Developing indicators for the Medicare Quality Indicator System (MQIS): challenges and lessons learned.

BACKGROUND: The Medicare Quality Indicator System (MQIS), initiated in 1993, was intended to form the basis for the Health Care Financing Administration's (HCFA's; Washington, DC) hospital-based quality measurement system. The MEDSTAT Group (Washington, DC) established quality indicator development methods for five clinical conditions, including early-stage breast cancer, diabetes mellitus, and upper gastrointestinal and lower intestinal bleeding. DEVELOPMENT OF QUALITY INDICATORS: Five national expert panels were convened to identify scientifically based hospital processes of care related to positive patient outcomes, as described in research studies and/or broadly accepted by the medical and allied health professions as indicators of high-quality care. LESSONS AND CHALLENGES: One of the most important lessons learned was the need for a narrow clinical focus for each quality indicator module. If the focus was adequately narrowed, all other tasks could be accomplished more efficiently and less ambiguously. Also, selection of the expert panel chair should occur early because of the chair's pivotal contributions to the clinical focus selection, literature review, and development of draft quality indicators. Perhaps the most starting finding and biggest challenge was the paucity of scientifically sound literature to support the clinical topics; except for the breast cancer studies, most of the available literature was categorized as flawed to some degree. A critical challenge for development of quality indicators is to translate what has been learned from the best scientific efficacy studies into indicators of effective rather than efficacious care. CONCLUSIONS: In choosing quality indicators, disease incidence and potential for quality improvement are important parameters. Although the former is relatively easy to define, the latter is not.

Mammography use in older women with regular physicians: what are the predictors?

Results of a longitudinal study of 1,050 women 65 years of age and older under the care of faculty internists demonstrate that individual and sociodemographic factors predict mammography use. The majority of the sample (79%) had received a mammogram within the past two years. Multiple logistic regression analyses of findings from a telephone survey showed that mammographic screening was significantly higher among women who had recently received a Pap smear, whose annual household incomes exceeded $30,000, and whose personal health care habits were preventively oriented; they used more preventive services such as routine dental care and engaged in more preventive behaviors such as seat-belt use and routine exercise. Mammography use was significantly higher among those 75 or younger. These findings suggest that enhancing patient preventive orientations should be part of interventions designed to increase screening. Medical Subject Headings (MeSH): appropriateness review, preventive health services, utilization, aged, Medicare.

Health promotion and disease prevention for older adults: opportunity for change or preaching to the converted?

Health promotion and disease prevention for the elderly offer the potential for improving the quality of life for the growing population of older adults, while reducing the economic burden on the health system. Whether this potential can be realized depends, in part, on whether those older adults whose health behaviors put them at risk actually use preventive services when offered the opportunity. In 1988 the Health Care Financing Administration began a series of health promotion demonstrations to address health issues related to older adults. This article reports on program participation at one of the five demonstration sites. Over 1,900 community-dwelling Medicare beneficiaries who receive their health care through fee-for-service providers were enrolled in the demonstration. These enrollees were randomly assigned to control and experimental groups, with the latter offered health screening and promotion services. Nine hundred-seventy-three of the experimental subjects and 938 of the control subjects completed a lengthy telephone interview, which determined each person's use of preventive services and practices. Although all 973 experimental subjects were invited by letter and follow-up telephone calls to attend these free-of-charge, Medicare-provided sessions, not everyone attended. We analyzed 17 health behaviors of those who attended the first session and those who did not. Those engaging in these preventive behaviors were neither more nor less likely to attend the screening services. Thus, we found neither "favorable" nor "unfavorable" selection bias in the use of these preventive services.

Benzodiazepine use in older, community-dwelling southern Californians: prevalence and clinical correlates.

