Patterns of change in self-reported oral health among dentate adults.

BACKGROUND: Although self-assessments of oral health have become useful tools in dental research, the use of self-reports to study changes in oral health over time has been limited. The aim of this investigation was to describe how oral disease and tissue damage, pain, functional limitation, disadvantage, and self-rated oral health change over time. METHODS: The Florida Dental Care Study (FDCS) (n = 873) is a longitudinal study of oral health among dentate adults (age, > or = 45 years). Incidence rates and transition probabilities were used to describe changes in oral health over a 24-month period. RESULTS: The probability of reporting a specific problem during the 24-month study ranged from 0.52 for perceived need for dental care to 0.07 for avoided eating with others. Only dental sensitivity and perceived need for dental care had transition probabilities >0.20. Decomposition of transition probabilities revealed moderate probabilities of onset coupled with relatively high probabilities of recovery. CONCLUSION: Although oral health status is clearly dynamic, no individual measure exhibited profound fluctuation. Most oral health problems were episodic rather than chronic. Patterns of change in oral health varied across dimensional lines.

Differences by race in the decline of health over time.

Previous research on race differences in health, we believe, has failed to take into account the initial state of health of the respondents. Other research has demonstrated that elders in poor health are more likely to experience a change in their health over time. It is unclear if the greater probability of decline in health observed among African Americans is a result of being more likely to begin such observations in health states that are worse than those for Whites. This investigation examines declines in health over a 30-month period in a sample of African American and White elders who began the study in similar "good health." Findings support the supposition that African Americans are more likely to report a decline in their health, regardless of the health measure used. Differences by race in the decline of health appear to be a consequence of economic and educational discrepancies between the two groups.

Selection of children to provide care: the effect of earlier parental transfers.

We use the first wave of data from the Asset and Health Dynamics Among the Oldest Old (AHEAD) study to examine the effects of past parent-to-child financial transfers on selection of a child to provide assistance with basic personal care for unmarried parents. We estimate a fixed-effects conditional logit model and find a positive and significant association between past financial transfers and a child's current helping behavior. The coefficient of past financial transfers is in the direction hypothesized, and its magnitude is 80% as large as that of gender, a well-documented powerful predictor of parental caregiving. There appears to be substantial evidence that earlier parent-to-child financial gifts play a role in determining which child in the family will provide assistance.

Race differences in the health of elders who live alone.

This investigation was initiated to determine whether older African Americans who live alone are in poorer health than their White counterparts who live under the same circumstances. Data on 5 measures of health were collected in telephone interviews with a stratified random sample of community-dwelling elders (n = 1,189). Analysis of weighted data indicate that there were fewer differences in health by race among older persons who lived alone compared to elders who lived with others. Where racial differences in health did exist among older adults who lived alone, the differences could only sometimes be accounted for by population composition factors that are known to influence health.

A longitudinal analysis of adherence and health status in childhood diabetes.

Applied structural equation modeling to a longitudinal data set of 193 youngsters with insulin-dependent diabetes mellitus assessed on two occasions, an average of 1.65 years apart. Six adherence constructs, Injection, Exercise, Diet Type, Testing-Eating Frequency, Calories Consumed, and Concentrated Sweets, were quantified from 24-hr recall interviews conducted with mother and child. Glycemic control was indexed by glycosylated hemoglobin (HA1C); lipid metabolism was indexed by fasting triglyceride levels (TRIG). The relationship of each adherence construct to metabolic control was tested separately. Patient age and disease duration served as exogenous variables in all models. Testing-Eating Frequency was associated with HA1C and Injection was associated with TRIG; in both cases better adherence was associated with better metabolic control. However, the standardized regression weights and variance accounted for were small. Patient age was a predictor of both adherence and metabolic control; older youngsters were less adherent and were in worse metabolic control. Inspection of models for younger versus older children suggested that age-homogeneous models improved prediction, but adherence and metabolic control linkages remained weak. Suggestions for refining the model are provided.

Retirement reason versus retirement process: examining the reasons for retirement typology.

Studies have often used reason for retirement as an indicator of the pathway leading to retirement. We discuss the conceptual basis for the retirement-reason typology and evaluate the distinctiveness of various reasons for labor force exit by predicting them in a standard model-based analysis. Data are from the 1982 Social Security New Beneficiary Study, and the analysis is limited to men. A number of factors in the model-based analysis have distinctive effects on exit for particular retirement reasons, but health limits increase the likelihood of all types of retirement. We conclude that reasons for retirement only partially capture distinctive retirement processes.

Adherence in childhood diabetes: results of a confirmatory factor analysis.

In an earlier study (Johnson, Silverstein, Rosenbloom, Carter, & Cunningham, 1986), an exploratory factor analysis identified five components of adherence in childhood diabetes. In this investigation, a simultaneous confirmatory analysis was used to test the equality of this factor pattern across two independent samples. Factors 1 through 4--Exercise, Injection, Diet Type, and Eating/Testing Frequency-were confirmed. Factor 5--Diet Amount--proved to be too complex; the adherence measures comprising this factor (total calories and concentrated sweets consumed) are best treated as separate, single-indicator constructs. The results support a multivariate conceptualization of adherence, offer insight into the nature of the components underlying diabetes adherence, and provide measurement information for reliable component estimation.

Labor-force participation of older married women.

This article utilizes the 1969, 1971, and 1973 waves of the Longitudinal Retirement History Study (LRHS) to examine stopping work by working wives of respondents. Different patterns of labor-force participation reveal that younger wives of respondents were more likely to work than were older wives. Most wives did not reenter the labor force after leaving it. The determinants of stopping or continuing work in 1969-73 for those wives who worked in 1969 were also examined. Although the patterns were somewhat different for younger wives, two factors stand out: Coverage of the wife by a private pension plan and providing for children or elderly parents have substantial effects on the probability of continuing work.