Measuring the quality of diabetes care using administrative data: is there bias?

OBJECTIVES

Health care organizations often measure processes of care using only administrative data. We assessed whether measuring processes of diabetes care using administrative data without medical record data is likely to underdetect compliance with accepted standards for certain groups of patients.

DATA SOURCES/STUDY SETTING

Assessment of quality indicators during 1998 using administrative and medical records data for a cohort of 1,335 diabetic patients enrolled in three Minnesota health plans.

STUDY DESIGN

Cross-sectional retrospective study assessing hemoglobin A1c testing, LDL cholesterol testing, and retinopathy screening from the two data sources. Analyses examined whether patient or clinic characteristics were associated with underdetection of quality indicators when administrative data were not supplemented with medical record data.

DATA COLLECTION/EXTRACTION METHODS

The health plans provided administrative data, and trained abstractors collected medical records data.

PRINCIPAL FINDINGS

Quality indicators that would be identified if administrative data were supplemented with medical records data are often not identified using administrative data alone. In adjusted analyses, older patients were more likely to have hemoglobin A1c testing underdetected in administrative data (compared to patients <45 years, OR 2.95, 95 percent CI 1.09 to 7.96 for patients 65 to 74 years, and OR 4.20, 95 percent CI 1.81 to 9.77 for patients 75 years and older). Black patients were more likely than white patients to have retinopathy screening underdetected using administrative data (2.57, 95 percent CI 1.16 to 5.70). Patients in different health plans also differed in the likelihood of having quality indicators underdetected.

CONCLUSIONS

Diabetes quality indicators may be underdetected more frequently for elderly and black patients and the physicians, clinics, and plans who care for such patients when quality measurement is based on administrative data alone. This suggests that providers who care for such patients may be disproportionately affected by public release of such data or by its use in determining the magnitude of financial incentives.

Trends in the Aggressiveness of Cancer Care Near the End of Life

Purpose

To characterize the aggressiveness of end-of-life cancer treatment for older adults on Medicare, and its relationship to the availability of healthcare resources.

Patients and Methods

We analyzed Medicare claims of 28,777 patients 65 years and older who died within 1 year of a diagnosis of lung, breast, colorectal, or other gastrointestinal cancer between 1993 and 1996 while living in one of 11 US regions monitored by the Surveillance, Epidemiology, and End Results Program.

Results

Rates of treatment with chemotherapy increased from 27.9% in 1993 to 29.5% in 1996 (P = .02). Among those who received chemotherapy, 15.7% were still receiving treatment within 2 weeks of death, increasing from 13.8% in 1993 to 18.5% in 1996 (P < .001). From 1993 to 1996, increasing proportions of patients had more than one emergency department visit (7.2% v 9.2%; P < .001), hospitalization (7.8% v 9.1%; P = .008), or were admitted to an intensive care unit (7.1% v 9.4%; P = .009) in the last month of life. Although fewer patients died in acute-care hospitals (32.9% v 29.5%; P < .001) and more used hospice services (28.3% v 38.8%; P < .001), an increasing proportion of patients who received hospice care initiated this service only within the last 3 days of life (14.3% v 17.0%; P = .004). Black patients were more likely than white patients to experience aggressive intervention in nonteaching hospitals but not in teaching hospitals. Greater local availability of hospices was associated with less aggressive treatment near death on multivariate analysis.

Conclusion

The treatment of cancer patients near death is becoming increasingly aggressive over time.

Consultation With a Medical Oncologist Before Surgery and Type of Surgery Among Elderly Women With Early-Stage Breast Cancer

Purpose: Prior studies have documented variation in breast cancer treatment and care that does not follow guideline recommendations, particularly for elderly women. We assessed whether consultation with a medical oncologist before surgery was associated with use of definitive surgery, axillary node dissection, and type of surgery.

Methods: We conducted a retrospective cohort study of a population-based sample of 9,630 women aged ≥ 66 years diagnosed with breast cancer during 1995 to 1996. We measured the adjusted proportion visiting a medical oncologist before surgery, identified factors associated with such visits, and assessed the association between visits with a medical oncologist and use of definitive surgery (mastectomy or breast-conserving surgery with radiation v breast-conserving surgery without radiation); axillary dissection; and breast-conserving surgery versus mastectomy among women undergoing definitive surgery.

