Health Care Financing Administration Data Research Articles

Interindividual Variation in Nucleotide Excision Repair Genes and Risk of Endometrial Cancer

Exposure to estrogens is a likely cause of endometrial cancer, but the means by which estrogens exert this effect are not entirely clear. One hypothesis is that certain estrogen metabolites bind to the DNA, forming bulky adducts that damage the DNA and initiate carcinogenesis. A woman's reduced capacity to repair such damage may increase her risk of endometrial cancer. We conducted a population-based case-control study in western Washington State to address the role of variation in nucleotide excision repair genes on the risk of endometrial cancer. Case women (n = 371), ages 50 to 69 years, were diagnosed with invasive endometrial cancer between 1994 and 1999. Control women (n = 420) were selected using random-digit dialing (ages 50-65 years) and by random selection from Health Care Financing Administration data files (ages 66-69 years). Genotyping assays were done for ERCC1, ERCC2 (XPD), ERCC4 (XPF), ERCC5 (XPG), XPA, and XPC. No appreciable differences between cases and controls were observed in the genotype distributions of ERCC1 (c8092a and c19007t), ERCC2 (D312N, K751Q, and c22541a), ERCC4 (R415Q and t30028c), or ERCC5 (D1104H). Carriage of at least one variant allele for XPA G23A was associated with decreased risk of endometrial cancer [odds ratio (OR), 0.70; 95% confidence interval (95% CI), 0.53-0.93]. Carriage of at least one XPC A499V variant allele was associated with a modest decrease in risk (OR, 0.79; 95% CI, 0.59-1.05). Women with variant alleles at both XPC A499V and K939Q had 58% of the risk of women with no XPC variant alleles (OR, 0.58; 95% CI, 0.35-0.96). Our data suggest that interindividual variation in XPA and XPC influences a woman's risk of endometrial cancer.

Comparison of mortality and intermediate outcomes between medicare dialysis patients in HMO and fee for service

End-stage renal disease (ESRD) is the only disease entitlement for Medicare; therefore, most patients with ESRD have Medicare coverage. Patients with ESRD are prohibited by law from enrolling in health maintenance organizations (HMOs), the only group prohibited within Medicare. However, they may remain in an HMO if they enrolled in such a plan before their kidneys failed. Thus, it is possible to compare patients with ESRD in HMOs with those in fee-for-service (FFS) plans. To determine whether mortality, transplantation rates, and intermediate outcomes differed between Medicare ESRD beneficiaries enrolled in HMO versus FFS providers, a retrospective cohort analysis was performed of patients with ESRD from three Health Care Financing Administration data sets containing administrative and outcome information for Medicare ESRD beneficiaries from 1990 to 1998. On December 31, 1998, a total of 278,510 prevalent patients with ESRD were in FFS plans, and 18,332 patients were in HMOs. HMO patients were older and more likely to be white and male and have diabetes mellitus and comorbid cardiovascular conditions than FFS patients. Unadjusted 2-year survival rates were 48.4% and 49.3% for FFS and HMO patients, respectively. In a multivariate model, HMO status had no significant effect on mortality, which was greater with older age, male sex, and white race. In 1998, unadjusted renal transplantation rates were 23.5% and 15.5% for FFS and HMO patients, respectively; age adjustment abrogated the apparent difference. For FFS and HMO patients, adequate hemodialysis was delivered to 72% and 82%, and 56% and 62% had hematocrits greater than the benchmark, respectively. There was no statistical difference in these outcomes in multivariate comparison. In conclusion, care by HMO for patients with an expensive chronic illness can achieve outcomes similar to those for FFS patients. Claims of poorer care and worse outcomes for patients with ESRD enrolled onto an HMO, an argument used to justify continued prohibition against widespread participation by patients with ESRD, are not supported. © 2002 by the National Kidney Foundation, Inc.

