Loading...
Please wait, while we are loading the content...
Similar Documents
How do we know how well we are doing?
| Content Provider | Semantic Scholar |
|---|---|
| Author | O'connor, Gerald T. Eagle, Kim A. |
| Copyright Year | 1998 |
| Abstract | It has been 88 years since the summer day in 1910 when Boston surgeon Ernest Amory Codman first proposed that hospitals should publish their mortality rates (1). He reasoned that both physicians and the public needed to know the outcomes of medical and surgical care. His ideas were immediately unpopular in the professional community. Concerns were expressed about differences in the patient case mix, about the effect that public data release would have on physicians’ willingness to tackle difficult cases and about migration of patients in response to data publication. In recent years many of these controversies were rekindled by a number of local or statewide public data releases (2,3). Proponents of such programs say that physicians need comparative data on outcomes to manage and improve clinical care. Some proponents say the public has a right to know the outcomes of clinical care when they choose the person or institution to provide their medical or surgical care. Opponents raise concerns about the validity of methods to adjust for differences in patient case mix, the statistical instability of low rates of adverse outcomes and the chilling effect that these data will have on physician decision making. The only fundamental changes in nearly a century of debate over this issue are the development of multivariate statistics and that these data release programs are now a mandated reality in many areas of the United States. Perhaps no data release program has undergone more intense scrutiny than the New York State Cardiac Surgery Reporting System. This program has been in place since 1988 and initially focused on the short-term outcomes of coronary artery bypass graft (CABG) surgery. The program has developed methods for multivariate adjustment for patient case mix (4) and has reported hospital and physician outcomes since 1991 (5,6). The program has been criticized for physician “gaming” of the data reports and for inadequacy of its statistical methods (7). However, an article by Jones et al. (8) shows similarity of results of the multivariate methods used by the New York State Program and those used in other locations. Further, validation studies conducted by the New York State Program showed that gaming the data collection system was not a likely cause of the differences in risk-adjusted mortality rates (9). In 1994 the New York State Program reported a substantial reduction in the perioperative mortality associated with CABG surgery and attributed this, at least in part, to the effects of the program (10). Since then two published articles have used indirect evidence to claim that the improved outcomes of CABG surgery seen in New York were a result of the migration of sicker patients to surrounding states or a result of a temporal trend toward improved outcomes in the United States. Data on consecutive cases of CABG surgery seen at the Cleveland Clinic between 1989 and 1993 were examined by Omoigui et al. (11). Based on 485 patients residing in New York but receiving CABG surgery at Cleveland Clinic, the authors concluded that New York patients were more likely than other patients to have had prior cardiac surgery, to be New York Heart Association functional class III or class IV and have experienced higher mortality rates. The authors concluded that public dissemination of outcome data may have been associated with increased referral of high-risk patients from New York. Examination of hospital discharge data from Massachusetts were used by Ghali et al. (12) to evaluate cardiac surgical mortality at 12 hospitals. These mortality rates improved from 4.7% in 1990 to 3.5% in 1992 and 3.3% in 1994. The authors concluded that since improvement occurred in Massachusetts, which had no statewide outcome reporting program, the improvement reported in New York may have been largely a consequence of a regional decline in cardiac surgery mortality. In this issue of the Journal of the American College of Cardiology, Peterson et al. (13) present an analysis of the CABG outcomes in New York State. These findings are based on the national Medicare claims data. They studied claims and clinical outcomes on over 700,000 CABG procedures to assess both migration from New York State and average annual improvement in CABG surgery during the period 1987 to 1992. The results do not support either migration or temporal change as the cause of the apparent improvement in CABG mortality in New York State. Peterson et al. found that the overall migration from New York State actually decreased during that period. Further, the comparison with data from other states showed that New York State had the lowest CABG mortality in the United States and was the most improved of the low mortality states during 1987 to 1992. Why did Peterson et al. reach a different conclusion than did Omoigui et al. and Ghali et al.? Omoigui et al. showed that, compared to patients from other areas and to historical controls, there was an increased severity of CABG patients referred to the Cleveland Clinic from New York during 1988 to 1992. This was almost certainly true but it may have been largely a local effect. Omoigui et al. believed this was evidence of a general migration of CABG patients from New York: it was not. More comprehensive data presented by Peterson et al. actually show a decline in migration from New York during this time period. Ghali et al. showed improvement in CABG *Editorials published in Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology. From the *Dartmouth Medical School, Hanover, New Hampshire and †The University of Michigan Health System, Ann Arbor, Michigan. Address for correspondence: Dr. Gerald T. O’Connor, 330 Strasenburgh Hall, Box 7250, Dartmouth Medical School, Hanover, New Hampshire 03756. JACC Vol. 32, No. 4 October 1998:1000–1 1000 |
| File Format | PDF HTM / HTML |
| DOI | 10.1016/S0735-1097(98)00331-3 |
| PubMed reference number | 9768724 |
| Journal | Medline |
| Volume Number | 32 |
| Issue Number | 4 |
| Alternate Webpage(s) | http://www.onlinejacc.org/content/accj/32/4/1000.full.pdf?origin=publication_detail |
| Alternate Webpage(s) | https://doi.org/10.1016/S0735-1097%2898%2900331-3 |
| Journal | Journal of the American College of Cardiology |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
