The quality of health data obtained in a new survey of elderly Americans: a validation study of the proposed Medicare Beneficiary Health Status Registry (MBHSR)

Predictors of self-report of heart failure in a population-based survey of older adults

BACKGROUND

Little research has been conducted on the predictors of self-report or patient awareness of heart failure (HF) in a population-based survey. The objective of this study was to (1) test the agreement between Medicare administrative and Health and Retirement Study (HRS) survey data and (2) determine predictors associated with self-report of HF, using a validated Medicare claims algorithm as the reference standard. We hypothesized that those who self-reported HF were more likely to have a higher number of HF-related claims.

METHODS AND RESULTS

Secondary data analysis was conducted using the 2004 wave of the HRS linked to 2002 to 2004 Medicare claims (n=5573 respondents aged ≥ 67 years). Concordance between self-report of HF in the HRS and Medicare claims was calculated. Logistic regression was performed to identify predictors associated with self-report HF. HF prevalence by self-report was 4.6%. Self-report of HF and claims agreement was 87% (κ=0.34). The presence of >1 HF inpatient claims was associated with greater odds of self-report (odds ratio [OR], 1.92; 95% CI, 1.23-3.00). Greater odds of self-reporting HF was also associated with ≥ 4 HF claims (OR, 2.74; 95% CI, 1.36-5.52). Blacks (OR, 0.28; 95% CI, 0.14-0.55) and Hispanics (OR, 0.30; 95% CI, 0.11-0.83) were less likely to self-report HF compared with whites in the final model.

CONCLUSIONS

Self-report of HF is an insensitive method for accurately identifying HF cases, especially in those with less-severe disease and who are nonwhite. There may be limited awareness of HF among older minority patients despite having clinical encounters during which HF is coded as a diagnosis.

Social Status, Risky Health Behaviors, and Diabetes in Middle-Aged and Older Adults

OBJECTIVE

This article investigates: (a) how social status influences diabetes prevalence and incidence; (b) how risky health behaviors contribute to the prediction of incident diabetes; (c) if the effects of health behaviors mediate the effects of social status on incident diabetes; and (d) if these effects differ in midlife and older age.

METHODS

We examined nationally representative data from the 1992/1993-1998 panels of the Health and Retirement Study for middle-aged and older adults using logistic regression analyses.

RESULT

The odds of prevalent diabetes were higher for people of older age, men, Black adults, and Latino adults. Higher early-life social status (e.g., parental schooling) and achieved social status (e.g., respondent schooling, economic resources) reduced the odds in both age groups. We observed similar patterns for incident diabetes in midlife but not in older age. Risky health behaviors--particularly obesity--increased the odds of incident diabetes in both age groups independent of social status. The increased odds of incident diabetes in midlife persisted for Black and Latino adults net of other social status factors.

DISCUSSION

Risky health behaviors are key predictors of incident diabetes in both age groups. Economic resources also play an important protective role in incident diabetes in midlife but not in older age.

Specificity and sensitivity of claims-based algorithms for identifying members of Medicare+Choice health plans that have chronic medical conditions

OBJECTIVE

To examine the effects of varying diagnostic and pharmaceutical criteria on the performance of claims-based algorithms for identifying beneficiaries with hypertension, heart failure, chronic lung disease, arthritis, glaucoma, and diabetes.

STUDY SETTING

Secondary 1999-2000 data from two Medicare+Choice health plans.

STUDY DESIGN

Retrospective analysis of algorithm specificity and sensitivity.

DATA COLLECTION

Physician, facility, and pharmacy claims data were extracted from electronic records for a sample of 3,633 continuously enrolled beneficiaries who responded to an independent survey that included questions about chronic diseases.

PRINCIPAL FINDINGS

Compared to an algorithm that required a single medical claim in a one-year period that listed the diagnosis, either requiring that the diagnosis be listed on two separate claims or that the diagnosis to be listed on one claim for a face-to-face encounter with a health care provider significantly increased specificity for the conditions studied by 0.03 to 0.11. Specificity of algorithms was significantly improved by 0.03 to 0.17 when both a medical claim with a diagnosis and a pharmacy claim for a medication commonly used to treat the condition were required. Sensitivity improved significantly by 0.01 to 0.20 when the algorithm relied on a medical claim with a diagnosis or a pharmacy claim, and by 0.05 to 0.17 when two years rather than one year of claims data were analyzed. Algorithms that had specificity more than 0.95 were found for all six conditions. Sensitivity above 0.90 was not achieved all conditions.