OBJECTIVE: To determine the use of benzodiazepines (BZDs) in an older, community-dwelling sample and to examine the sociodemographic and clinical correlates of BZD use. DESIGN: A cross-sectional study of 1752 elderly people (aged > or = 65 y) who completed a mailed medication survey and a telephone health status survey. PARTICIPANTS: Participants were invited to participate in a large Medicare demonstration project on prevention by their private physicians, who were also enrolled in the larger study. Participants had to be English-speaking, could not have dementia or a terminal illness, and had to give informed consent to participate in the study. MAIN OUTCOME MEASURES: Sociodemographic and health status variables that predicted BZD use were examined. Sociodemographic variables included age, gender, ethnicity, education, and income. Health status variables included functional status, with measures of mental, social, and physical health. Influenza immunization status was used as an indicator for preventive health services use and self-reported chronic illness was used as a measure of comorbidity. RESULTS: Twenty percent of the participants used BZDs at least twice in the past year. We found that those who used BZDs were more than twice as likely to take ten or more drugs, two-and-a-half times more likely to have difficulty falling asleep, and over twice as likely to be depressed. BZD users were also more likely to be white, to have a college education, and to have received a recent influenza shot, but were not more likely to be women when controlled for health status. CONCLUSIONS: Further clinical research should explore the relationship between BZD use among older patients and the BZD-associated adverse clinical factors we observed, as well as the association between multiple drug use and potential adverse outcomes in older BZD users.

Comparison of oral health ratings by dentists and dentate elders.

Previous studies suggest a discrepancy between the way dentists and patients measure oral health. The purpose of this study was to determine the relationship between a dentist's rating of an older dentate person's oral health and the patient self-rating using a single-item indicator, and to compare the clinical (i.e., number of teeth, caries, etc.) and subjective (problems with function, pain, etc.) factors that influence the rating. The study sample consisted of 776 older dentate people. Results showed that dentists judged subjects' oral health significantly more positively than the self-ratings. Approximately 30 percent of the elders rated their oral health identically to the dentist and half rated their oral health lower than the dentist. Bivariate comparisons showed that similar clinical and subjective variables were associated with the dentist and patient ratings. Multiple regression findings, however, highlighted differences in the factors that influenced the ratings. In addition, the proportion of variance accounted for by the clinical factors as opposed to the subjective factors was greater for the dentist rating (R2 = .28 of .33) than the subject self-rating (R2 = .18 of .43).

The relationship between dental utilization and preventive participation among a well-elderly sample.

Previous studies have shown that dental utilization by older people is lower than for the general population. This study hypothesizes that an elder's tendency toward participating in preventive health activities may be an important factor in explaining the likelihood of accessing the dentist. Subjects included 1,911 older individuals who enrolled in the UCLA Medicare Screening and Health Promotion Trial. All were interviewed about their utilization of preventive health services and participation in preventive behaviors. A recent dental visit was positively associated with all sociodemographic variables examined except age and sex. It was also related to health status questions and utilization of the preventive health services and health behaviors studied. Logistic regression analysis showed that both summary preventive health behavior and preventive service utilization variables were important factors in explaining a recent dental visit (model chi-square = 221.4, P = .001) along with income, not having a removable prosthesis, and perceiving the need for dental care. This study showed that dental utilization is related to older people's participation in other preventive activities. When in contact with older people, health care professionals should consider current oral and general preventive health status and encourage appropriate referral for preventive activities.

Patient-related predictors of rehabilitation use for community-dwelling older Americans.

OBJECTIVE: To determine patient factors that predict use of physical or occupational therapy (PT/OT) services by elderly people. DESIGN: Time-series study of the relationship of PT/OT use to a variety of characteristics present at baseline. PARTICIPANTS: Eight-hundred nine community-dwelling men and women aged 65 and older. SETTING: The Medicare Screening and Health Promotion Trial at UCLA. MAIN OUTCOME MEASURES: Use of PT/OT services, as reported in a telephone survey 1 year after initial information was obtained on the same subjects by the baseline telephonic survey. RESULTS: Fifteen percent of the sample used PT/OT within 12 months of baseline. In multiple logistic regression analysis, PT/OT use was significantly less frequent among racial minorities, less well educated groups, and the oldest age group. PT/OT use was higher among those who had both functional disability and the presence of arthritis, heart or lung disease, or a prior history of stroke. However, neither functional disability alone nor the presence of arthritis, cardiovascular or lung disease, in the absence of limited functioning, was associated with PT/OT use. Patients who had obtained a pneumococcal vaccination or used transportation services were also more likely to receive PT/OT. CONCLUSIONS: Our findings suggest that there may be important sociodemographic inequalities in the use of rehabilitation services and raise the possibility of inappropriate underuse in certain subgroups. Additional studies are needed to determine whether similar inequalities of PT/OT use are found in other populations and whether rehabilitation is effective in various subgroups.

