Original Article
How to weight chronic diseases in multimorbidity indices? Development of a new method on the basis of individual data from five population-based studies

https://doi.org/10.1016/j.jclinepi.2011.11.006Get rights and content

Abstract

Objective

In multimorbidity indices, chronic conditions are often weighted according to their severity or their impact on different outcomes. These weights are mostly developed on the basis of only one study population by using very specific study participants, such as hospital patients. To overcome the limited validity of the indices, mean weights across five population-based studies were calculated according to the impact of diseases on self-reported health status.

Study Design and Setting

Individual data was provided from the National Health Interview and Examination Survey (n = 1,010), Dortmund Health Study (n = 281), Memory and Morbidity in Augsburg Elderly Study (n = 385), Survey of Health, Aging and Retirement in Europe Study (n = 1,278), and Study of Health in Pomerania Study (n = 962). By using logistic regression analysis, odds ratios (ORs) were calculated for reporting a fair or poor health status resulting from one of 10 different chronic conditions compared with a reference group without the specific disease, controlling for age and sex. If the results were homogenous across studies (I2 < 40%), significant pooled ORs were considered valid weights for a multimorbidity index.

Results

Myocardial infarction has the highest impact on self-reported health status across studies with a pooled OR of 3.9, followed by chronic obstructive pulmonary disease (pooled OR: 3.1). A medium impact was observed for arthrosis, asthma, diabetes mellitus, and osteoporosis.

Conclusion

This method provided valid weights for seven chronic conditions.

Section snippets

Objective

What is new?

  • -

    Individual data from five population-based studies were used to calculate valid weights for diseases included in multimorbidity indices.

  • -

    Homogenous results for the impact of seven out of 10 chronic diseases on self-reported health status across five studies were obtained.

  • -

    Compared with past approaches, where weights were mostly calculated on the basis of very specific study population such as hospital patients, it is the first attempt to use representative data from more than one

Setting and sample

For the calculation of weights, we used individual data from five population-based studies conducted in Germany including the National Health Interview and Examination Survey 1998 (NHIES) [25], the Dortmund Health Study (DHS) [26], Memory and Morbidity in Augsburg Elderly (MEMO) Study [27], German participants of the Survey of Health, Aging and Retirement in Europe (SHARE) [28], and the Study of Health in Pomerania (SHIP) [29]. To increase the homogeneity between the data sources, we only

Results

The characteristics of the five studies are remarkably similar with respect to age, sex, and education (Table 1). The mean age is between 70.2 years in the DHS and 73.0 years both in SHARE and in SHIP. As expected for this age group, most of the people have finished school after 8 years of education with a general secondary school certificate. Only the proportion of male participants in the NHIES is slightly lower than in the other four studies.

Regarding the prevalence of single diseases (Table

Discussion

We report results of a simple approach to determine the common impact of selected diseases on self-reported health status of elderly participants in five population-based studies. We restricted the analysis to older participants, because these individuals are frequently affected by multimorbidity [14], [20], [21]. We pooled results from studies that used similar or identical methods to collect information on the prevalence of diseases and on self-reported health status. We were able to define

Acknowledgments

The study is part of the “Priscus-Network” and was funded by the Ministry of Education and Research (BMBF) within the research cooperation “Health in old age.” The RKI, Berlin, provided the public-use file for the National Health Survey (BGS), which was funded by the Federal Ministry of Research and Education. The DHS was supported by the German Migraine & Headache Society (DMKG) and by unrestricted grants of equal share from Astra Zeneca, Berlin Chemie, Boots Health care, Glaxo-Smith-Kline,

References (40)

  • M. van den Akker et al.

    Problems in determining occurrence rates of multimorbidity

    J Clin Epidemiol

    (2001)
  • Y. Okura et al.

    Agreement between self-report questionnaires and medical record data was substantial for diabetes, hypertension, myocardial infarction and stroke but not for heart failure

    J Clin Epidemiol

    (2004)
  • S. Susser et al.

    Comorbidity information in older patients at an emergency visit: self-report vs. administrative data had poor agreement but similar predictive validity

    J Clin Epidemiol

    (2008)
  • M. Van den Akker et al.

    Comorbidity or multimorbidity: what’s in a name? A review of literature

    Eur J Gen Pract

    (1996)
  • G. Anderson et al.

    The growing burden of chronic disease in America

    Public Health Rep

    (2004)
  • A.B. Newman et al.

    A physiological index of comorbidity: relationship to mortality and disability

    J Gerontol A Biol Sci Med Sci

    (2008)
  • M. Fortin et al.

    Multimorbidity and quality of life in primary care: a systematic review

    Health Qual Life Outcomes

    (2004)
  • A. Marengoni et al.

    The impact of chronic multimorbidity and disability on functional decline and survival in elderly persons. A community-based, longitudinal study

    J Intern Med

    (2008)
  • K.M. Schneider et al.

    Prevalence of multiple chronic conditions in the United States Medicare population

    Health Qual Life Outcomes

    (2009)
  • G. Fillenbaum et al.

    Comorbidity of five chronic health conditions in elderly community residents: determinants and impact on mortality

    J Gerontol A Biol Sci Med Sci

    (2000)
  • Cited by (0)

    View full text