Article
Plant-based dietary patterns and risk of all-cause mortality in diabetes subgroups: a prospective cohort study from the UK Biobank
Search Medline for
Authors
-
Edyta Szczerba
- Deutsches Diabetes-Zentrum (DDZ), Leibniz-Zentrum für Diabetes-Forschung an der Heinrich-Heine-Universität Düsseldorf, Düsseldorf, Germany; Deutsches Zentrum für Diabetesforschung, Neuherberg, Germany
-
Janett Barbaresko
- Deutsches Diabetes-Zentrum (DDZ), Leibniz-Zentrum für Diabetes-Forschung an der Heinrich-Heine-Universität Düsseldorf, Düsseldorf, Germany
-
Sabrina Schlesinger
- Deutsches Diabetes-Zentrum (DDZ), Düsseldorf, Germany; Deutsches Zentrum für Diabetesforschung, Neuherberg, Germany
| Published: | September 6, 2024 |
|---|
Outline
Text
Background: Plant-based diets have been shown to improve cardiovascular risk markers in persons with type 2 diabetes (T2D). However, it is unknown whether plant-based diets also have the potential to improve T2D progression, and therefore reduce the risk of premature death. Moreover, T2D is a heterogenous disease and different disease characteristics can lead to different disease progression, risk of diabetes complications and premature death. We aimed to investigate the association between adherence to an overall plant-based dietary index (PDI), as well as a healthy-PDI and unhealthy-PDI and all-cause mortality in persons with T2D, and whether these associations differ in T2D subgroups.
Methods: We included 4,829 individuals with existing T2D and at least two available 24-hour dietary recalls from the UK Biobank cohort. We generated an overall PDI that rated plant-based foods positively and animal-based foods negatively; furthermore, healthy-PDI only scored healthy plant-based products positively, while unhealthy-PDI only scored less healthy plant-based foods positively. Multivariable adjusted Cox regression was conducted to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for all-cause mortality, comparing highest (T3) with lowest tertile (T1) of adherence and per 5 points of PDI, healthy-PDI, and unhealthy-PDI adherence. Non-linearity of the associations was investigated with restricted cubic splines. We investigated effect modification by diabetes subgroups, which were defined as early age at diagnosis (≤45 years), high waist circumference (women: ≥88; men: ≥102 cm) and elevated HbA1c (≥48 mmol/l).
Results: During mean follow-up time of 11.3 years, 679 cases of all-cause mortality occured. Higher PDI was associated with lower all-cause mortality risk (per 5 points: HR (95% CI) 0.92 (0.85, 0.99); T3 vs T1: 0.92 (0.85, 0.99)). The association for healthy-PDI was inverse but imprecisely estimated (per 5 points: 0.95 (0.89, 1.01); T3 vs T1: 0.82 (0.67, 1.02)). Opposite direction of the association was observed for unhealthy-PDI (per 5 points: 1.08 (1.01, 1.15); T3 vs T1: 1.24 (1.00, 1.54)). The association between healthy-PDI and all-cause mortality was not linear, with a decrease of relative risk observed after reaching higher levels of healthy-PDI adherence. There was no clear indication for effect modification of the associations between PDI, healthy-PDI, unhealthy-PDI and all-cause mortality across T2D subgroups, however the associations were stronger for all three dietary scores in the subgroup of younger age at diagnosis and better glycemic control.
Conclusions: We showed that higher adherence to PDI, particullarly healthfy-PDI, associated with decreased, while higher adherence to unhealthy-PDI associated with an increased all-cause mortality risk. There was no indication for effect modification, however effect estimates were stronger in the subgroup of younger age at diagnosis and in those with better gylcemic control.
The authors declare that they have no competing interests.
The authors declare that an ethics committee vote is not required.
