might include efforts to prospectively ascertain the proportion of poorly adherent patients identified by alternative methods and to then compare these methods with the MPRs derived from pharmacy data.

Several researchers have suggested that the MPR could be used to inform clinical practice, not only for research and quality improvement but also during clinical encounters. This study takes a first step in establishing the MPR as a clinically useful way to estimate adherence among individual patients by providing the proportion of patients with MPRs less than .80 who are likely not filling their prescriptions more than 80% of the time. Providing systematic, timely information regarding usage of antipsychotic medication may facilitate better patient-physician communication regarding medication use and may allow patients to benefit from targeted, evidence-based interventions to address adherence challenges.

This research was supported by grants ARCD-98-350 and IIR-01-174-01 from the Health Services Research and Development Service of the Department of Veterans Affairs. The authors acknowledge the generous help and active participation of the mental health staff, pharmacy staff, and enrolled patients in the Veterans Affairs facilities in this study.

The authors report no competing interests.

Lehman AF, Kreyenbuhl J, Buchanan RW, et al: The Schizophrenia Patient Outcomes Research Team (PORT): updated treatment recommendations 2003. Schizophrenia Bulletin 30:193—217, 2004
 
American Psychiatric Association: Practice Guideline for the Treatment of Patients With Bipolar Disorder. Arlington, Va, American Psychiatric Association, 2002
 
Viguera AC, Baldessarini RJ, Hegarty JD, et al: Clinical risk following abrupt and gradual withdrawal of maintenance neuroleptic treatment. Archives of General Psychiatry 54:49—55, 1997
 
Lacro JP, Dunn LB, Dolder CR, et al: Prevalence of and risk factors for medication nonadherence in patients with schizophrenia: a comprehensive review of recent literature. Journal of Clinical Psychiatry 63:892—909, 2002
 
Valenstein M, Barry KL, Blow FC, et al: Agreement between seriously mentally ill veterans and their clinicians about medication compliance. Psychiatric Services 49:1043—1048, 1998
 
Weiden PJ, Kozma C, Grogg A, et al: Partial compliance and risk of rehospitalization among California Medicaid patients with schizophrenia. Psychiatric Services 55:886—891, 2004
 
Eaddy M, Grogg A, Locklear J: Assessment of compliance with antipsychotic treatment and resource utilization in a Medicaid population. Clinical Therapeutics 27:263—272, 2005
 
Svarstad BL, Shireman TI, Sweeney JK: Using drug claims data to assess the relationship of medication adherence with hospitalization and costs. Psychiatric Services 52:805—811, 2001
 
Gilmer TP, Dolder CR, Lacro JP, et al: Adherence to treatment with antipsychotic medication and health care costs among Medicaid beneficiaries with schizophrenia. American Journal of Psychiatry 161:692—699, 2004
 
Valenstein M, Copeland LA, Blow FC, et al: Pharmacy data identify poorly adherent patients with schizophrenia at increased risk for admission. Medical Care 40:630—639, 2002
 
Szeto HC, Coleman RK, Gholami P, et al: Accuracy of computerized outpatient diagnoses in a Veterans Affairs general medicine clinic. American Journal of Managed Care 8:37—43, 2002
 
Lurie N, Popkin M, Dysken M, et al: Accuracy of diagnoses of schizophrenia in Medicaid claims. Hospital and Community Psychiatry 43:69—71, 1992
 
Desai RA, Rosenheck RA, Rothbard A: Cross-system service use among VA mental health patients living in Philadelphia. Admissions Policy and Mental Health 28:299—309, 2001
 
Osterberg L, Blaschke T: Adherence to medication. New England Journal of Medicine 353:487—497, 2005
 
Agha Z, Lofgren RP, VanRuiswyk JV, et al: Are patients at Veterans Affairs Medical Centers sicker? A comparative analysis of health status and medical resource use. Archives of Internal Medicine 160:3252—3257, 2000
 
Table 1  Proportion of patients with low medication possession ratios who may have been adherent to antipsychotic medication
Table 1  Proportion of patients with low medication possession ratios who may have been adherent to antipsychotic medication
+

