Counting the Healthy Lifetime Cost of Obesity

1. Executive Summary

This study by Stenholm et al. (2017) attempts to investigate the association between body mass index (BMI) and health expectancy, specifically healthy life expectancy (HLE) and chronic disease-free life expectancy (CDFLE) from ages 50 to 75 across four European countries: England, Finland, France, and Sweden.

Utilising longitudinal data the study tries to evaluate the proportion of life individuals spend in good health or free of chronic diseases across five BMI categories.

The analysis revealed that individuals with normal BMI (18.5–24.9 kg/m²) experienced the highest proportion of healthy and disease-free life between ages 50 and 75. Conversely, individuals with Class II obesity (≥35 kg/m²) spent significantly fewer years in good health and were more likely to suffer from chronic conditions such as cardiovascular disease, cancer, respiratory illness, and diabetes.

Normal weight individuals experienced 7 to 9 more healthy years compared to those with Class II obesity. These findings were consistent across all cohorts and sexes.

This study illustrates that while overall life expectancy may not differ dramatically across BMI categories, the quality of those years varies substantially and therefore should be taken into account when assessing any future losses within a civil damages claim in England and Wales.

2. Introduction and Background

The adverse consequences of excess weight—including elevated risks for cardiovascular disease, type 2 diabetes, certain cancers, respiratory conditions, and musculoskeletal disorders have been studied (Ng et al., 2014; Aune et al., 2016) as has the correlation between obesity and all-cause mortality (Flegal et al., 2013; Aune et al., 2016).

It seems to me, however (and this is true of many conditions) that less is known about how increased BMI affects quality of life during aging. Particularly how BMI influences the number of years individuals lived in good health.

Health expectancy goes beyond mere longevity. It estimates the average years a person can expect to live free from disability or chronic illness, thus offering a more nuanced perspective on population health (Sanders, 1964; Wood et al., 2006) but one which I would suggest is highly relevant when looking at future losses.

The Stenholm Study applies longitudinal analysis to large cohort datasets evaluating two distinct health outcomes: self-rated health and chronic disease incidence. This approach attempts to provide a clearer picture of how body weight affects not just how long people live, but how well they live during their later years.

3. Study Objectives

The primary objective of this study was to quantify how BMI influences two types of health expectancy: healthy life expectancy (HLE) and chronic disease-free life expectancy (CDFLE) between the ages of 50 and 75, using data from four national cohort studies in Europe.

Key goals included:

• Estimating and comparing partial life expectancy (LE), HLE, and CDFLE across BMI categories (normal weight, overweight, obesity classes I and II).
• Assessing gender-specific patterns in health expectancy across BMI groups.
• Identifying cross-country similarities and differences in health expectancy outcomes using harmonised methodology and indicators.
• Evaluating the robustness of associations after accounting for confounding variables, including smoking status and socioeconomic factors.

4. Methodology

4.1 Study Design and Population

This investigation used data from four longitudinal cohort studies spanning England, Finland, France, and Sweden:

• ELSA (English Longitudinal Study of Ageing): A nationally representative, biennial survey of English residents aged 50+.
• FPS (Finnish Public Sector study): An occupational cohort of public sector employees from Finland.
• GAZEL (French Cohort): Employees of the French utility company EDF-GDF, surveyed annually since 1989.
• SLOSH (Swedish Longitudinal Occupational Survey of Health): A follow-up of Swedish Work Environment Survey respondents.

Participants aged 50–75 with valid BMI and health data at baseline were included, with total sample sizes ranging from approximately 7,000 (ELSA) to over 40,000 (FPS). The upper limit of age 75 was imposed to synchronise follow-up durations across cohorts.

4.2 BMI Categorization

Body mass index (BMI) was the primary exposure variable. It was categorised based on the World Health Organisation’s standard classification system:

• Normal weight: 18.5–24.9 kg/m²
• Overweight: 25.0–29.9 kg/m²
• Obese class I: 30.0–34.9 kg/m²
• Obese class II: ≥35.0 kg/m²
• Underweight (<18.5 kg/m²) individuals were excluded due to low sample sizes.

BMI was derived from objectively measured height and weight in ELSA, while the other studies relied on self-reported data.

4.3 Health Expectancy Metrics

1. Healthy Life Expectancy (HLE)

Defined via self-rated health, a widely validated subjective measure of general well-being. Respondents classified their health using Likert scales:

• In ELSA, FPS, and SLOSH: a 5-point scale (collapsed into good [1–2] vs. sub-optimal [3–5]).
• In GAZEL: an 8-point scale (collapsed into good [1–4] vs. sub-optimal [5–8]).

2. Chronic Disease-Free Life Expectancy (CDFLE)

Determined by self-reported physician diagnoses of major chronic illnesses:

• Cardiovascular disease (including heart attack, coronary artery disease, angina, etc.)
• Stroke or transient ischaemic attack
• Chronic lung disease (e.g., asthma, emphysema)
• Diabetes or high blood sugar
• Cancer (excluding skin cancer)

CDFLE reflects the number of years a participant is expected to live without any of these conditions between ages 50 and 75. Unlike HLE, recovery from disease is not modelled in CDFLE estimates.

