Quantifying the contribution of 31 risk factors to the increasing prevalence of diabetes among US adults, 2005–2018

Yue Huang, Yaqing Xu, Yongxia Qiao, Hui Wang, Victor W. ZhongFrontiers in Public Health 2023 · DOI

Abstract

The concurrently changing risk factors accounted for the increasing diabetes prevalence. However, the contribution of each risk factor domain varied. Findings may inform planning cost-effective and targeted public health programs for diabetes prevention.

Introduction

Diabetes remains a leading cause of mortality and disability in the United States, with an annual economic burden exceeding $320 billion. Despite its known impact, the specific drivers behind the sharp rise in prevalence—from 12.2% in 2005–2006 to 17.1% in 2017–2018—have been difficult to isolate. This study, published in Frontiers in Public Health, provides a "collective accounting" of 31 risk factors to determine how much of this increase is driven by aging, obesity, social determinants, and biological markers.

Methods

Researchers analyzed data from seven cycles of the National Health and Nutrition Examination Survey (NHANES) spanning 2005 to 2018. The primary analysis included 16,091 participants who were at least 20 years old, not pregnant, and had complete reconstructed weights.

NHANES participant selection flowchart across seven survey cycles.

The 31 risk factors were categorized into seven domains: Genetic (family history), Demographic (age, sex, race), Social Determinants of Health (SDOH), Lifestyle, Obesity, Biological (blood pressure, cholesterol, medication use), and Psychosocial factors. The study employed Poisson regression to measure the "percent reduction in the $\beta$ coefficient," quantifying how much the trend of increasing diabetes flattened when adjusting for specific domains.

Results

The most significant finding was that the 31 factors collectively accounted for 97.3% of the rise in diabetes prevalence. After adjusting for all domains, the upward trend became statistically non-significant, suggesting that the primary drivers of the epidemic are well-mapped.

Trends in participant characteristics and risk factors across NHANES cycles from 2005 to 2018.

Key Domain Contributions

  • Biological Factors (46.2%): This was the largest single contributor. Biological markers like blood pressure and cholesterol are considered "proximal" factors—the final pathway through which environment and lifestyle impact health.
  • Demographic Shifts (41.5%): A substantial portion of the increase is attributed to an aging population and changes in ethnic composition.
  • Obesity (35.3%): BMI and waist circumference remain critical drivers of the trend.
  • Modifiable vs. Non-Modifiable: While aging and genetics are non-modifiable, modifiable factors (obesity, lifestyle, etc.) together accounted for 65% of the prevalence increase.

Interestingly, the study noted a "statin paradox": increased use of statins contributed to the biological domain's impact. While statins are vital for cardiovascular health, they are associated with a marginal increase in diabetes risk, which contributed to the rising prevalence statistics.

Discussion and Conclusion

This comprehensive decomposition highlights that the rising prevalence of diabetes in the US is not a mystery but a result of measurable shifts in population health and demographics. The findings underscore the importance of targeted public health interventions focusing on modifiable risk factors, particularly obesity and lifestyle, while also acknowledging the significant impact of an aging society. Future research should explore these drivers within specific subgroups to further refine targeted prevention strategies.

Summary updated

Abstract

The concurrently changing risk factors accounted for the increasing diabetes prevalence. However, the contribution of each risk factor domain varied. Findings may inform planning cost-effective and targeted public health programs for diabetes prevention.
BibTeX
@inproceedings{huang2023quantifying,
  author = {Yue Huang and Yaqing Xu and Yongxia Qiao and Hui Wang and Victor W. Zhong},
  title = {{Quantifying the contribution of 31 risk factors to the increasing prevalence of diabetes among US adults, 2005–2018}},
  booktitle = {Frontiers in Public Health},
  year = {2023},
  doi = {10.3389/fpubh.2023.1174632},
}
Figure