Results: Nineteen percent of women visited a medical oncologist before surgery; these women were younger, more often had larger or more poorly differentiated cancers, had more comorbid illnesses, and were treated more often at a teaching hospital (all P < .05). Women who saw a medical oncologist before surgery were more likely than others to undergo definitive surgery (adjusted odds ratio [OR], 1.28; 95% CI, 1.05 to 1.56) and axillary dissection (adjusted OR, 1.44; 95% CI, 1.19 to 1.73), but less likely to undergo breast-conserving surgery among women undergoing definitive surgery (OR, 0.84; 95% CI, 0.75 to 0.95).

Conclusion: Elderly women who consulted with a medical oncologist before surgery were more likely to receive guideline-recommended care. Additional research is needed allow a better understanding of the quality and content of discussions that elderly women have with various providers about breast-conserving surgery and mastectomy.

Breast Cancer in Older Women: Quality of Life and Psychosocial Adjustment in the 15 Months After Diagnosis

Purpose: We examined the health-related quality of life (QOL) of a cohort of older women with breast cancer after their diagnosis.

Patients and Methods: Six hundred ninety-one women aged 65 years and older were interviewed approximately 3 months after breast cancer surgery and two additional times in the following year using standardized QOL measures. Demographic factors, breast cancer treatments, and comorbid conditions were used to model ratings of health-related QOL over time. Self-perceived health and psychosocial adjustment at 15 months after surgery were modeled.

Results: Physical and mental health scores declined significantly in the follow-up year, independent of age. However, a cancer-specific psychosocial instrument showed significant improvement in scores. Better 3-month physical and mental health scores, as well as better emotional social support, predicted more favorable self-perceived health 15 months after surgery. Psychosocial adjustment at 15 months was significantly predicted by better mental health, emotional social support, and better self-rated interaction with health care providers assessed at 3 months.

Conclusion: Contrary to reports from younger women with breast cancer, we observed significant declines in the physical and mental health of older women in the 15 months after breast cancer surgery, whereas scores on a cancer-specific psychosocial QOL measure improved over time, consistent with patterns in younger women. Predictive models indicate that older women with impaired physical functioning, mental health, and emotional social support after surgery have poorer self-perceived health and psychosocial adjustment 1 year later. Interventions to address the physical and emotional needs of older women with breast cancer should be developed and evaluated to determine their impact on subsequent health-related QOL.

Specialty of ambulatory care physicians and mortality among elderly patients after myocardial infarction

BACKGROUND

The outcome after myocardial infarction may be influenced by the type of physician providing ambulatory care.

METHODS

We studied 35,520 patients 65 years of age or older who were hospitalized for myocardial infarction in seven states during 1994 and 1995 and who survived for at least three months after discharge. From Medicare claims, we identified ambulatory visits to cardiologists, internists, and family practitioners. Using propensity scores to adjust for demographic, clinical, and hospital characteristics, we analyzed treatment and mortality at two years among patients matched according to their estimated propensity to receive care from a cardiologist within three months after discharge.

RESULTS

As compared with patients who saw only an internist or a family practitioner in the three months after discharge, patients who saw a cardiologist were younger, were more likely to be white, were more likely to be male, had fewer coexisting conditions, and were more likely to have undergone invasive cardiac procedures while hospitalized (P<0.01 for all comparisons). Patients who saw a cardiologist were more likely to undergo cardiac procedures and rehabilitation after discharge. Patients who saw a cardiologist had a lower two-year mortality rate than matched patients who saw only an internist or a family practitioner (14.6 percent vs. 18.3 percent, P<0.001). Patients who saw both a cardiologist and an internist or a family practitioner had a lower mortality rate than matched patients who saw only a cardiologist (11.1 percent vs. 12.1 percent, P=0.02).