Calcium channel blockers and cancer

PURPOSE: We sought to explore the relation that has been previously reported between calcium channel blockers and an increased risk of cancer. SUBJECTS AND METHODS: We followed 3,511 participants, age 65 years or older, in the Duke Established Populations for Epidemiologic Studies of the Elderly for up to 10 years. Information about use of medications was obtained at baseline and 3 and 6 years later. Information about hospitalization for cancer, or death from cancer, was obtained from Health Care Financing Administration data and death certificates. RESULTS: Of the 133 users of calcium channel blockers, 16 (12%) developed cancer, compared with 548 (16%) of 3,378 nonusers (hazard ratio = 0.9; 95% confidence interval, 0.5 to 1.5). Adjusting for baseline and time-dependent covariates, such as race, diabetes, or blood pressure, for dose or class of calcium channel blockers, or for length of follow-up, had no effect. CONCLUSIONS: Use of calcium channel blockers does not appear to be related to cancer risk. Earlier reports showing such a relation may have been the result of chance.

Incidence of small-bowel obstruction and adhesiolysis after open colorectal and general surgery.

The study contained herein was undertaken to establish the incidence of small-bowel obstruction, adhesiolysis for obstruction, and additional abdominal surgery after open colorectal and general surgery. A retrospective cohort study was performed using patient-specific Health Care Financing Administration data to evaluate a random 5 percent sample of all Medicare patients who underwent surgery in 1993. Of these, 18,912 patients had an index abdominal procedure. Two-year follow-up data documented outcomes of hospitalizations with obstruction, adhesiolysis for obstruction, and/or additional open colorectal or general surgery. Within two years of incision, excision, and anastomosis of intestine (International Classification of Dis eases (ICD)-9 code 45), 14.3 percent of patients had obstructions, 2.6 percent required adhesiolysis for obstructions, and 12.9 percent underwent additional open colorectal or general surgery. After other operations of intestine (ICD code 46), 17 percent of patients had obstructions, 3.1 percent required adhesiolysis for obstructions, and 20.2 percent underwent additional open colorectal or general surgery. After operations of rectum, rectosigmoid, and perirectal tissue (ICD code 48), 15.3 percent of patients had obstructions, 5.1 percent required adhesiolysis for obstructions, and 16.4 percent underwent additional open colorectal or general surgery. After other operations on the abdominal region (ICD code 54), 12.4 percent of patients had obstructions, 2.3 percent required adhesiolysis for obstructions, and 8.8 percent underwent additional open colorectal or general surgery. In this retrospective study of Medicare patients, we learned that bowel obstruction, adhesiolysis for obstructions, and additional abdominal surgery occurred more often after abdominal surgery than was previously published.

Mammography screening among California Medicare beneficiaries: 1993–1994

Background: Over half of all breast cancer deaths occur among women 65 years of age or older. However, mammography screening decreases with increasing age, despite better survival rates for tumors detected early. Methods: Health Care Financing Administration data from 1993 and 1994, and 1990 United States Census data were used to assess the impact of race, age, Medicaid coverage, and community-level socioeconomic indices on mammography screening for over 800,000 California Medicare beneficiaries. Results: Women who were African American, older, or had Medicaid coverage were significantly less likely to have a biennial mammogram than their counterparts. Women living in areas with fewer college educated residents, with a higher proportion of Mexican or Asian residents had lower use of mammography. However, African-American and Caucasian women with Medicaid coverage had equally low mammography rates (AOR = 1.01, 95%CI .97–1.04), while African-American women with and without Medicaid had similarly low mammography rates (AOR = .96, 95%CI .92–1.01). Conclusions: Despite dual coverage, Medicare beneficiaries enrolled in Medicaid had few mammograms. African-American Medicare beneficiaries, with and without Medicaid, had low mammography rates. Intervention efforts should be targeted toward these women.

Geographic distribution of constipation in the United States.

Despite its frequent occurrence, the etiology of constipation has remained poorly understood. The influence of widely accepted risk factors such as inadequate dietary fiber intake, immobility, insufficient fluid intake, and poor muscle tone is unclear. This study examined the geographic distribution of constipation among Medicare beneficiaries to identify potential environmental risk factors. All Medicare beneficiaries with a diagnosis of constipation were extracted from the total Health Care Financing Administration data file of 1987 and stratified by sex, race, and state of residence. The population of each state by sex, race, and age >65 yr served as the denominator to calculate sex- and race-specific morbidity rates. A distinct geographic distribution was observed. When stratified by individual states, hospital discharges for constipation were more common in rural as compared with urban states. Constipation also appeared to be more common in northern and in poorer states. The distinct geographic pattern of constipation suggests the influence of three global environmental factors: rural living, colder temperature, and lower socioeconomic status.