Predictors of mortality in older patients following medical intensive care: the importance of functional status.

OBJECTIVE: We examined predictors of hospital and 6-month mortality in older Medical Intensive Care Unit (MICU) patients with particular attention to age and functional status. Age is generally thought to be strongly associated with intensive care outcomes, but this relationship may be confounded by age-related changes. These age-related changes may be approximated by changes in functional status (FS). DESIGN: We conducted a retrospective chart review and collected severity of illness data using the Acute Physiology Score (APS), pre-hospitalization FS dichotomized as limited or not limited, and hospital mortality. County death records were reviewed for 6-month mortality. SETTING: Three community hospital MICUs. PATIENTS: Four-hundred MICU patients aged 50 and older admitted during the study period. RESULTS: Limited FS was found in 42% of the 227 patients who had FS data in the chart. Mortality was significantly associated with APS, age, FS, immunocompromise state, comorbidity, and nursing home residence. In logistic regression analyses, while controlling for important variables, APS (P less than 0.001) and age greater than or equal to 75 with limited FS (P less than 0.05) were associated with hospital mortality. Six-month mortality predictors were APS (P less than 0.001), hospital (P less than 0.05), immunocompromised state (P less than 0.05) and age greater than or equal to 75 with limited FS (P less than 0.05). CONCLUSIONS: We found that among patients without functional limitations, the oldest group was no more likely to die than the youngest group. Age and functional status had a significant interaction: patients older than 75 years with functional limitations were almost six times more likely to die in hospital compared to the reference group of patients between 50-64 years old without functional limitations. We conclude that functional status is an important predictor of outcome in older MICU patients.

Differences in health status between older and younger homeless adults.

The Institute of Medicine has placed a priority on determining the special health-care needs of elderly homeless persons. As part of a community-based study of 521 homeless adults in two beach communities of Los Angeles, we compared the demographic characteristics and health of older (age rangek 50-78, n = 61) and younger (age range, 18-49, n = 460) homeless individuals. Compared with younger adults, older adults were more likely to be white (85% versus 61%), veterans (59% versus 27%), retired (36% versus 3%), and living in a vehicle (21% versus 8%). Older adults were more likely to report having a chronic disease (69% versus 49%), functional disabilities, no informal social contacts during the previous month (49% versus 27%), observed high blood pressure (42% versus 22%), elevated creatinine (11% versus 2%), and elevated cholesterol (57% versus 36%). Older adults were less likely to have a toothache (3% versus 30%), report psychotic symptoms (25% versus 42%), and to be illegal drug users (15% versus 55%). Although they are chronologically younger, the constellation of health and functional problems of older homeless adults resemble those of geriatric persons in the general population. We suggest that geriatricians could play a significant role in training other primary-care providers to evaluate and treat socially isolated older homeless adults in a more comprehensive way than is currently standard in practice (e.g., interdisciplinary team care and emphasis on functional status, rehabilitative medicine, and assessment for sensory impairment).

The early effect of Medicare's prospective payment system on the use of medical intensive care services in three community hospitals.

To assess the impact of Medicare's prospective payment system (PPS) on patient care and outcome from the medical intensive care unit (MICU), we reviewed the medical records of 400 MICU patients from three community hospitals: 200 patients were admitted before the PPS and 200 were admitted after the PPS. We sampled Medicare patients, aged 65 years and over, and non-Medicare comparison patients, aged 50 to 64 years, collecting data on case mix, treatment intensity and discharge disposition, hospital and six-month mortality, length of stay, and number of intensive care unit beds. After the PPS, the number of intensive care unit beds decreased 31%, without changes in MICU patients' illness severity or treatment intensity. Hospital length of stay decreased 15% in the Medicare group and 43% in the comparison group. For both Medicare and comparison patients combined, MICU length of stay decreased 14% and patients after the PPS were less likely to be discharged to go home. There were no significant changes in in-hospital or six-month mortality. Thus, clinically meaningful decreases in length of stay among seriously ill patients did not result in a change in in-hospital or six-month mortality.