CONCLUSIONS

Ambulatory visits to cardiologists were associated with greater use of cardiac procedures and decreased mortality after myocardial infarction. Concurrent care by an internist or a family practitioner was associated with a further reduction in mortality.

Understanding individual and small area variation in the underuse of coronary angiography following acute myocardial infarction

BACKGROUND

Underuse of coronary angiography is common among patients with acute myocardial infarction (AMI) and the magnitude of underuse varies across geographic areas.

OBJECTIVES

To examine the influence of patient demographic, clinical and hospital characteristics on underuse of coronary angiography, and the contribution of these factors to variation in underuse across geographic regions.

RESEARCH DESIGN

Cohort study using data from the Cooperative Cardiovascular Project.

SUBJECTS

Nine thousand four hundred fifty-eight patients in 95 hospital referral regions (HRRs) hospitalized for AMI in 1994 to 1995 and for whom angiography was rated necessary.

MEASURES

Odds ratios (95% confidence intervals) associated with underuse of angiography according to patient and hospital characteristics. The difference between low and high rates of underuse of angiography across regions after controlling for regional differences in patient and hospital characteristics.

RESULTS

Of those for whom angiography was rated necessary, 42% did not undergo the procedure. Underuse of angiography was associated with several patient demographic and hospital attributes (eg, female gender, black race, treatment in a hospital without angiography, treatment by a general practitioner) as well as with prevalent clinical characteristics, such as renal insufficiency, congestive heart failure, prior coronary artery bypass surgery, and chronic obstructive pulmonary disease. Across HRRs, variation in underuse ranged from 24.0% to 58.3%. The difference between low and high rates did not decline significantly after controlling for regional differences in patient or hospital characteristics.

CONCLUSIONS

At the patient-level, rates of necessary angiography may be improved if we address disparities in care related to sociodemographic characteristics and to the technological capabilities of hospitals. In addition, practice guidelines should be updated to reflect clinical concerns about the risks and benefits of angiography and subsequent revascularization in certain patient sub-groups, both to provide appropriate guidance to physicians and to facilitate better estimates of underuse. The causes of regional variation in underuse do not appear to be related to regional differences in patient or hospital characteristics, and therefore, require further study.

Adjuvant tamoxifen prescription in women 65 years and older with primary breast cancer

PURPOSE

We examined patterns of adjuvant tamoxifen discussion and prescription among breast cancer patients age 65 years and older.

METHODS

We selected from women diagnosed with primary breast cancer those with (1) stage I (tumor diameter > or = 1 cm), stage II, or stage IIIa disease; (2) age 65 years or older on the date of diagnosis; and (3) permission from the attending physician to contact. Data were collected from consenting patients' medical records, telephone interviews with patients, and mailed questionnaires completed by their physicians.

RESULTS

We obtained medical record and interview data for 698 patients. The oldest patients (odds ratio [OR], 0.45; 95% confidence interval [CI], 0.23 to 0.87 for those aged 80+ relative to those aged 65 to 69 years old), those with more comorbid conditions (each additional comorbid condition reduced the odds of discussion by 0.84; 95% CI, 0.73 to 0.96), and those who were estrogen receptor-negative (OR, 0.56; 95% CI, 0.32 to 0.99) were less likely to report discussion of tamoxifen therapy with a physician. Older patients (OR, 2.17; 95% CI, 1.18 to 4.01 for 70- to 79-year-olds relative to 65- to 69-year-olds; OR, 2.44; 95% CI, 1.11 to 5.34 for those aged 80+ relative to those aged 65 to 69 years old), those who reported a greater influence of information about tamoxifen on decision-making (an increase in 1 SD increased the odds by 7.43; 95% CI, 4.36 to 12.65), and those whose physicians believed that the benefits of tamoxifen outweighed its risks (an increase in 1 SD increased the odds by 1.87; 95% CI, 1.34 to 2.62) were more likely to be prescribed tamoxifen.

CONCLUSION

These findings highlight the key role of communication in the care of older women with breast cancer and its ultimate influence on the receipt of therapy.