Outcome Implications for the Timing of Bilateral Total Knee Arthroplasties

Health Care Financing Administration data from 1985 to 1990 revealed 339,152 total knee arthroplasties of which 62,730 (18.6%) were bilateral procedures (simultaneous 112,922; staged 6 weeks, 4354; staged 3 months, 4524; staged 6 months, 9829; and staged 1 year 31,401). Medicare beneficiaries undergoing bilateral procedures were an average of 73 years of age; demographics revealed that among the various simultaneous and staged groups 57% to 69% were females, 90% were white, 85% to 90% had a diagnosis of osteoarthritis, and 30% to 40% were performed in rural hospitals. Between 1985 and 1990, surgical and vascular complications ranged from 2.4% to 4% and 4.1% to 6.8%, respectively, for all types of bilateral staged and simultaneous total knee arthroplasties. All differences were statistically significant. After controlling statistically for demographic variables and diagnoses, a surrogate for case mix, it was found that individuals electing simultaneous bilateral arthroplasties experienced twice the number of intensive care days than those choosing staged procedures. Days in the intensive care unit were double when done simultaneously instead of staged (0.48 versus 0.21). Nosocomial infections were similar within groups (10% versus 13%); however, wound infections were nearly half when done simultaneously (0.5% versus 1%) versus in a staged fashion. Length of stay and cost were much less for the simultaneous procedure group who were sicker as measured by the number of diagnoses. Mortality at 30 days was highest for the simultaneous procedure group (.99%) versus staged 3 or 6 months (0.30%); however, by 2 years it was close to 4% for all groups. Staging the procedure 3 to 6 months seems to offer the fewest disadvantages, is only slightly more expensive, and has the lowest mortality rate.

Some of the Small Print on Managed CareProposals for End-Stage Renal Disease

In this article we discuss selected issues related to Medicare's end-stage renal disease (ESRD) managed care demonstration project and Congressional proposals to remove the barrier to ESRD patients enrolling in Medicare managed care plans. We discuss financial incentives to keep patients healthy; beneficiary obligations under fee-for-service and managed care; risk selection by beneficiaries among plans; and the baseline determination of a capitation rate. The ESRD demonstration offers the opportunity to evaluate the consequences of making Medicare managed care options available to a high cost and clinically vulnerable population. Careful evaluation is necessary to ensure that ESRD managed care options are structured to be beneficial to taxpayers, caregivers, and, most importantly, the beneficiaries choosing these options. Certainly, the potential exists for managed care to benefit patients by changing the fractured system in which each provider only has an incentive to worry about its own costs. However, the possible unintended consequences highlighted in this article strongly suggest that the evaluation of the demonstration project be undertaken before managed care options are made widely available outside the demonstration sites. Problems of a more technical nature, such as how to best use available Health Care Financing Administration data in the rate-setting process, are likely to be overcome, but the time and effort necessary to resolve them should not be underestimated.

Factors that enhance continued trauma center participation in trauma systems.

To examine hospital, trauma system, and reimbursement factors that offset the financial burdens of trauma care delivery and to assess how proposed Medicaid and Medicare budget cuts may affect the ability of hospitals to alleviate financial pressures related to trauma care delivery. In-depth interviews and data collection for trauma centers in 12 metropolitan areas with populations of 1 million or more. Seventy trauma centers in these large urban communities that indicated a continuing commitment to providing trauma services for the foreseeable future. Hospital, trauma system, and reimbursement characteristics that distinguish hospitals that are better able to alleviate the financial burdens of indigent trauma care and a financial analysis that assesses payment adequacy for different payers and overall financial outcomes. Data from a variety of sources were obtained to measure the factors that affect the operation and financing of trauma centers: published and unpublished hospital data from the American Hospital Association; trauma center level, length of operation, and the availability of alternative centers from a recently published study; Health Care Financing Administration data on Medicare and Medicaid program characteristics; automobile insurance requirements; and patient discharge data. Most data are reflective of 1992. Public hospitals, teaching hospitals, and institutions receiving supplemental indigent care payments appear to be best able to mitigate the financial burdens of uncompensated trauma care, especially those with moderate indigent care loads. A detailed financial analysis found that private hospitals with trauma centers were near break-even in 1992 for trauma care delivery and public hospitals experienced financial losses. Proposals to reduce Medicaid and Medicare would create substantial reductions in hospital payments for hospital-wide patient care and trauma patients specifically. Proposed Medicaid and Medicare payment cuts are likely to eliminate the delicate financial balance that many urban hospitals have achieved in providing trauma care. The erosion in funding from public programs may portend a new wave of trauma center closures as hospitals seek to deal with reduced reimbursement by eliminating unprofitable services.

