TY - JOUR
T1 - Prediabetes transitions to normoglycaemia or type 2 diabetes and associated risk factors in the Obesity, Diabetes and Cardiovascular Disease Collaboration: an individual-level pooled analysis of 19 prospective cohort studies
AU - Davoodian, Najmeh
AU - Lotfaliany, Mojtaba
AU - Huxley, Rachel R.
AU - Lee, Crystal Man Ying
AU - Pasco, Julie A.
AU - Adams, Robert J.
AU - Azizi, Fereidoun
AU - Bertoni, Alain G.
AU - Björkelund, Cecilia
AU - Colagiuri, Stephen
AU - Fahimfar, Noushin
AU - Gabriel, Rafael
AU - Giedraitis, Vilmantas
AU - Gill, Tiffany K.
AU - González, Clicerio
AU - Gregg, Edward W.
AU - Hadaegh, Farzad
AU - Hange, Dominique
AU - Harland, Jacob W.
AU - Hodge, Allison M.
AU - Holloway-Kew, Kara L.
AU - Hosseini, Seyed Reza
AU - Jacobs, David R.
AU - Khalili, Davood
AU - Magliano, Dianna J.
AU - Mongraw-Chaffin, Morgana
AU - Mishra, Gita D.
AU - Lissner, Lauren
AU - Mehlig, Kirsten
AU - Najafipour, Hamid
AU - Nieto-Martinez, Ramfis
AU - Ostovar, Afshin
AU - Shadkam, Mitra
AU - Shaw, Jonathan E.
AU - Sundh, Valter
AU - Schreiner, Pamela J.
AU - Sakurai, Masaru
AU - Wittert, Gary A.
AU - Yatsuya, Hiroshi
AU - Zethelius, Bjorn
AU - Mohebbi, Mohammadreza
N1 - Publisher Copyright:
© 2025 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY-NC-ND 4.0 license
PY - 2025/9
Y1 - 2025/9
N2 - Background: With the increasing global burden of type 2 diabetes, prevention strategies that target prediabetes, a state of hyperglycaemia that puts individuals at high risk of type 2 diabetes, are required. We aimed to estimate global rates of transition from prediabetes to normoglycaemia or type 2 diabetes, stratified by age, sex, and race and ethnicity. We also aimed to quantify the effect of modifiable and non-modifiable risk factors on these transitions. Methods: In this pooled analysis of individual-level data, we included original data from 19 prospective cohort studies conducted in Asia (Iran and Japan), Australia, Europe (Spain and Sweden), North America (USA and Mexico), and South America (Venezuela). We applied discrete-time hidden Markov models to estimate rates and ratios of prediabetes transitions to type 2 diabetes and normoglycaemia specific to age, sex, and race and ethnicity. We used Fine–Gray competing risk models to derive cohort-specific subhazard ratios (SHRs) for potential risk factors influencing these transitions. We subsequently pooled these SHRs using a random-effects meta-analysis. In subgroup analyses stratified by age, sex, race and ethnicity, and recruitment period, we used multivariate Cox models to investigate the degree of heterogeneity between studies. Findings: 76 092 participants (39 842 [52·3%] women and 36 250 [47·6%] men; mean age 51·1 years [SD 12·7]) with available data on glycaemic status from at least one follow-up visit were included in the analysis, of whom 56 837 (74·7%) had normoglycaemia and 19 255 (25·3%) had prediabetes. Median follow-up was 9·8 years (IQR 5·8–12·5). Within 10 years, individuals with prediabetes had a 12·5% probability of progressing to type 2 diabetes, whereas the probability of reverting to normoglycaemia was 36·1%. However, in the highest fasting plasma glucose quartile, the probability of progression increased to 16·1% and reversion decreased to 13·4%. Male sex, older age (≥55 years), and Latinx populations were associated with an increased risk of transitioning to type 2 diabetes. Risk factors that significantly reduced prediabetes reversion to normoglycaemia were overweight (SHR 0·88 [95% CI 0·76–0·99]), obesity (0·66 [0·52–0·81]), elevated waist-to-height ratio (0·82 [0·70–0·95]), elevated waist-to-hip ratio (0·79 [0·68–0·91]), and reduced HDL concentration (0·72 [0·59–0·84]). Interpretation: Our findings highlight that reversion to normoglycaemia was more common than progression to type 2 diabetes among individuals with prediabetes, and that these transitions were strongly influenced by modifiable risk factors. The increased risk of progression with advancing age and among men underscores the importance of early identification and targeted interventions in population groups at high risk of type 2 diabetes. Furthermore, the elevated progression risk in individuals with higher fasting plasma glucose concentrations at baseline reinforces the need for timely detection and intervention during this crucial clinical window. Funding: Deakin University Postgraduate Research Scholarship.