Use of cholesterol-lowering therapy by elderly adults after myocardial infarction

BACKGROUND

Use of cholesterol-lowering drugs reduces mortality and adverse cardiac events among people aged 65 to 75 years with coronary heart disease, but previous studies have shown that most patients have not received this treatment.

METHODS

We conducted a telephone survey during 1999 and 2000 of 815 Medicare beneficiaries aged 65 to 74 years hospitalized for an acute myocardial infarction in California, Florida, Massachusetts, New York, or Pennsylvania during 1994 and 1995. Outcome measures included use of cholesterol-lowering drugs, beliefs about the importance of lowering cholesterol levels, and knowledge of personal cholesterol levels, adjusting for demographic and clinical factors using logistic regression.

RESULTS

Among respondents, 59.4% reported taking a cholesterol-lowering drug, but most were not aware of potential adverse effects. In adjusted analyses, drug treatment was significantly more common among women, patients aged 65 to 69 years, and those who reported that a cardiologist was mainly responsible for their cholesterol management. Lowering cholesterol levels was viewed as "very important" by 77.2% of respondents, but significantly less often by men, older patients, and those with diabetes mellitus or congestive heart failure. Only 33.1% of respondents knew their cholesterol level, and this knowledge was significantly less common among black patients and those with diabetes mellitus or congestive heart failure.

CONCLUSIONS

Use of cholesterol-lowering drugs was much greater than in previous studies of elderly patients after myocardial infarction, demonstrating increased attention to secondary prevention. However, most patients were unaware of their cholesterol level or potential adverse effects of drug treatment, indicating that they may benefit from greater education about cholesterol testing and treatment.

Longitudinal profiles of health care providers

Provider profiling is the activity of collecting, comparing and reporting quality of care measures for individuals, groups, agencies and institutions that provide health care services. Univariate provider profiles, such as hospital-specific mortality rates, have been constructed using cross-sectional data based on posterior summaries or maximum likelihood estimates. As data continue to be collected over time, the construction and interpretation of longitudinal profiles of health care providers will become increasingly important. Longitudinal series can be used to improve the precision of estimates - a feature that is particularly important for providers who treat a small number of patients per year. We extend and apply hierarchical models to examine and classify provider performance over time using two examples, one in the area of cardiology and the other in mental health. Performance is evaluated using the squared Mahalanobis distance and posterior probabilities based on this distance. By comparing providers based on level and temporal trend simultaneously, conservative but comprehensive assessments of performance are possible. Furthermore, the longitudinal profiles developed are easily interpreted and flexible, making them of practical use to policy-makers.

Treatment decision making in early-stage breast cancer: should surgeons match patients' desired level of involvement?

PURPOSE

To describe desired and actual roles in treatment decision making among patients with early-stage breast cancer, identify how often patients' actual roles matched their desired roles, and examine whether matching of actual and desired roles was associated with type of treatment received and satisfaction.

PATIENTS AND METHODS

We surveyed 1,081 women (response, 70%) diagnosed with early-stage breast cancer in Massachusetts or Minnesota about their desired and actual roles in treatment decision making with their surgeon and used logistic regression to assess whether matching of actual to desired roles was associated with type of surgery and satisfaction.

RESULTS

Most patients (64%) desired a collaborative role in decision making, but only 33% reported actually having such a collaborative role when they discussed treatments with their surgeons. Overall, 49% of women reported an actual role that matched the desired role they reported, 25% had a less active role than desired, and 26% had a more active role than desired. In adjusted analyses, patients whose reported actual role matched their desired role were no more likely than others to undergo breast-conserving surgery (P >.2), but these women were more satisfied with their treatment choice (83.5% very satisfied; reference) than those whose role was less active than desired (72.9% very satisfied; P =.02) or more active than desired (72.2% very satisfied; P =.005).

CONCLUSION

Only approximately half of patients reported an actual role in decision making that matched the desired role they reported. These patients were more satisfied with their treatment choice than other patients, suggesting that women with early-stage breast cancer may benefit from surgeons' efforts to identify their preferences for participation in decisions and tailor the decision-making process to them.

