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
Column   |    
State Mental Health Policy: Implementation of Computerized Medication Prescribing Algorithms in a Community Mental Health System
Karen K. Milner, M.D.; Daniel Healy, M.D.; Kristen L. Barry, Ph.D.; Frederic C. Blow, Ph.D.; Cheryl Irmiter, Ph.D.; Peter De Chavez, M.S.
Psychiatric Services 2009; doi: 10.1176/appi.ps.60.8.1010
View Author and Article Information

The authors are affiliated with the Department of Psychiatry, University of Michigan, Rachel Upjohn Building, 4250 Plymouth Rd., SPC 5740, Ann Arbor, MI 48109 (e-mail: kmilner@med.umich.edu). Dr. Milner and Dr. Healy are also with the Washtenaw Community Health Organization, Ypsilanti, Michigan. Dr. Barry and Dr. Blow are also with the Serious Mental Illness Treatment Research and Evaluation Center, Department of Veterans Affairs, Ann Arbor. Dr. Irmiter is also with the American Medical Association, Chicago. Fred C. Osher, M.D., is editor of this column.

This column describes a Michigan initiative to implement medication prescribing algorithms for schizophrenia, bipolar disorder, and major depression. The algorithms were incorporated into the electronic medical records system of a four-county community mental health system. Guideline adherence of 30 providers who treated nearly 3,000 patients was measured at mid- and endpoints of the first year. They were adherent for about a third of their patients in the first six months (32%) and more than half in the second (52%). Scores on scales measuring providers' perceptions of algorithm ease of use and usefulness were in the midrange at both time points. (Psychiatric Services 60:1010–1012, 2009)

Abstract Teaser
Figures in this Article

The mental health care field is in the midst of a comprehensive transformation, including a broad-based restructuring of care for maximum effectiveness and efficiency. Use of evidence-based practices is critical to this transformation. The appropriate and consistent use of psychiatric medications is a vital concern in the care of individuals with mental illnesses. An evidence-based practice for medication management, the Texas Medication Algorithm Project (TMAP), was developed to give physicians methods and tools for the appropriate pharmaceutical management of bipolar disorder, schizophrenia, and depression among consumers with complex treatment issues (1,2,3,4,5,6).

The translation of best practices such as TMAP into common clinical practice remains a challenge for the mental health field. Meta-analyses have identified barriers and solutions to implementing guidelines and practices (7,8). Organizational factors may facilitate or hinder the implementation, dissemination, and adoption of evidence-based practices (5,6,8,9,10). Organizational culture can influence attitudes toward adoption of innovation in general and evidence-based practices in particular (9). Use of local opinion leaders to champion an innovative evidence-based practice can also be a key to adoption of that practice (7).

This column describes a year-long study that examined provider adherence to and attitudes toward the implementation of algorithms, following TMAP guidelines, for prescribing psychotropic medications to patients diagnosed as having schizophrenia, bipolar disorder, or major depression. The algorithms were incorporated into electronic medical records of a four-county community mental health system in Michigan. The study is one of the first systematic efforts to address implementation of computerized medication algorithms to improve and standardize prescriptive practices for consumers with severe and persistent mental illness. It was approved by the University of Michigan's Institutional Review Board. Introduction of the algorithms were part of the Michigan Mental Health Evidence-Based Practice Initiative.

Before implementation began in January 2007, TMAP guidelines were used to stage all consumers—that is, medical records for each consumer were reviewed by one of the clinician authors to determine which psychotropic medication trials a consumer had undergone. On the basis of this information, the consumer was placed in the appropriate algorithm stage. If a prescriber chose not to follow algorithm recommendations (for example, if the patient required injectable medications), the prescriber documented the reasons in the electronic medical record from a drop-down comprehensive list of reasons (for example, patient tolerating current regimen, patient refused change in medications, and severe side effects). Prescribers could then change the staging—or the algorithm (by changing the diagnosis)—after the initial implementation. Treatment of patients who were psychiatrically stable did not require progression through the algorithm in order to fit TMAP staging recommendations.

In this column "adherence" refers to decision making about prescribing psychotropic medications. The provider was considered to be adherent if he or she followed the medication guidelines as given for each stage and recorded that information in the patient's record or if he or she consulted the guidelines and recorded the reasons for not using them in the electronic medical record or on a paper-and-pencil form.