AB - Background: With the increasing global burden of type 2 diabetes, prevention strategies that target prediabetes, a state of hyperglycaemia that puts individuals at high risk of type 2 diabetes, are required. We aimed to estimate global rates of transition from prediabetes to normoglycaemia or type 2 diabetes, stratified by age, sex, and race and ethnicity. We also aimed to quantify the effect of modifiable and non-modifiable risk factors on these transitions. Methods: In this pooled analysis of individual-level data, we included original data from 19 prospective cohort studies conducted in Asia (Iran and Japan), Australia, Europe (Spain and Sweden), North America (USA and Mexico), and South America (Venezuela). We applied discrete-time hidden Markov models to estimate rates and ratios of prediabetes transitions to type 2 diabetes and normoglycaemia specific to age, sex, and race and ethnicity. We used Fine–Gray competing risk models to derive cohort-specific subhazard ratios (SHRs) for potential risk factors influencing these transitions. We subsequently pooled these SHRs using a random-effects meta-analysis. In subgroup analyses stratified by age, sex, race and ethnicity, and recruitment period, we used multivariate Cox models to investigate the degree of heterogeneity between studies. Findings: 76 092 participants (39 842 [52·3%] women and 36 250 [47·6%] men; mean age 51·1 years [SD 12·7]) with available data on glycaemic status from at least one follow-up visit were included in the analysis, of whom 56 837 (74·7%) had normoglycaemia and 19 255 (25·3%) had prediabetes. Median follow-up was 9·8 years (IQR 5·8–12·5). Within 10 years, individuals with prediabetes had a 12·5% probability of progressing to type 2 diabetes, whereas the probability of reverting to normoglycaemia was 36·1%. However, in the highest fasting plasma glucose quartile, the probability of progression increased to 16·1% and reversion decreased to 13·4%. Male sex, older age (≥55 years), and Latinx populations were associated with an increased risk of transitioning to type 2 diabetes. Risk factors that significantly reduced prediabetes reversion to normoglycaemia were overweight (SHR 0·88 [95% CI 0·76–0·99]), obesity (0·66 [0·52–0·81]), elevated waist-to-height ratio (0·82 [0·70–0·95]), elevated waist-to-hip ratio (0·79 [0·68–0·91]), and reduced HDL concentration (0·72 [0·59–0·84]). Interpretation: Our findings highlight that reversion to normoglycaemia was more common than progression to type 2 diabetes among individuals with prediabetes, and that these transitions were strongly influenced by modifiable risk factors. The increased risk of progression with advancing age and among men underscores the importance of early identification and targeted interventions in population groups at high risk of type 2 diabetes. Furthermore, the elevated progression risk in individuals with higher fasting plasma glucose concentrations at baseline reinforces the need for timely detection and intervention during this crucial clinical window. Funding: Deakin University Postgraduate Research Scholarship.
UR - https://www.scopus.com/pages/publications/105013853776
U2 - 10.1016/S2214-109X(25)00237-2
DO - 10.1016/S2214-109X(25)00237-2
M3 - Article
C2 - 40845880
AN - SCOPUS:105013853776
SN - 2572-116X
VL - 13
SP - e1533-e1542
JO - The Lancet Global Health
JF - The Lancet Global Health
IS - 9
ER -