Article Text

Download PDFPDF

RF05 The impact of distribution shifts in a population’s cardiovascular risk factors on healthy life expectancy in japan
  1. R Tsukinoki1,
  2. Y Murakami2,
  3. K Miura3,4,
  4. T Okamura5,
  5. A Kadota3,4,
  6. T Hayakawa6,
  7. A Okayama7,
  8. H Ueshima3,4
  1. 1Department of Community Health Nursing, apanese Red Cross College of Nursing, Tokyo, Japan
  2. 2Department of Medical Statistics, School of Medicine, Toho University, Tokyo, Japan
  3. 3Department of Public Health, Shiga University of Medical Sciences, Otsu, Japan
  4. 4Center for Epidemiologic Research in Asia, Shiga University of Medical Sciences, Otsu, Japan
  5. 5Department of Preventive Medicine and Public Health, Keio University, Tokyo, Japan
  6. 6The Kinugasa Research Organization, Ritsumeikan University, Kyoto, Japan
  7. 7Research Institute of Strategy for Prevention, Tokyo, Japan


Background Healthy life expectancy (HLE) is an important indicator of population health. Although the distribution of risk factors within populations has shifted during the past few decades (e.g., smoking rates have declined in developed countries), little is known if these changes affect population HLE. Our study aimed to estimate the impact of distribution shifts in a population’s cardiovascular risk factors on HLE in Japan.

Methods Data for analysis were obtained from NIPPON DATA90, a nationwide cohort study of over 9,000 Japanese people initiated in 1990. Using activities of daily living scores, we estimated the HLEs of study participants aged ≥60 years at the baseline survey. We examined shifts in the following three risk factors: systolic blood pressure (SBP), body mass index (BMI), and smoking status. All HLE calculations were performed using maximum likelihood approach with interpolated Markov chains (iMaCh 0.98r7). First, risk factor-specific HLEs were calculated using multistate life tables. A total of 24 combinations of cardiovascular risk factors were estimated. Next, we calculated the population’s baseline HLE (Scenario 0) using the weighted average of the risk factor-specific HLEs. The averaging weight was derived from the multivariable distribution of baseline data from NIPPON DATA90. Finally, scenarios were set to investigate the impact of distribution shifts in risk factors. In Scenario 1, the distributions of SBP and BMI were reduced by 4.0 mmHg and 2.0 kg/m2, respectively. In Scenario 2, we added the condition to Scenario 1 that all current smokers who wanted to quit smoking (men: 25%, women: 40%) were successful in quitting.

Results The analysis was performed using 6,676 participants (2,840 men and 3,836 women). The baseline HLE at age 60 years (Scenario 0) was 20.02 years in men and 24.32 years in women. After the shift in risk factors, the adjusted HLE at age 60 years in Scenario 1 was 20.12 years in men and 24.45 years in women, and the corresponding HLE in Scenario 2 was 20.46 years in men and 24.53 years in women. When compared with the baseline HLE, Scenario 1 presented a gain of 0.10 years in men and 0.14 years in women, and Scenario 2 presented a gain of 0.43 years in men and 0.21 years in women.

Conclusion We examined the impact of distribution shifts in SBP, BMI, and smoking status on population HLE in Japan, and found that smoking cessation substantially extended HLE.

  • Healthy life expectancy
  • cardiovascular risk factors
  • maximum likelihood approach

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.