The medical director and a key staff psychiatrist provided leadership to a group of psychiatrists that met monthly for a year (October 2005 to October 2006) to incorporate the algorithms into the regional electronic medical record. Continuing medical education credits were offered at prescriber training events scheduled biannually to retrain prescribers on the algorithms and update them about implementation.

This column describes an evaluation of provider adherence to the three algorithms—for schizophrenia, bipolar disorder (euphoric-mixed or depressed), or major depression (psychotic or nonpsychotic)—for all patients with these disorders who were seen in two six-month periods during the initial implementation year: April to September 2007, and October 2007 to March 2008. A total of 2,591 patients were seen in the community mental health system during the first period, and 2,615 were seen during the second period. A total of 2,170 individuals were seen in both periods, 359 were seen in the first period only, and 445 were seen in the second period only. This degree of overlap indicates the overall stability of the program and the patient caseload. More than half of the patients in the system had schizophrenia, about a fifth had bipolar disorder, and about a fourth had major depressive disorder.

The variables tracked in the electronic medical records system included the number of consumers being treated according to each algorithm and their demographic and diagnostic characteristics, the number of clinicians using each algorithm, their adherence to the algorithm guidelines, and their reasons for not following recommendations.

The Ease of Use Scale, adapted from a scale used in the TMAP evaluation (9), includes 12 questions about use of the algorithm—for example, How easy were the computerized algorithms to use? How easy was it correct mistakes? How easy were the algorithms to use with your daily workflow? How easy was it to change the stage in the algorithm? Responses were based on a 7-point Likert scale, ranging from 1, very difficult, to 7, very easy. Possible total scores range from 12 to 84, with lower scores indicating more difficulty in using the algorithms. The Usefulness Scale, also adapted from a scales used in the TMAP evaluation (9), consists of 11 questions. Responses are also based on 7-point Likert scales. Response options follow the general format of 1, not at all useful, to 7, very useful. Possible total scores range from 11 to 77, with lower scores indicating less perceived usefulness.

The evaluation included a total of 30 providers. In general, most patients clustered in stages 1–4 in each algorithm. The early stages of the algorithms are grounded in the evidence from randomized controlled trials, whereas later stages are less rigorously derived and influenced by expert consensus. The fact that patients clustered in early stages suggests either that they were responsive to medications in these stages or that they had yet to progress to later stages. In the first six months, providers followed the algorithm guidelines for appropriate medication prescribing for about a third of their patient visits (32%). In the second six months, they followed the guidelines for more than half of their patient visits (52%).

The providers completed the Ease of Use Scale and the Usefulness Scale at booster training events at the end of the first six months (N=20, 66% of the 30 participating providers) and the second six months (N=24, 80%). Because booster training was not mandatory, not all providers attended each session. However, attendance improved by the end of the first full year of implementation. On the Ease of Use Scale the mean±SD score for the first six months was 41.11±13.35 (range 17–60), and the score for the second six months was 40.50±11.75 (range 12–61). Thus providers' mean score was in the midrange in terms of the difficulty of using the electronic medical record with the algorithms. The mean scores indicated that their overall perceptions did not change over the implementation year. Results were similar for the Usefulness Scale. The mean score for the group in the first six months was 33.24± 14.22 (range 9–58), and it was 33.71± 13.56 (range 11–61) in the second six months. Variation in scores on both scales indicated that providers had a wide range of perceptions about ease of use and usefulness.

A scale measuring provider satisfaction was administered at the end of each six-month period. Although there was a relatively wide range of scores on the items, mean scores were in the midrange of possible scores, indicating that prescribers found the embedded guidelines somewhat helpful, although their helpfulness was perhaps influenced by difficulty of use. A final open-ended question that addressed overall satisfaction and asked for suggestions for improvements yielded a range of answers from negative to positive. The written comments were generally more positive at the end of the second six-month period.

The data were analyzed to determine whether provider attitudes varied by years in practice. Although providers with ten or fewer years in practice scored consistently higher (greater ease of use and greater satisfaction), this analysis found no significant differences.

These findings indicate that it was feasible to embed medication algorithms into electronic medical records of a four-county community mental health system. The computerized algorithms included decision points, stages, recommended medications for each stage, reasons for not following the recommendations, and symptom scales. Organization-wide support for implementation was key to the initiative's success. These findings are promising because use of evidence-based practices in the treatment of individuals with mental illnesses is critical to positive long-term outcomes.

The authors thank the Ethel and James Flinn Family Foundation for generous support of this project.

The authors report no competing interests.