CONCLUSIONS

Varying claims criteria improved the performance of case-finding algorithms for six chronic conditions. Highly specific, and sometimes sensitive, algorithms for identifying members of health plans with several chronic conditions can be developed using claims data.

Review of the performance of methods to identify diabetes cases among vital statistics, administrative, and survey data

PURPOSE

The ability to identify prevalent cases of diagnosed diabetes is crucial to monitoring preventative care practices and health outcomes among persons with diagnosed diabetes.

METHODS

We conducted a comprehensive literature review to assess and summarize the validity of various strategies for identifying individuals with diagnosed diabetes and to examine the factors influencing the validity of these strategies.

RESULTS

We found that studies using either administrative data or survey data were both adequately sensitive (i.e., identified the majority of cases of diagnosed diabetes) and highly specific (i.e., did not identify the individuals as having diabetes if they did not). In contrast, studies based on cause-of-death data from death certificates were not sensitive, failing to identify about 60% of decedents with diabetes and in most of these studies, researchers did not report specificity or positive predictive value.

CONCLUSIONS

Surveillance is critical for tracking trends in diabetes and targeting diabetes prevention efforts. Several approaches can provide valuable data, although each has limitations. By understanding the limitations of the data, investigators will be able to estimate diabetes prevalence and improve surveillance of diabetes in the population.

A dynamic Markov model for forecasting diabetes prevalence in the United States through 2050

This study develops forecasts of the number of people with diagnosed diabetes and diagnosed diabetes prevalence in the United States through the year 2050. A Markov modeling framework is used to generate forecasts by age, race and ethnicity, and sex. The model forecasts the number of individuals in each of three states (diagnosed with diabetes, not diagnosed with diabetes, and death) in each year using inputs of estimated diagnosed diabetes prevalence and incidence; the relative risk of mortality from diabetes compared with no diabetes; and U.S. Census Bureau estimates of current population, live births, net migration, and the mortality rate of the general population. The projected number of people with diagnosed diabetes rises from 12.0 million in 2000 to 39.0 million in 2050, implying an increase in diagnosed diabetes prevalence from 4.4% in 2000 to 9.7% in 2050.

Regional differences in functional status among the aged

This study investigated regional differences in functional status among aged Medicare beneficiaries in the United States, and the degree to which population risk factors and certain geographic/environmental attributes of communities accounted for the regional differences. Four years of the Medicare Current Beneficiary Survey (1992-1995) were pooled together yielding 37,150 person-year observations of functional status for a sample of aged Medicare beneficiaries residing in the community or nursing homes. Multinomial logit models, estimated on a four-category functional status scale, produced strong empirical evidence of substantial regional differences in the prevalence of functional independence, functional limitations, IADL limitations, and ADL limitations, that could not be attributed to regional population composition, socio-demographic factors, lifestyle characteristics, and chronic medical conditions. Although such population risk factors accounted for much of the regional variations in functional status among older men, the notably higher prevalence of IADL and ADL limitations among older women residing in the Deep South could not be similarly attributed to such risk factors. Rather, the empirical results suggest that a significant portion of the harmful effects associated with residence in the Deep South among older women may be attributed to a higher prevalence of residence in counties characterized by lower population density and/or higher poverty concentration.

Foot care practices, services and perceptions of risk among medicare beneficiaries with diabetes at high and low risk for future foot complications

A cohort of Medicare beneficiaries with diabetes was identified from inpatient and outpatient claims data and their risk for foot complications was estimated based on claims reflecting services for recent foot problems. A telephone survey of a random sample from this cohort was conducted to assess their foot care practices, barriers, and perceptions of risk. Eight percent of respondents reported a history of foot ulcers and 7% a history of lower extremity amputation. Based on claims data, 30% of respondents were at high risk for future foot complications. Compared to those at low risk, those at high risk were more likely to report having an annual foot exam, using protective footwear, and perceiving themselves to be high risk for future foot complications. However, 50% of those with claims indicating a high risk perceived themselves to be at low risk for future foot complications. Overall, 20% of respondents seldom checked their feet daily for sores or irritations. Among this group, 60% felt that it was unimportant and 9% reported they were limited by poor vision or physical problems. Our findings suggest that strategies are needed to improve the delivery of preventive foot care services to older persons with diabetes. Additionally, emphasis is needed to help individuals understand their risk and seek and perform appropriate preventive foot care.

Disability outcomes of older Medicare HMO enrollees and fee-for-service Medicare beneficiaries

OBJECTIVES

To investigate whether older Medicare beneficiaries enrolled in Medicare risk health maintenance organizations (HMOs) have different rates of disablement than fee-for-service (FFS) beneficiaries.