Use of cholesterol-lowering therapy and related beliefs among middle-aged adults after myocardial infarction

OBJECTIVE

To assess use of cholesterol-lowering therapy and related beliefs among middle-aged adults after myocardial infarction.

DESIGN

Telephone survey and administrative data.

SETTING

National managed-care company.

PARTICIPANTS

Six hundred ninety-six adults age 30 to 64 surveyed in 1999, approximately 1 to 2 years after a myocardial infarction.

MEASUREMENTS

Use of cholesterol-lowering drugs, beliefs about the importance of lowering cholesterol, and knowledge of personal cholesterol level, adjusting for demographic and clinical factors with logistic regression.

MAIN RESULTS

Among respondents, 62.5% reported they were taking a cholesterol-lowering drug. In adjusted analyses, these drugs were used significantly less often by African-American patients and those with congestive heart failure or peripheral vascular disease, and more often by college graduates, patients with hypertension, and those who had seen a cardiologist since their myocardial infarction. Lowering cholesterol was viewed as "very important"; by 87.1% of patients, but significantly less often by smokers and more often by those who had undergone coronary angioplasty or bypass surgery. Only 42.5% of respondents knew their cholesterol level, and this knowledge was significantly less common among less-educated or less-affluent patients, African-American patients, and patients who smoked or had diabetes or peripheral vascular disease.

CONCLUSIONS

Although most patients recognized the importance of lowering cholesterol after myocardial infarction, several clinical and demographic subgroups were less likely to receive cholesterol-lowering therapy, and many patients were unaware of their cholesterol level. Health-care providers and managed-care plans can use these findings to promote cholesterol testing and treatment for patients with coronary heart disease who are most likely to benefit from these efforts.

The economic impacts of the tobacco settlement

Recent litigation against the major tobacco companies culminated in a master settlement agreement (MSA) under which the participating companies agreed to compensate most states for Medicaid expenses. Here the terms of the settlement are outlined and its economic implications analyzed using data from Massachusetts. The financial compensation to Massachusetts (and other states) under the MSA is substantial. However, this compensation is dwarfed by the value of the health impacts induced by the settlement. Specifically, Medicaid spending will fall, but only by a modest amount. More importantly, the value of health benefits ($65 billion through 2025 in 1999 dollars) from increased longevity is an order of magnitude greater than any other impacts or payments. The net efficiency implications of the settlement turn mainly on a comparison of the value of these health benefits relative to a valuation of the foregone pleasure of smoking. To the extent that the value of the health benefits is not offset by the value of the pleasure foregone, the economic impacts of the MSA will include a share of these health benefits.

The treating physician as active gatekeeper in the recruitment of research subjects

BACKGROUND

Institutional Review Boards vary in regard to the conditions imposed on investigators concerning contacting potential subjects to participate in health-services research studies.

OBJECTIVE

The impact of more active involvement of the treating physician was examined in the approval process for recruiting study subjects.

DESIGN

In recruiting subjects for a Massachusetts-based, multihospital (n = 17), health-services research study of treatment patterns for early stage breast cancer that required patient interviews, four hospitals stipulated that the treating surgeon provide written permission to the investigators to allow any contact with a potential study subject for the purpose of recruitment (active physician involvement group); the remaining 13 hospitals stipulated that the treating surgeon need only respond to the investigators if contact with a potential subject was forbidden (passive physician involvement group).

SUBJECTS

Of the 1401 potential subjects treated for early stage breast cancer, 697 were in the active physician involvement group and 704 were in the passive physician involvement group.

MEASURES

The percentages of patients for whom contact was allowed for recruitment purposes and who enrolled in the study were determined for the active physician involvement group and the passive physician involvement group, respectively. Logistic regression models were used to assess the independent effect of physician involvement on study enrollment.

RESULTS

Of the 697 patients in the active physician involvement group, contact was approved by the treating surgeon for 72% (n = 505), compared with 91% (n = 638) of the passive physician involvement group (P <0.001). After adjustment for a variety of patient, physician, and hospital-level variables, patients in the passive physician involvement group were found to be significantly more likely to be enrolled in the study (adjusted OR 2.61; 95% CI, 1.53-4.45). However, among those patients approved for investigator contact, there were no significant differences between patients who were enrolled and patients who were not enrolled in the study with regard to physician involvement in the recruitment process (adjusted OR 1.13; 95% CI, 0.70-1.81).