Suppes T, Swann AC, Dennehy EB, et al: Texas Medication Algorithm Project: development and feasibility testing of a treatment algorithm for patients with bipolar disorder. Journal of Clinical Psychiatry 62:439–447, 2001
 
Kashner TM, Carmody TJ, Suppes T, et al: Catching up on health outcomes: the Texas Medication Algorithm Project. Health Services Research 38:311–331, 2003
 
Rush AJ, Crismon ML, Kashner TM, et al: Texas Medication Algorithm Project, phase 3 (TMAP-3): rationale and study design. Journal of Clinical Psychiatry 64:357–369, 2003
 
Dennehy E, Suppes T: Medication algorithms for bipolar disorder. Journal of Practical Psychiatry and Behavioral Health 5:142–152, 1999
 
Trivedi MH, Rush AJ, Crismon ML, et al: Clinical results for patients with major depressive disorder in the Texas Medication Algorithm Project. Archives of General Psychiatry 61:669–680, 2004
 
Suppes T, Rush AJ, Dennehy EB, et al: Texas Medication Algorithm Project, phase 3 (TMAP-3): clinical results for patients with a history of mania. Journal of Clinical Psychiatry 64:370–382, 2003
 
Greenhalgh T, Robert G, Macfarlane F, et al: Diffusion of innovations in service organizations: systematic review and recommendations. Milbank Quarterly 82:581–629, 2004
 
Fixsen DL, Blase KA: Creating new realities: program development and dissemination. Journal of Applied Behavioral Analysis 26:597–615, 1993
 
Trivedi MH, Claassen CA, Grannemann BD, et al: Assessing physicians' use of treatment algorithms: Project IMPACTS study design and rationale. Contemporary Clinical Trials 28:192–212, 2007
 
Miller AL, Crismon ML, Rush AJ, et al: The Texas Medication Algorithm Project: clinical results for schizophrenia. Schizophrenia Bulletin 30:627–647, 2004
 
+

References

Suppes T, Swann AC, Dennehy EB, et al: Texas Medication Algorithm Project: development and feasibility testing of a treatment algorithm for patients with bipolar disorder. Journal of Clinical Psychiatry 62:439–447, 2001
 
Kashner TM, Carmody TJ, Suppes T, et al: Catching up on health outcomes: the Texas Medication Algorithm Project. Health Services Research 38:311–331, 2003
 
Rush AJ, Crismon ML, Kashner TM, et al: Texas Medication Algorithm Project, phase 3 (TMAP-3): rationale and study design. Journal of Clinical Psychiatry 64:357–369, 2003
 
Dennehy E, Suppes T: Medication algorithms for bipolar disorder. Journal of Practical Psychiatry and Behavioral Health 5:142–152, 1999
 
Trivedi MH, Rush AJ, Crismon ML, et al: Clinical results for patients with major depressive disorder in the Texas Medication Algorithm Project. Archives of General Psychiatry 61:669–680, 2004
 
Suppes T, Rush AJ, Dennehy EB, et al: Texas Medication Algorithm Project, phase 3 (TMAP-3): clinical results for patients with a history of mania. Journal of Clinical Psychiatry 64:370–382, 2003
 
Greenhalgh T, Robert G, Macfarlane F, et al: Diffusion of innovations in service organizations: systematic review and recommendations. Milbank Quarterly 82:581–629, 2004
 
Fixsen DL, Blase KA: Creating new realities: program development and dissemination. Journal of Applied Behavioral Analysis 26:597–615, 1993
 
Trivedi MH, Claassen CA, Grannemann BD, et al: Assessing physicians' use of treatment algorithms: Project IMPACTS study design and rationale. Contemporary Clinical Trials 28:192–212, 2007
 
Miller AL, Crismon ML, Rush AJ, et al: The Texas Medication Algorithm Project: clinical results for schizophrenia. Schizophrenia Bulletin 30:627–647, 2004
 
+
+

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



Related Content
Articles
Books
The American Psychiatric Publishing Textbook of Geriatric Psychiatry, 4th Edition > Chapter 33.  >
The American Psychiatric Publishing Textbook of Geriatric Psychiatry, 4th Edition > Chapter 33.  >
The American Psychiatric Publishing Textbook of Geriatric Psychiatry, 4th Edition > Chapter 33.  >
The American Psychiatric Publishing Textbook of Substance Abuse Treatment, 4th Edition > Chapter 42.  >
Dulcan's Textbook of Child and Adolescent Psychiatry > Chapter 30.  >
Topic Collections
Psychiatric News
PubMed Articles