DESIGN

Secondary analysis of annual functional status transitions using the Medicare Current Beneficiary Survey, 1991 to 1996.

SETTING

Telephone interviews.

PARTICIPANTS

Forty-four thousand seven hundred and sixty-five person-years of annual functional status transitions for noninstitutionalized older Medicare beneficiaries who were either risk HMO enrollees or FFS beneficiaries with or without private supplementary insurance.

MEASUREMENTS

Five multinomial logit models were estimated as single-state transition models, with five functional states, death, and censored as outcomes. The probability of being in a certain functional state the following year was specified as a function of individual risk factors and HMO versus FFS supplementary insurance status.

RESULTS

Among functionally independent beneficiaries, the odds of becoming disabled in activities of daily living (ADLs) within a year were lower among FFS individuals with supplementary insurance (odds ratios (OR) = 0.67, P <.01) and HMO enrollees (OR = 0.58, P <.01). Among older people who were functionally impaired, neither HMO enrollment nor private supplementary insurance affected the risk of further functional decline or functional improvement. Supplementary insurance, but not HMO enrollment, was associated with lower mortality risk among beneficiaries with functional limitations (OR = 0.65, P <.05) or moderate ADL disability (OR = 0.72, P <.05).

CONCLUSION

Medicare risk HMO enrollment and FFS private supplementary insurance convey similar benefits of slowing functional decline and extending life span for nonseverely disabled older people. That no association was found between adverse functional status outcomes and risk HMO enrollment has favorable implications regarding the quality of care of managed care plans.

The use of proxy respondents in studies of older adults: lessons, challenges, and opportunities

OBJECTIVE

Proxies play a critical role as sources of health information for older persons with cognitive impairment and other chronic debilitating conditions. This paper reviews the validity of proxy responses for people older than age 60 in the following areas: functioning, physical and mental health, cognition, medical care utilization, and preferences for types of care and health states.

DESIGN

A Medline review identified 24 clinical studies from 1990 to 1999 that use proxy data as a source of information about older adults.

RESULTS

In general, studies report fairly good agreement between subjects and proxies in assessments of functioning, physical health, and cognitive status, and fair-to-poor agreement in assessments of psychological well-being. Proxies tend to describe more impairment in functioning and emotional well-being, relative to subjects, a pattern that is particularly marked among persons with cognitive impairment. In addition, proxies who report more caregiver responsibilities and subjective stress from caregiver duties provide more negative assessments of subjects' health and well-being.

CONCLUSIONS

Findings tend to support the use of proxy ratings among older adults in many areas but not all when self-reports are not feasible. There is a need for more evaluation of proxy data in relation to other measures, such as performance assessments, medical records, and claims data, which may be less subject to respondent biases.

Monitoring the health status and impact of treatment on Americans: the Medicare Beneficiary Health Status Registry

OBJECTIVE

A major new survey program, the Medicare Beneficiary Health Status Registry (MBHSR), has been proposed to improve the monitoring of the health status of Medicare beneficiaries. The MBHSR would collect data by mail with telephone follow up of nonrespondents to permit economical assessment of a total Registry of approximately 200,000 Medicare beneficiaries, approximately 54,000 of whom would be surveyed in any given year. (Surveys would be conducted of samples of new enrollees who would be reinterviewed every five years.)

METHOD

To assess the feasibility of that approach, a field test was conducted with a probability sample (n = 1,922) that comprised approximately equal numbers of new Medicare enrollees (aged, 65) and current beneficiaries (age range, 76-80). The field test was designed to assess the quality of the data that this design would produce.

FINDINGS

Results indicate that the proposed design of the MBHSR could achieve response rates of approximately 80% among both age cohorts using a survey instrument that took 30 minutes to complete. Internal reliability of Activities of Daily Living, Instrumental Activities of Daily Living, Mobility, Mental Health Index, General Health, and Prostate Symptomatology scales ranged from 0.77 to 0.93. When measurements were repeated approximately 30 days after the initial survey, moderate to high levels of cross temporal correlation (range, 0.64-0.96) were found for most indexes, with the exception of prostate symptomatology. In addition, an earlier comparison of survey responses in the MBHSR field test to Medicare payment records indicated that the MBHSR field test obtained highly accurate reports of most of the major surgeries that were recorded in Medicare claims files.

CONCLUSION

The design proposed for the MBHSR is feasible. If implemented, it should produce acceptably high rates of response and data quality.