This study retrospectively evaluated patient records from two cardiac rehabilitation (CR) service centers located in large urban hospitals using a Process Evaluation System (PES) recently developed through a collaborative project of the American Association of Cardiovascular and Pulmonary Rehabilitation (AACVPR), Madison, WI, and the Center for Clinical Quality Evaluation (CCQE), Washington, DC. The major aims were to: 1) evaluate the utility of the PES as an audit instrument for assessment of adherence to the 24 quality process criteria that comprised the PES; and 2) determine whether adherence to the PES criteria resulted in different patient outcomes for those cases where intervention need was documented at patient admission. Using the data abstraction manual and audit procedures developed by AACVPR/CCQE, a trained medical technician audited 150 CR records for consecutively treated outpatients who typically received 36 sessions of treatment in either Moses H. Cone Memorial Hospital, N.C. Heart Institute, Greensboro, NC, or Carolina's Medical Center, Charlotte, NC, covering a calendar period between 1995-97. The data were pooled from both sites for analyses and included patients with one or more of the following diagnoses: MI (37%), angina (14%), coronary revascularization (76%), and other (18%). The cost of utilizing the PES was assessed by evaluating the technician time required to abstract a patient record and this was observed to improve over the course of the review period, i.e., mean abstraction time for initial versus final 20 records = 13.2 min. and 4.6 min., respectively. Experience with the PES suggested areas where instrument revision should be considered, e.g., the operational guidelines for extracting acceptable markers were not always clear enough or sufficiently flexible to allow determination of adherence of a record to the 24 quality process criteria. Adherence to the PES was determined, case by case, for each of the 24 criteria. In 129 cases (86% of the sample), complete adherence was found, i.e. 100% adherence to all 24 criteria that included indicators of key clinical steps for patient intake, treatment planning, and follow-up. The remaining 21 records (14%) showed adherence to at least 21 of the 24 criteria (87.5%). Given the uniformly high levels of adherence to the PES documented by these two program sites, the data could not resolve the question of whether patient outcome effects were different between cases of high versus low adherence to PES. Nonetheless, outcome data were examined to evaluate achievement levels in four different areas widely considered by clinicians as important to treatment success: blood cholesterol, smoking status, exercise tolerance, and body mass index (BMI). Of the study patients diagnosed with dyslipidemia 12 of 27 (44%) had levels < 200 mg/dl by exit. Seven of 14 documented smokers (50%) reported quitting at exit from treatment. Forty-nine patients of 117 (42%) who initially could only maintain treadmill walking for 10 min. at levels below 4 METs, were able to exceed this level by treatment end. Six of 104 (6%) with BMI values > 24.9 kg/m2 had a documented decrease in this indicator of overweight by treatment end. The threshold levels for outcome criteria used here to describe achievement levels in this data set are somewhat arbitrary. However, the criteria are reflective of the standards typically suggested as meaningful for effective secondary risk reduction in CR programs (Franklin et al., 1996). The PES system was developed to audit the quality of CR process in treatment centers, as standardized by a consensus panel to reflect the content of the evidenced-based CR guideline recently published by the US Agency for Health Care Policy and Research (Cardiac Rehabilitation as Secondary Prevention: #17, 1995). The findings of this study suggest that the content markers of quality process in the PES are relevant and the instrument is efficient to administer. When field tested against two urban centers in North Carolina where state statutes require program certification for CR treatment centers, these centers demonstrated uniformly high adherence to the PES and a pattern of good achievement for several patient outcome measures accepted as relevant to evaluation of treatment success for individual patients.