Medicare-certified home health services: national and regional supply in the 1980s.

The number of Medicare-certified home health agencies nearly doubled from 1980 to 1990. Using Health Care Financing Administration data, this study documented national and regional patterns of entry and exist by Medicare home health providers from 1980 to 1990. Nationally, agency origination rates accelerated during the early 1980s and then dropped abruptly in the second half of the decade. The proprietary sector, accounting for approximately 42% of agencies in existence during the period of the study, exhibited the greatest volatility. Regional differences are also evident. Both expansion and contraction in Medicare home health services appear to be a response to the incentives of legislation implemented during this period.

The Use of Health Care Financing Administration Data for the Development of a Quality Improvement Project on the Treatment of Anemia

The Health Care Financing Administration maintains a wide array of data systems that are essential to the functioning of the Medicare program. These data, collected and maintained for the purposes of ensuring entitlements and payment for services, also can be used to monitor programmatic changes and to define potential problem areas. The end-stage renal disease (ESRD) Program Management and Medical Information System (PMMIS) is a subset of the larger Medicare statistical system. It is a historic record of all Medicare ESRD beneficiaries dating back to 1978. Basic Medicare enrollment information on ESRD beneficiaries is enhanced with the addition of information on the cause of renal failure, type of dialysis therapy, transplantation history, and cause of death. The ESRD PMMIS has been put to a number of uses in the past decade or so, ranging from basic descriptive epidemiology to analyses of mortality rates to assessments of programmatic issues such as the composite rate and dialyzer reuse. Because of the limited clinical detail in the PMMIS, there are many specific questions that cannot be adequately addressed. With approval of the Food and Drug Administration and Medicare coverage of erythropoietin, a erythropoietin monitoring system was developed to assess utilization trends of this anemia control drug. Within a few months it became evident that dosing levels for erythropoietin were much lower than expected from the clinical trials. Following a change in the payment method from a fixed amount to one based on dose level, dosing has increased markedly. However, hematocrit levels still remain below optimal levels. This lack of hematocrit response has led the Health Care Financing Administration, in concert with the renal community, to target anemia control as a potential health care quality improvement project. This paper presents an example of the type of data presentation that can be derived from the current PMMIS. The Health Standards and Quality Bureau has made a commitment to a program of continuous quality improvement. Part of this process is the provision of descriptive data that can be the starting point for an iterative approach to quality improvement.

Validation of the Cause of Renal Failure of Patients in the Medicare End-Stage Renal Disease Program

Studies addressing the epidemiological issues of end-stage renal disease (ESRD) and the implications for health resource allocations are valuable, especially because the Medicare ESRD program is expected to continue growing during the next decade. Studies on trends in the causes of renal failure have benefited from Medicare's extensive data files on ESRD patients. However, no studies have been published that validate the cause of renal failure field in the Medicare files. The primary disease causing renal failure for over 10,000 New York State patients in the Medicare ESRD program was compared with their hospital discharge diagnoses. Of these patients, 8,730 (83%) had a known cause of renal failure in the Health Care Financing Administration (HCFA) data files. Eighty-nine percent of these patients' primary cause of renal failure was matched with the same major hospital diagnostic code. Patients with diabetes and glomerulonephritis had the highest overall match rates (96% to 97%). Patients with polycystic kidney disease and causes of renal failure other than the four major causes had the lowest match rates (75% to 76%), but these match rates increased to 84% to 87% for patients hospitalized more than five times. Some differences in match rate by age and race were found. These findings suggest that HCFA data on the causes of renal failure of ESRD patients are reasonably accurate and can be used successfully to study a variety of issues related to the diseases leading to chronic renal failure.

