Assessing outcomes of cardiac procedures

Study identifies age as significant factor

By Harold D. Taylor, PhD

Statistician

GE Medical Systems

Englewood, CO

Mary Jo Strobel, BSN, MBA

Past Product Manager

Functional Outcomes Monitoring Program

LBA/MECON

Englewood, CO

Current Senior Business Analyst

Clinical Information Systems

Kaiser Permanente

Denver

Over the past 20 years of the health care quality evolution, the definition of quality has remained fairly stable. The Institute of Medicine in Washington, DC, defines quality as "the degree to which health services for individuals and populations increase the likelihood of desired outcomes and are consistent with current professional knowledge."1

Comprehensive quality programs generally target the key areas of clinical, professional, and service quality including patient satisfaction. Yet, the need for innovative quality programs has changed over the past several years, moving toward evidence-based validation of results.

One reason for this phenomenon is shorter lengths of stay resulting from stringent managed care cost containment programs. Patients now are discharged far sooner, and usually before the full benefits of a procedure can be ascertained. Emphasis has shifted to managing the patient after discharge.

A second factor is the Internet, where patients can access sophisticated clinical information. Increasingly, these better-informed patients take a more active role in selecting health care options and request data demonstrating long-term benefits of their treatments. Both of these forces drive the need for developing databases that focus on monitoring the functional status of patients over time and linking these results with clinical data. Such a tool will provide a thorough evaluation of the value of care rendered.

This is the new paradigm: "The true value of health care can be determined only by a systematic examination of patient outcomes."2 The following statements were made by physicians engaged in a February 1999 Internet discussion on the need for outcomes measurements:

• "No one knows the outcomes or value of many of the listed procedures in the elderly, but they are being done at ever-increasing rates. Until payers can demand outcomes data including pre- and post-procedure functional status and pay based on this, we will not substantially improve our performance."

• "The lack of data to demonstrate patient or population benefit is striking."

• "Without data that [demonstrate that] the health status of over 85-year-[old] females has been improved by the rise in hip surgery, one believes that the rise in surgery has other causes — not relating to health status or medical conditions."

LBA Consulting Group developed CORE (Clinical Outcomes Review and Evaluation)3 in 1997 to track the outcomes of certain cardiac and orthopedic patients. Because such data are only of benefit if collected by a standardized, statistically valid psychometric tool, the 36-Item Short Form4 developed out of the Medical Outcomes Study was used. The survey instrument contains 36 questions, which then are rolled up into eight scales. Published benchmarks are available for all scales.

Scales include:

1. Role-physical

2. Bodily pain

3. General health

4. Physical functioning

5. Vitality

6. Social functioning

7. Mental health

8. Role-emotional

Prior to admission, client hospitals distribute a modified 36-Item Short Form survey instrument, to which several additional questions pertaining to a cardiac population have been appended. Patients are informed that their physician and the hospital wish to track their progress after discharge and request their cooperation in completing and returning similar surveys throughout the coming year.

Follow-up surveys are printed, coded, and mailed by LBA at three-, six-, and 12-month intervals. The information is then combined with the UB92 Discharge Summary data. Comparative reports are periodically distributed to client hospitals. This provides an analysis of hospital performance, as well as performance against other hospitals and national benchmarks.

The purpose of our study was to assess the longitudinal functional status of three distinct invasive cardiology patient populations: patients undergoing PTCA (percutaneous transluminal coronary angioplasty) with stent (a wire mesh implant, designed to hold the artery open), patients undergoing PTCA without stent, and patients undergoing CABG (coronary artery bypass graft). These patient populations were selected based on their similar clinical presentations on admission.5 This study concentrates on the improvements of functional status three months following treatment.

The data set used for this study consisted of 974 patients from 11 hospitals with discharge dates ranging from July 22, 1997 through Sept. 30, 1999 (468 CABG, 91 PTCA, and 416 stent). These patients completed both the initial and the three-month follow-up questionnaire, and the hospital was able to provide clinical data via a legitimate UB92 record. The clinical data were used to determine age, sex, the procedure performed, and whether or not the patient had experienced acute myocardial infarction (AMI).

Surprisingly, the three treatment groups had similar scores upon admission. In fact, the only scale showing a statistically significant difference among the groups was physical functioning, where the CABG patients had significantly lower scores. Age and/or sex showed statistically significant effects for all scales, with females and older patients scoring lower. As one example, initial physical functioning scores for females averaged 42.7, vs. 58.4 for males, adjusted for other factors. (See graphs, pp. 69-71. All data shown in the graphs have been statistically adjusted for the influences of other factors.)

The differences between males and females were surprisingly large, raising the question, once again, of whether females are treated as aggressively as males.6,7

Differences among the three treatment groups are also worth noting. The CABG patients, in general, started out with lower scores than the PTCA or stent patients. Their improvement after treatment, however, resulted in generally higher scores after three months.

The notable exception was role-physical, in which the CABG patients showed much less improvement, perhaps because of the longer recovery time for CABG surgery. The questions for this measure relate to whether the patient had difficulty in performing activities or had to cut down the amount of time spent on activities, so the longer rehabilitation period could easily have an effect. The results at six months will be interesting.

For two other measures, physical functioning and vitality, the CABG patients showed much greater improvement. They started out with lower scores but had higher scores after three months.

Several findings were of interest because of the lack of statistical significance. Presence of AMI was only significant in the improvement of the bodily pain metric, and age was not significantly associated with improvement in any of the measures. Apparently, the benefits of these procedures span all age groups.

References

1. Bodenheimer T. The American health care system — the movement for improved quality in health Care. N Engl J Med 1999; 340:6

2. Gerszten PC. Outcomes research: A review. Neurosurgery 1998; 43(5):1,146-1,156.

3. The CORE program (formerly known as Response Plus) was originally developed as a joint venture between two HCIA divisions: LBA Consulting Group and HCIA Response office. In February 1999, after a corporate spinoff, the LBA Consulting Group took over all CORE operations. The LBA Consulting Group has recently been acquired by GE Medical Systems. For more information, contact: GE Medical Systems, Englewood, CO. Telephone: (303) 740-7779.

4. Sherbourne CD, McHorney CA, Ware JE, et al. The MOS 36-item Short Form Survey, III: Tests of data quality, scaling assumptions, and reliability across diverse patient groups. Med Care 1994; 32(1):40-66.

5. Note: Limitations in the current data preclude ability to assess significant clinical findings on admission such as the degree and location of coronary occlusion. An opportunity exists in the future to combine the functional outcomes data with key preoperative clinical findings.

6. Roger VL, Farkouh ME, et al. Sex differences in evaluation and outcome of unstable angina. JAMA 2000; 283:5.

7. Steingart RM, Packer M, et al Sex differences in the management of coronary artery disease. N Engl J Med 1991; 325:226-230.

[Editor’s note: For additional information or updates on this research, contact Harold Taylor, GE Medical Systems, 6300 S. Syracuse Way, Suite 630, Englewood, CO 80111. Telephone: (303) 714-9532. E-mail: harold. taylor@med.ge.com. For further insights into managing cardiac conditions as well as other chronic care concerns of the elderly, see QI/TQM’s two-part series "DM (disease management) and chronic care, in the January and February 2000 issues.]