4.4 Statistical Techniques

To assess transitions between health states and estimate health expectancy, the researchers employed discrete-time multi-state life table models (MSLT) based on incident data.

Key Model Components:

• For HLE: healthy → unhealthy; unhealthy → healthy; healthy → death; unhealthy → death
• For CDFLE: disease-free → diseased; disease-free → death; diseased → death
• Analyses were repeated for never-smokers to control for the confounding effect of smoking.
• Interaction terms for sex × BMI were included to evaluate gender-specific effects.

5. Key Findings

The clearly lots of ways that the Study can be criticised but, nevertheless, analysis of data from over 72,000 individuals aged 50–75 across four European cohorts yielded consistent evidence linking higher BMI with reduced healthy and chronic disease-free life expectancy: the proportion of life spent in good health or free of chronic disease diminished substantially with increasing adiposity.

5.1 Healthy Life Expectancy (Self-rated Health)

One of the central insights from this study is the progressive decline in healthy life expectancy (HLE) as BMI increases. Across all cohorts, men and women of normal weight (BMI 18.5–24.9 kg/m²) spent approximately 81% of their lives between ages 50–75 in good health.

In contrast:

• Class I obese individuals (BMI 30–34.9 kg/m²) spent only about 64% of those years in good health.
• Class II obese individuals (BMI ≥35 kg/m²) spent just 53% of that period in good health.

These percentages translated into a loss of up to 7 years of good health for the most obese participants compared to their normal-weight peers.

5.2 Chronic Disease-free Life Expectancy

The decline in chronic disease-free life expectancy (CDFLE) was even more pronounced. Normal-weight individuals could expect to live:

• 62–65% (men) and 59–73% (women) of their lives between ages 50–75 free from chronic diseases.

However:

• Class I obese individuals experienced between 36–40% of their lives without chronic disease.
• Class II obese individuals spent just 29% (men) and 36% (women) of that time free from conditions like diabetes, cardiovascular disease, or cancer.

These statistics equate to:

• 9 fewer disease-free years for men, and
• 7 fewer disease-free years for women in the Class II obesity category versus those of normal weight.

6. Discussion

The findings demonstrate a consistent relationship between BMI and declines in both healthy life expectancy (HLE) and chronic disease-free life expectancy (CDFLE).

The most striking implication is that individuals with class II obesity are likely to live substantially more years with poor health and chronic illness even if their overall lifespan is only marginally shorter than those of normal weight.

7. Conclusions

This study by Stenholm et al. delivers a concerning message: while modern healthcare has extended lifespan, rising BMI levels threaten to erode the quality of those added years.

Individuals with class II obesity can expect to live as many as 7–9 fewer healthy years between the ages of 50 and 75, compared to their normal-weight peers. These losses in health expectancy are consistent across European contexts and sexes and should be factored in to calculations for future loss as the ‘but for’ situation.

8. References

1. Ng, M., et al. (2014). Global, regional, and national prevalence of overweight and obesity... Lancet, 384(9945), 766–781. https://doi.org/10.1016/S0140-6736(14)60460-8
2. Aune, D., et al. (2016). Body mass index and all-cause mortality... BMJ, 353, i2156. https://doi.org/10.1136/bmj.i2156
3. Flegal, K. M., et al. (2013). Association of all-cause mortality with overweight and obesity... JAMA, 309(1), 71–82. https://doi.org/10.1001/jama.2012.113905
4. Renehan, A. G., et al. (2008). Body-mass index and cancer incidence... Lancet, 371(9612), 569–578. https://doi.org/10.1016/S0140-6736(08)60269-X
5. Wood, R., et al. (2006). Measuring inequalities in health: the case for healthy life expectancy. JECH, 60(12), 1089–1092. https://doi.org/10.1136/jech.2005.043794
6. Majer, I. M., et al. (2011). Life expectancy with disability by BMI category. Obesity, 19(7), 1451–1459. https://doi.org/10.1038/oby.2011.56
7. Fransen, H. P., et al. (2014). Lifestyle factors and QALYs in EPIC-NL. PLoS ONE, 9(11), e111480. https://doi.org/10.1371/journal.pone.0111480
8. Gorber, S. C., et al. (2007). A comparison of direct vs self-report BMI. Obes Rev, 8(4), 307–326. https://doi.org/10.1111/j.1467-789X.2007.00347.x
9. Stommel, M., & Schoenborn, C. A. (2009). Accuracy and usefulness of BMI measures based on self-report. BMC Public Health, 9, 421. https://doi.org/10.1186/1471-2458-9-421
10. Christensen, K., et al. (2009). Ageing populations: the challenges ahead. Lancet, 374(9696), 1196–1208. https://doi.org/10.1016/S0140-6736(09)61460-4
11. Stenholm, S., et al. (2017). Body mass index as a predictor of healthy and disease-free life expectancy between ages 50 and 75: a multicohort study. International Journal of Obesity, 41, 769–775. https://www.nature.com/articles/ijo201729