The Fungible Value of Government and Employer Provided In-Kind Health Benefits 1

In this paper we discuss the issue of whether and how to value government and employer provided in-kind medical benefits in determining who is poor. Consideration of this issue leads to the suggestion that the fungible portion of medical care transfers should be included in the income of the poor as they constitute a significant percentage of their resources available for the purchase of goods and services. Fungible medical care transfers are defined as third party payments for medical benefits except those for hospital and nursing home care, on the premise that the latter are for nonroutine medical services that do not increase an individual's well-being. Estimates of the fungible portion of government and private in-kind medical benefits are developed from Health Care Financing Administration data that indicate that of the two, the former is less fungible.

Understanding the Recent Growth in Medicare Physician Expenditures

This study employs several large Health Care Financing Administration data sets for 1983 and 1985 to examine the recent growth in Medicare physician services. The study concludes that the recent growth (approximately 15% in real terms between 1983 and 1985) has been more rapid in areas with higher incomes per capita and suggests that this may be related to faster adoption and diffusion of new medical technologies in these areas. The volume of physician services had grown considerably faster for those specialists who utilize newer procedures and technologies than for those who do not. The study also provides evidence that the sharp increase in assignment rates in recent years because of the introduction of the physician participation program also contributed to the growth in physician services during this period. Medicare's prospective payment system, which controlled hospital payments and encouraged hospitals to become more efficient, had at most a small positive impact on the growth in Part B spending. Finally, the freeze on physician's fees did not seem to have had a major impact on the volume of physician services.

Understanding the recent growth in Medicare physician expenditures

This study employs several large Health Care Financing Administration data sets for 1983 and 1985 to examine the recent growth in Medicare physician services. The study concludes that the recent growth (approximately 15% in real terms between 1983 and 1985) has been more rapid in areas with higher incomes per capita and suggests that this may be related to faster adoption and diffusion of new medical technologies in these areas. The volume of physician services had grown considerably faster for those specialists who utilize newer procedures and technologies than for those who do not. The study also provides evidence that the sharp increase in assignment rates in recent years because of the introduction of the physician participation program also contributed to the growth in physician services during this period. Medicare's prospective payment system, which controlled hospital payments and encouraged hospitals to become more efficient, had at most a small positive impact on the growth in Part B spending. Finally, the freeze on physician's fees did not seem to have had a major impact on the volume of physician services.

Sex-specific and race-specific hip fracture rates.

Sex-, race- and age-specific hip fracture rates were determined using Health Care Financing Administration data for Medicare-reimbursed hip fracture hospitalizations from 1980 to 1982. Rates were highest in White women, lowest in Black men, and intermediate in White men and Black women. Proportions of hip fracture patients dying during hospitalization and those discharged to nursing homes, respectively, were: White men (10.5%; 49%); Black men (9.3%; 32%); White women (5.0%; 54%); and Black women (8.2%; 30%).

Hospital Leaders' Opinions of the HCFA Mortality Data

The release of hospital-specific mortality data by the Health Care Financing Administration has stirred controversy about the adequacy of current case-mix adjustment models and about the wisdom in general of public release of outcome data. We surveyed a national sample of hospitals, stratified by measured mortality rate, in the 1987 Health Care Financing Administration data release to determine the reactions of hospital leaders to the data and to learn if hospitals with high mortality had different attitudes from those of hospitals with average or low mortality as measured by the Health Care Financing Administration. Seventy-eight percent (N = 195) responded. All hospitals, regardless of mortality rate, shared an extremely negative view of the accuracy, usefulness, and interpretability of the Health Care Financing Administration's mortality data. The lowest possible rating (poor) was given by 70% of the respondents on the question of usefulness of the data to the hospital, by 54% on accuracy of the data, and by 85% on usefulness of the data to consumers. Only 31% of the respondents said that they had used the data at all for internal purposes and 20% reported that the data release had caused problems for the hospital. Hospitals in the high-mortality group were more likely than others to report both use of the data and problems from its release. Publication of outcome data to encourage quality improvement may face severe and pervasive barriers in the attitudes and reactions of hospital leaders who are potential clients for such data.