References

Lehman AF, Kreyenbuhl J, Buchanan RW, et al: The Schizophrenia Patient Outcomes Research Team (PORT): updated treatment recommendations 2003. Schizophrenia Bulletin 30:193—217, 2004
 
American Psychiatric Association: Practice Guideline for the Treatment of Patients With Bipolar Disorder. Arlington, Va, American Psychiatric Association, 2002
 
Viguera AC, Baldessarini RJ, Hegarty JD, et al: Clinical risk following abrupt and gradual withdrawal of maintenance neuroleptic treatment. Archives of General Psychiatry 54:49—55, 1997
 
Lacro JP, Dunn LB, Dolder CR, et al: Prevalence of and risk factors for medication nonadherence in patients with schizophrenia: a comprehensive review of recent literature. Journal of Clinical Psychiatry 63:892—909, 2002
 
Valenstein M, Barry KL, Blow FC, et al: Agreement between seriously mentally ill veterans and their clinicians about medication compliance. Psychiatric Services 49:1043—1048, 1998
 
Weiden PJ, Kozma C, Grogg A, et al: Partial compliance and risk of rehospitalization among California Medicaid patients with schizophrenia. Psychiatric Services 55:886—891, 2004
 
Eaddy M, Grogg A, Locklear J: Assessment of compliance with antipsychotic treatment and resource utilization in a Medicaid population. Clinical Therapeutics 27:263—272, 2005
 
Svarstad BL, Shireman TI, Sweeney JK: Using drug claims data to assess the relationship of medication adherence with hospitalization and costs. Psychiatric Services 52:805—811, 2001
 
Gilmer TP, Dolder CR, Lacro JP, et al: Adherence to treatment with antipsychotic medication and health care costs among Medicaid beneficiaries with schizophrenia. American Journal of Psychiatry 161:692—699, 2004
 
Valenstein M, Copeland LA, Blow FC, et al: Pharmacy data identify poorly adherent patients with schizophrenia at increased risk for admission. Medical Care 40:630—639, 2002
 
Szeto HC, Coleman RK, Gholami P, et al: Accuracy of computerized outpatient diagnoses in a Veterans Affairs general medicine clinic. American Journal of Managed Care 8:37—43, 2002
 
Lurie N, Popkin M, Dysken M, et al: Accuracy of diagnoses of schizophrenia in Medicaid claims. Hospital and Community Psychiatry 43:69—71, 1992
 
Desai RA, Rosenheck RA, Rothbard A: Cross-system service use among VA mental health patients living in Philadelphia. Admissions Policy and Mental Health 28:299—309, 2001
 
Osterberg L, Blaschke T: Adherence to medication. New England Journal of Medicine 353:487—497, 2005
 
Agha Z, Lofgren RP, VanRuiswyk JV, et al: Are patients at Veterans Affairs Medical Centers sicker? A comparative analysis of health status and medical resource use. Archives of Internal Medicine 160:3252—3257, 2000
 
+
+

CME Activity

There is currently no quiz available for this resource. Please click here to go to the CME page to find another.
Submit a Comments
Please read the other comments before you post yours. Contributors must reveal any conflict of interest.
Comments are moderated and will appear on the site at the discertion of APA editorial staff.

* = Required Field
(if multiple authors, separate names by comma)
Example: John Doe



Web of Science® Times Cited: 15

Related Content
Articles
Books
Manual of Clinical Psychopharmacology, 7th Edition > Chapter 1.  >
The American Psychiatric Publishing Textbook of Psychopharmacology, 4th Edition > Chapter 62.  >
The American Psychiatric Publishing Textbook of Psychopharmacology, 4th Edition > Chapter 62.  >
The American Psychiatric Publishing Textbook of Substance Abuse Treatment, 4th Edition > Chapter 38.  >
The American Psychiatric Publishing Textbook of Psychopharmacology, 4th Edition > Chapter 29.  >
Topic Collections
Psychiatric News
Read more at Psychiatric News >>
APA Guidelines