CONCLUSION

Our findings demonstrate that more stringent IRB requirements on health services researchers to verify permission from the treating physician to access patients for recruitment purposes adversely impact on the enrollment of patients even in noninterventional research studies. Current procedures for involving the treating physician as a gatekeeper in the recruitment of research subjects may limit access to patient participation in research studies from the perspectives of both researchers and potential subjects.

A multiple imputation strategy for incomplete longitudinal data

Longitudinal studies are commonly used to study processes of change. Because data are collected over time, missing data are pervasive in longitudinal studies, and complete ascertainment of all variables is rare. In this paper a new imputation strategy for completing longitudinal data sets is proposed. The proposed methodology makes use of shrinkage estimators for pooling information across geographic entities, and of model averaging for pooling predictions across different statistical models. Bayes factors are used to compute weights (probabilities) for a set of models considered to be reasonable for at least some of the units for which imputations must be produced, imputations are produced by draws from the predictive distributions of the missing data, and multiple imputations are used to better reflect selected sources of uncertainty in the imputation process. The imputation strategy is developed within the context of an application to completing incomplete longitudinal variables in the so-called Area Resource File. The proposed procedure is compared with several other imputation procedures in terms of inferences derived with the imputations, and the proposed methodology is demonstrated to provide valid estimates of model parameters when the completed data are analysed. Extensions to other missing data problems in longitudinal studies are straightforward so long as the missing data mechanism can be assumed to be ignorable.

Discussion of treatment options for early-stage breast cancer: effect of provider specialty on type of surgery and satisfaction

OBJECTIVES

To describe the physicians with whom breast cancer patients discuss treatment options and assess whether discussing surgical options with a medical oncologist is associated with type of surgery and satisfaction.

RESEARCH DESIGN

Medical record abstraction and survey.

SUBJECTS

Women with early-stage breast cancer numbering 2,426 in two states-Massachusetts, where the rate of breast-conserving surgery is high, and Minnesota, where it is lower.

MEASURES

Receipt of breast-conserving surgery and satisfaction.

RESULTS

Women in Massachusetts discussed breast cancer treatments with more physicians than women in Minnesota (mean 3.5 vs. 2.8; P <0.001) and more often discussed surgical options with a medical oncologist (52% vs. 28%; P <0.001). Using propensity score analyses, in Massachusetts, discussing surgical options with a medical oncologist was not related to type of surgery (adjusted difference in rate of breast-conserving surgery: 3.9%, 95% CI -3.6% to 11.5%) but was associated with greater satisfaction (adjusted difference: 8.1, 95% CI 2.0% to 14.2%). In Minnesota, discussing surgical options with a medical oncologist was associated with breast-conserving surgery (adjusted difference: 12.6%, 95% CI 5.6% to 19.7%) with no difference in satisfaction (adjusted difference: -1.5%, 95% CI -6.8% to 3.8%).

CONCLUSIONS

Outcomes associated with discussing surgical treatments with a medical oncologist vary with local care patterns. Where breast-conserving surgery is standard care, seeing a medical oncologist is not related to type of surgery, but is associated with greater satisfaction. Where it is not the standard, seeing a medical oncologist is associated with more breast-conserving surgery and equivalent satisfaction. These findings suggest that collaborative care may benefit women with respect to treatment selection or satisfaction.

Quality of ambulatory care after myocardial infarction among Medicare patients by type of insurance and region

PURPOSE

To evaluate use of effective cardiac medications and rehabilitation after myocardial infarction in the ambulatory setting in health maintenance organizations (HMOs) and fee-for-service care, and by region.

SUBJECTS AND METHODS

We surveyed elderly Medicare patients during 1996 and 1997 in California (n = 516), Florida (n = 304), and the Northeast (n = 220; Massachusetts, New York, and Pennsylvania) approximately 18 months after myocardial infarction. We assessed use of cardiac medications and rehabilitation for HMO (n = 520) and fee-for-service (n = 520) patients matched by age, sex, month of infarct, and region.

