Sex, age, education, marital status, obesity, and smoking have been found to affect health transitions between non-disabled, disabled, and death. Our aim is to review the research literature on this topic and provide structured information, first on the availability of results for each risk factor and transition, and then on detailed study characteristics and disability measures. We use expert recommendations and the electronic databases Medline, PsycINFO, and SOCA. The search is confined to the years 1985-2005, and produced a total of 7,778 articles. Sixty-three articles met the selection criteria regarding study population, longitudinal design, risk factors, transition, and outcome measures.
It is well known that the risk of disability increases exponentially with age, that higher education reduces the risk of disability, that married people are less likely to experience disability than unmarried people, and that smoking is a clear risk factor for disability. Results are less clear regarding the impact of sex and body mass index (BMI) on the onset of disability, as well as for the interaction between age and sex and the various risk factors. In general, the effects of risk factors tend to become smaller with age, which is partly caused by selection effects (Hoffmann 2008). However, there are exceptions. For example, a series of studies report positive effects of being overweight on the health of the elderly, while effects at younger ages are generally negative (Losonczy et al. 1995, Himes 2000, Greenberg 2001).
In order to make reliable projections for the population, and to promote and influence political decision processes regarding health care and health systems, an instrument that provides reliable information must be available. Despite the presence of a large and complex body of literature on the effects of various socio-demographic factors, as well as of particular risk factors on disability, a suitable instrument for summarizing and using this knowledge does not yet exist. The ongoing EU-financed project "MicMac - Bridging the Micro-Macro Gap in Population Forecasting" (www.micmac-projections.org) is seeking to overcome this problem. MicMac is a multistage population projection approach aimed at European countries that combines cohort data, usually by age and sex (macro), with individual biographic data (micro). With this approach it is possible to make detailed demographic forecasts that give reliable information about the development of health care and pension systems. MicMac consists of a methodology, a set of algorithms, and user-friendly software. Within this framework we conducted a systematic literature review, focusing on the effects of age, sex, education, marital status, smoking, and obesity on various indicators of disability and mortality. The literature review serves as a source of background information for the forecasts and scenarios within the project. The present collection of research material shows the search procedure and offers structured information, first on the availability of results for each risk factor and transition, and then on detailed study characteristics and disability measures of the selected studies.
The following paper presents details and results of our review approach. In the following three sections we introduce our search strategy, outlining the processing of the articles and the underlying criteria for the choice of articles. After providing in Section 5 a detailed description of outcome variables and measurement issues, as well as a definition of transitions and their measurement, we offer in Section 6 a theoretical discussion of considered risk factors. Section 7 is devoted to the overview of specific study characteristics of the 63 selected articles, and is based on the table in the appendix.
2. Literature review
We started our literature search by analyzing the article, "Risk Factors for Functional Status Decline in Community-Living Elderly People: A Systematic Literature Review" by Stuck et al. …