To the Editor: Dr. James Morrison's article (
1) in the August 1998 issue describing a staffing model based on workload at three Veterans Affairs mental health clinics was of great interest to us, as we are grappling with such issues at the state-operated prepaid mental health plan in New York State. We appreciated the author's rigorous effort to measure productivity while taking many variables into account. However, we would like to raise questions about some issues that were not clarified in the article.
Dr. Morrison concluded that two psychiatrists could evaluate 400 new patients and provide ongoing treatment for 2,300 patients a year. He thus implies that a single psychiatrist could have the responsibility for managing a caseload of 1,150 patients in addition to evaluating 200 new patients a year. However, the data presented support the conclusion that each psychiatrist could have only 2,808 half-hour patient time slots in a year, without any reference to standards for frequency of patient visits with the psychiatrist.
Subtracting 400 slots for the 200 one-hour new evaluations leaves 2,408 half-hour slots, which can be consumed by a caseload of 200 patients seen once a month, a frequency that would seem clinically prudent with a population of patients with serious mental illness that the VA purports to serve. According to this calculation, the psychiatric staffing recommended in the model appears to be quite inadequate.
Furthermore, the issues of missed appointments, outreach, crisis intervention, and arrangements for transfer to partial hospital or inpatient settings were seemingly unaccounted for or lumped under the 1.5 hours per day allotted for meetings and emergencies.
It is well known that the need for psychiatric staffing in mental health clinics can be arbitrarily changed by altering the standards for required frequency of patient visits with the psychiatrist and by supplying medications for longer periods. Similar concerns about the frequency of visits apply to the staff members who are not psychiatrists.
Another issue is the quality of care provided when psychiatrists and other staff have large caseloads. It would have been useful to include typical outcome measures such as rehospitalization rates, consumer satisfaction, and symptom improvement in assessing this staffing model. Recently, Young and others (
2) have made some effort to measure the quality of care given in large outpatient centers, including VA clinics. Their findings, considered in the context of staffing patterns at these clinics, would be of interest to consumers and policy makers.
We believe that further research is needed to correlate quality of care and improved outcomes with staffing models. While financial issues are important, administrators and clinicians still need to review the clinical effects of having high patient-to-staff ratios at clinics treating seriously ill psychiatric patients.
