Opinions on provider profiling: telephone survey of stakeholders.

The views of producers, purchasers, and users of provider profiling concerning this practice were studied. A snowball sample of individuals representing seven groups with a stake in retrospective provider profiling were interviewed by telephone over a 12-week period in 1997. Participants were asked what they believed were the most important uses for profiles, who should receive copies of profiles, and what the limitations of profiles are. A semi-structured format was used to ensure that each interview was comparable and complete. The responses were aggregated, and qualitative research approaches were used to analyze them. A total of 40 people were interviewed. A majority of the respondents cited physician education, changing physician behavior, and monitoring and improving the quality of care as valid uses of provider profiles. A majority believed that the recipients of profile data should include the individual providers being profiled, medical administrative staff, people directly involved in the profiling program, pharmacists, and health plan administrators. The respondents acknowledged many limitations of profiles, with the top concern being inherent problems in the use of billing and administrative databases for profiling. Interviews with stakeholders in provider profiling yielded insights into the strengths and weaknesses of profiling, as well as echoing findings reported elsewhere. Health system administrators and health care professionals need to be aware of these issues as they use and modify profiling.

Data linkages for research on outcomes of long-term care.

Medicaid claims were linked with a sample of data gathered for nursing home quality assurance and case mix reimbursement (N = 14,917). This generated patient level records of medical resource use combined with standardized assessments of health and functional status. We describe the linkage process and the characteristics of the combined set of information-its strengths, limitations, and potential uses. The two data sources, one cross-sectional and the other longitudinal, complemented one another and provided a more complete description of patient resource use and health status. However, methodological and ethical issues must be resolved before data linkages are used routinely for research.

Evaluation of a DUR intervention: a case study of histamine antagonists.

Drug Utilization Review (DUR) programs aim to control inappropriate prescribing and drug costs, but there are few rigorous studies of their impact. This paper presents the evaluation results of a DUR intervention (excessive duration of full-dose histamine-2 receptor antagonists) using a quasi-experimental design. We observe a greater reduction of H2RA use in the intervention group than in the comparison group, and a reduction in drug expenditures which exceeded the cost of targeting and notifying prescribers. The success of DUR programs depends on both their efficiency and the drug issues under consideration; the use of a comparison group is critical in the evaluation of their impact.

Pharmacists' reactions to the Wisconsin Medicaid drug-use review program.

Pharmacists' reactions to inquiries on Medicaid drug-use review (DUR) and how the information supplied by such inquiries influenced pharmacists' interactions with physicians were studied. A questionnaire was mailed in September 1991 to all 137 pharmacies that had been contacted in 1989 by the Wisconsin Medicaid DUR program about potentially inappropriate use of histamine H2-receptor antagonists. The survey sought to determine whether pharmacists used information in the DUR letters and, if so, how. A section for consultants was included to identify possible differences in responses between consultant and nonconsultant pharmacists. A response was received from 96 of the 137 pharmacies contacted. Pharmacists reported that the issues raised by the DUR letter frequently were discussed with the prescriber. Most respondents attempted to make changes in drug therapy based on the information supplied in the DUR letter. Consultant pharmacists were less likely than nonconsultants to refer to the DUR letter but were more likely to consider inclusion of copies of references to the medical literature helpful, to contact the prescriber in writing, to discuss alternative therapies or discontinuation of therapy, and to suggest patient-monitoring activities. Although consultants and nonconsultants tend to use DUR information in different ways, DUR does offer helpful information that is used by Wisconsin pharmacists in making drug therapy recommendations to prescribers.

Presentation of drug prescribing guidelines and physician response.

Most Medicaid Drug Use Review (DUR) Programs disseminate prescribing practice guidelines by providing written reviews to physicians about specific prescribing practices that do not follow established criteria and inviting their comment. This study reviewed physicians' written responses to explore relationships between DUR communications and the degree to which these communications affect awareness of prescribing issues, increase knowledge, change erroneous beliefs, and generate intentions to apply the information appropriately. The form, content, and effective tone for all responses received in 1990 regarding five different drug therapy issues (N = 240) was recorded. Response rates, comprehension, counterarguments, and prescribing intentions varied by drug therapy issue. Attention to the communication process could improve the dissemination of prescribing practice guidelines by DUR Programs.

An examination of histamine-2 receptor antagonist use by Medicaid recipients in Wisconsin long-term care facilities.

The Wisconsin Drug Utilization Review Project used Medicaid claims data to examine the use of long-term, full doses histamine-2 receptor antagonists (H2RAs) among long-term care facility (LTCF) residents, ie, doses exceeding 800 mg per day of cimetidine or 150 mg per day of ranitidine for 12 weeks or more. Additional information about a small but representative sample of patients was obtained from consultant pharmacists by mail questionnaire. Of approximately 35,000 residents, there were 1,046 receiving long-term, full-dose H2RAs during 1984 at a cost of +405,848. No differences were noted between treated patients and the overall LTCF population by gender, age, level of care, or geographic locality. Among 387 patients receiving therapy for the last 3 months of 1984, over one half received antacids and over 20% received nonsteroidal anti-inflammatory drugs concurrently. According to survey respondents, almost one half of the patients had no diagnosis amenable to treatment by H2RAs. Diagnostic procedures conducted within 30 days prior to the initiation of H2RAs were hemoglobin/hematocrit assessment (performed on 53% of patients), stool testing (34%), radiological exam (23%) and endoscopy (8%). The results suggest that there is widespread use of long-term full-dose H2RAs in LTCFs for undocumented, inappropriate, and poorly supported diagnoses. Such findings are of concern to all persons charged with monitoring drug use in nursing homes because of the implications for both quality and cost of care. Information from this study will be used in designing a program to reduce inappropriate utilization of H2RAs, and Medicaid claims data will be used to monitor trends in utilization.

Use of a Medicaid database to analyze pharmaceutical services in long-term-care facilities.

A profile of Wisconsin pharmacies that provide services to Medicaid recipients in long-term-care facilities (LTCFs) was developed, and the relationships among pharmacy characteristics and drug use and reimbursement for pharmaceutical services in these pharmacies were studied. Computer-generated reports of Medicaid claims of all pharmacies that had received $1000 or more for services provided to LTCFs from July through September 1984 were included in the study; these claims were submitted by 255 pharmacies on behalf of 32,410 recipients. The reports contained the following individual pharmacy data: geographic location, total dollar payments, total number of Medicaid recipients served, percentage of recipients residing in LTCFs, total number of payments from Medicaid for prescriptions, amount paid to the pharmacy for LTCF recipients as a percentage of total payments, percentage of prescriptions paid at the unit dose rate, and average number of prescriptions and cost per LTCF recipient. An average of $137.30 per recipient was reimbursed, representing an average of 12.3 prescriptions per recipient. The average number of prescriptions per recipient did not vary among pharmacies on the basis of unit dose or traditional distribution system, vendor-consultant status of the pharmacist, location of the pharmacy, or the proportion of LTCF claims to total Medicaid prescription claims. The average amount paid per recipient was related to the average number of prescriptions per recipient, the use of a unit dose system, and the proportion of nursing home claims to total claims submitted. A Medicaid database can be useful for characterizing pharmaceutical services and drug use in LTCFs.