RESULTS

Across all regions, similar proportions of HMO and fee-for-service patients were using aspirin (72%, n = 374 vs. 74%, n = 387), beta-blockers (38%, n = 195 vs. 32%, n = 168), angiotensin-converting enzyme inhibitors (31%, n = 159 vs. 29%, n = 148), cholesterol-lowering agents (28%, n = 146 vs. 30%, n = 157), and calcium channel blockers (31%, n = 162 vs. 31%, n = 159; all P >0.07), except in California where more HMO patients received beta-blockers (36%, n = 93 vs. 26%, n = 66, P = 0.01). In adjusted analyses, use of these drugs did not differ significantly between HMO and fee-for-service patients. Substantial regional differences were evident in the use of beta-blockers (Northeast 46%, n = 102; Florida 34%, n = 102; California 31%, n = 159) and cholesterol-lowering agents (California 35%, n = 182; Florida 24%, n = 73; Northeast 22%, n = 48; each P <0.001). Fee-for-service patients were more likely than HMO patients to receive cardiac rehabilitation in unadjusted (32%, n = 167, vs. 22%, n = 141, P = 0.001) and adjusted analyses.

CONCLUSIONS

Both HMO and fee-for-service patients would likely benefit from greater use of beta-blockers and cholesterol-lowering agents. Professional fees for cardiac rehabilitation may promote increased use among fee-for-service patients. Future studies should assess the quality of ambulatory cardiac care in different types of HMOs and the reasons for geographic variations in cardiac drug use.

Impact of underuse, overuse, and discretionary use on geographic variation in the use of coronary angiography after acute myocardial infarction

BACKGROUND

Geographic variation in the use of medical procedures has been well documented. However, it is not known whether this variation is due to differences in use when procedures are indicated, discretionary, or contraindicated.

OBJECTIVES

To examine whether use of coronary angiography after acute myocardial infarction (AMI) according to appropriateness criteria varied across geographic regions and whether underuse, overuse, or discretionary use accounted for variation in overall use.

DESIGN

Retrospective cohort study using data from the Cooperative Cardiovascular Project.

SETTING

Ninety-five hospital referral regions.

PATIENTS

There were 44,294 Medicare patients hospitalized with AMI during 1994 or 1995, classified according to appropriateness for angiography.

MAIN OUTCOME MEASURE

Variation in use of angiography, as measured by the difference between high and low rates of use across regions.

RESULTS

Across regions, variation in the use of angiography was similar for indications judged necessary; appropriate, but not necessary; or uncertain. Variation was lowest for indications judged unsuitable (difference between high rate and low rate across regions = 16.3%; 95% CI = 12.6%; 20.6%). The primary cause of variation in the overall rate of angiography was due to use for indications judged appropriate, but not necessary or uncertain. When variation associated with these indications was accounted for, the difference between the resulting high and low overall rates was 10.8% (9.4%, 12.4%). In contrast, variation in the overall rate remained high when underuse in necessary situations or overuse in unsuitable situations was accounted for.

CONCLUSIONS

Across regions, practice was more similar for patients categorized unsuitable for angiography than for patients with other indications. Variation in overall use of angiography appeared to be driven by utilization for discretionary indications rather than by underuse or overuse. If equivalent rates across geographic areas are judged desirable, then greater effort must be directed toward defining care for patients with discretionary indications.

Validating recommendations for coronary angiography following acute myocardial infarction in the elderly: a matched analysis using propensity scores

We determined whether adherence to recommendations for coronary angiography more than 12 h after symptom onset but prior to hospital discharge after acute myocardial infarction (AMI) resulted in better survival. Using propensity scores, we created a matched retrospective sample of 19,568 Medicare patients hospitalized with AMI during 1994-1995 in the United States. Twenty-nine percent, 36%, and 34% of patients were judged necessary, appropriate, or uncertain, respectively, for angiography while 60% of those judged necessary received the procedure during the hospitalization. The 3-year survival benefit was largest for patients rated necessary [mean survival difference (95% CI): 17.6% (15.1, 20.1)] and smallest for those rated uncertain [8.8% (6.8, 10.7)]. Angiography recommendations appear to select patients who are likely to benefit from the procedure and the consequent interventions. Because of the magnitude of the benefit and of the number of patients involved, steps should be taken to replicate these findings.

