Concurrent Changes in Diet Quality and Physical Activity and Association With Adiposity in Adults.
JAMA network open 2025 ; 8: e2545232.
Aryannezhad S, Imamura F, De Lucia Rolfe E, Griffin SJ, Wareham NJ, Brage S, Forouhi NG, Forouhi N, Wareham N, Griffin S
DOI : 10.1001/jamanetworkopen.2025.45232
PubMed ID : 41269689
PMCID : PMC12639474
URL : https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2841742
Abstract
Diet and physical activity (PA) are both associated with body weight, but less is understood about how long-term changes in these behaviors may be associated with changes in adiposity in the general population.
To evaluate associations of concurrent changes in diet quality and PA with different body composition indices.
This cohort study was conducted among participants in the UK population-based Fenland study, with repeated measurements of health behaviors and adiposity. Data collection occurred in 2005 to 2015 and 2014 to 2020, with a mean (SD) follow-up of 7.2 (2.0) years. Data were analyzed from January 2024 through April 2025.
Mediterranean diet score (MDS), a measure of adherence to the Mediterranean diet (range, 0-15 points), was assessed with a food frequency questionnaire. PA energy expenditure (PAEE), measured in kilojoules per kilogram per day, was assessed and calibrated with heart rate and movement sensing. Change variables were derived.
Indices of total adiposity (weight, body mass index and body fat) and regional adiposity (waist circumference, visceral adipose tissue [VAT], and subcutaneous adipose tissue) were assessed with anthropometry and dual-energy x-ray absorptiometry. Hepatic steatosis was ascertained by abdominal ultrasonography.
The analysis included 7256 participants (mean [SD; range] age at recruitment, 48.8 [7.4; 29-65] years; 3748 female [51.7%]). In multivariable linear regression models after adjustment for potential confounders and baseline values, changes in both exposures were negatively associated with changes in all adiposity indices. For example, each 1-SD (1.27 points) increase in the change in MDS was negatively associated with changes in adiposity measures (change in body fat: β = -0.47 kg; 95% CI, -0.58 to -0.36 kg; change in VAT: β = -45 g; 95% CI, -55 to -35 g), and each 1-SD (19.0 kJ/kg/d) in the change in PAEE was similarly negatively associated with changes in adiposity measures (change in body fat: β = -1.40 kg; 95% CI, -1.51 to -1.26 kg; change in VAT: β = -108 g; 95% CI, -118 to -98 g) . Despite no evidence for a diet-PA interaction, simultaneous increases in MDS and PAEE were associated with a greater magnitude of decrease in adiposity, with a β of -149 g (95% CI, -187 to -111 g) for change in VAT among participants in higher joint change in MDS and change in PAEE tertiles. Associations had higher β values among individuals with overweight or obesity or who were physically inactive at baseline; for example, for each 1-SD increase in the change in PAEE, the β for change in body fat was -0.96 kg (95% CI, -1.10 to -0.81 kg) when baseline BMI was less than 25 and -1.74 kg (95% CI, -1.91 to -1.57 kg) when baseline BMI was 25 or greater (P for interaction < .001).
In this study, improved diet quality and increased PA were independently associated with weight loss or less weight gain and a healthier adiposity profile. Combining these health behaviors was associated with the greatest adiposity-related benefits.
Lay Summary
When people talk about changes in body weight, they often refer to a single number on the scale. But not all weight loss or gain is the same. Body fat is stored in different places, and some types are more harmful than others. Subcutaneous fat is stored under the skin and is relatively harmless, and can even be considered a healthy fat storage, while visceral fat is stored around the organs in the abdomen (like liver, causing fatty liver disease) and is linked to a higher risk of diseases such as type 2 diabetes and heart disease. So, when we gain or lose weight, it matters where these changes happen.
In our study, we followed 7,256 adults from the UK Fenland Study over about 7 years. We repeatedly measured their physical activity using wearable devices, assessed their diet quality using questionnaires, and measured different fat compartments using advanced imaging. We looked at how changes in diet and activity over time were linked to changes in weight, total fat mass, fat mass type and occurrence of fatty liver disease.
We found that improving either diet quality or physical activity was linked to less increase in total body fat, harmful visceral fat and less occurrence of fatty liver disease. For example, people who improved both their diet and activity gained on average about 150 g less visceral fat over the follow-up period compared to those who did not. This is roughly 16% of the average amount of visceral fat in the population at the start of the study. Importantly, improvements in diet and activity showed a stronger negative link with the harmful visceral fat than with the healthier subcutaneous fat in. In other words, these behaviours were relatively more connected to reduction or less gain of the in “bad” fat than “good” fat.
We also found that these benefits were strongest among people with overweight or obesity who were less active at the start. This is encouraging: even if someone begins from a less healthy starting point, improvements in diet and activity can lead to meaningful changes in body fat distribution over time.
In summary, combining a better diet with more physical activity is an effective way to improve not just weight, but where fat is stored in the body, potentially supporting healthier ageing and disease prevention.