Appropriateness of coronary angiography after myocardial infarction among Medicare beneficiaries. Managed care versus fee for service

BACKGROUND

Previous studies have documented that cardiac procedures are performed less frequently in patients enrolled in managed-care plans than in those with fee-for-service coverage. However, it is not known whether this difference is due to less frequent use of cardiac procedures when they are indicated or to less frequent use when they are not indicated.

METHODS

We compared the use of coronary angiography after acute myocardial infarction among Medicare beneficiaries who had traditional fee-for-service coverage with the use among Medicare beneficiaries enrolled in managed-care plans. The analysis was adjusted for differences in demographic and clinical characteristics of the patients and for characteristics of the hospitals to which they were admitted. We studied more than 50,000 beneficiaries in seven states and evaluated their care according to guidelines proposed by the American College of Cardiology and the American Heart Association (ACC-AHA).

RESULTS

Among the 44 percent of patients in both groups who had ACC-AHA class I indications (those for which angiography is useful and effective), more fee-for-service beneficiaries than managed-care enrollees underwent angiography (46 percent vs. 37 percent, P<0.001). The rate of angiography was very low among patients with class I indications who were admitted to hospitals without angiography facilities (31 percent in the fee-for-service group and 15 percent in the managed-care group, P<0.001). Among patients with class III indications (those for which angiography is not effective), the rate of use was low in both groups (approximately 13 percent).

CONCLUSIONS

In situations in which angiography is thought to be useful, it is used less often among Medicare beneficiaries enrolled in managed-care plans than among those with fee-for-service coverage. Moreover, rates of use among patients with class I indications are fairly low in both groups, suggesting that there is room for improving the care of elderly patients with myocardial infarction, especially those admitted to hospitals without angiography facilities.

Understanding variability in physician ratings of the appropriateness of coronary angiography after acute myocardial infarction

We examined variability in ratings of the appropriateness of coronary angiography for 890 clinical scenarios (indications) after an acute myocardial infarction (AMI) from a nine-member multispecialty panel as a function of panel characteristics and the attributes of the clinical indications. We documented a substantial degree of reliability in the ratings. However, key differences among the experts in terms of both their overall propensity to score high and their beliefs regarding the impact of clinical factors on appropriateness were identified. Age, cardiac complications, post-AMI angina, and noninvasive test results were the clinical factors most strongly related to appropriateness ratings for coronary angiography. Further research on the effectiveness of coronary angiography in older patients and in patients with shock, pulmonary edema, and silent ischemia is needed to improve our knowledge about the appropriateness of this procedure in these patients.

Applying Bayesian ideas to the development of medical guidelines

Measurements of the quality of health care, in particular the underuse and overuse of medical therapies and diagnostic tests, often involve employment of medical practice guidelines to assess the appropriateness of treatments. This paper presents a case study of a Bayesian analysis for the development of medical guidelines based on expert opinion, using ordinal categorical rater data. We develop guidelines for the use of coronary angiography following an acute myocardial infarction (AMI) for 890 clinical indications using statistical models fit to appropriateness ratings obtained from a nine-member expert panel. The main foci of our analyses were on the estimation of an appropriateness score for each of the clinical indications, an associated measure of precision, and functions of the underlying score. We considered two classes of models that assume the ratings are either in the form of grouped normal data or are ungrouped variables arising from a normal distribution, while permitting rater effects and indication heterogeneity in both. We estimated models using Markov chain Monte Carlo methods and constructed indices quantifying appropriateness based on posterior probabilities of selected model parameters. We compared our model-based approach to the standard approach currently employed in medical guideline development and found that the standard approach correctly identified 99 per cent of the appropriate indications while overestimating appropriateness 18 per cent of the time compared to our model-based approach.