The effect on body weight of people in England associated with stay-at-home policies introduced to control the spread of COVID-19 is not yet known.
In 2016, the National Health Service (NHS) in England established The Healthier You: NHS Diabetes Prevention Programme (NHS DPP) and about 2 years later, universal population coverage in England was achieved. The NHS DPP was developed to prevent or delay the onset of type 2 diabetes in identified adults with non-diabetic hyperglycaemia (HbA1c 42–47 mmol/mol [6·0–6·4%] or fasting plasma glucose 5·5–6·9mmol/L), through structured lifestyle interventions. Programme rationale, justification, implementation, and early outcomes have been described previously.1 More than 750 000 people have now been referred into the programme.
In March, 2020, in response to the COVID-19 pandemic, programme delivery switched from group-based face-to-face sessions to remote sessions via telephone, video conferencing, or specific digital interventions. Body weight measurements changed from independently recorded to self-reported measurements.
We compared body weight of participants at entry to the programme between April 1, 2020, and March 31, 2021, with participants' weights at entry over the preceding 3 years (between April 1, 2017, and March 31, 2020), to estimate the effect of the COVID-19 pandemic on the body weight of people with non-diabetic hyperglycaemia (appendix p 1).
The NHS DPP minimum dataset contains data on all patients with non-diabetic hyperglycaemia referred to the programme and was used to identify the recorded weight of people at programme entry accessing at least one intervention session between April 1, 2017, and March 31, 2021 (appendix p 1). The legal basis for the data collections, data flows, and service evaluations have been described previously.1
72 611 participants remotely accessed an initial intervention NHS DPP session between April 1, 2020, and March 31, 2021, and 217 181 participants attended an initial intervention session during the previous 3 years. Of those, 46 069 (63%) and 208 054 (96%) participants respectively had a valid body weight measurement recorded at programme entry. The proportion of missing weight data in 2020–21 was higher for Black, Mixed, other, and unknown ethnic groups, but otherwise proportions of missing data were broadly similar across participant characteristics (appendix p 2).
The mean BMI was 30·8 kg/m2 (95% CI 30·8–30·9) and mean body weight was 86·8 kg (86·6–87·0) during 2020–21, compared with a mean BMI of 30·4 kg/m2 (30·4–30·4) and mean body weight of 84·4 kg (84·3–84·5) over the previous 3 years; an unadjusted weight difference of 2·4 kg (2·2–2·6; p<0·0001; (table ). This weight difference was substantially larger than any yearly increase in 2017–19 (appendix pp 2–3), and the higher baseline weights were apparent from the onset of the pandemic. Participants with baseline weight data in 2020–21 were younger than those in the previous 3 years, whereas other characteristics were broadly similar.
Table.
Mean body weight at entry to the National Health Service Diabetes Prevention Programme during the COVID-19 pandemic and in the preceding 3 years
| Participants entering the programme in 2017–20 | Participants entering the programme in 2020–21 | Mean baseline weight in 2017–20, kg | Mean baseline weight in 2020–21, kg | Unadjusted weight difference, kg | Adjusted weight difference, kg* | p value for adjusted weight difference | ||
|---|---|---|---|---|---|---|---|---|
| Age, years | ||||||||
| <40 | 7169 (3%) | 3258 (7%) | 91·4 (90·8–92·0) | 95·3 (94·4–96·3) | 3·9 (2·8 to 5·0) | 3·00 (2·22 to 3·78) | <0·0001 | |
| 40–64 | 84 641 (41%) | 24 180 (52%) | 88·6 (88·5–88·8) | 90·6 (90·3–90·9) | 1·9 (1·6 to 2·2) | 1·67 (1·4 to 1·94) | <0·0001 | |
| 65–74 | 73 703 (35%) | 12 402 (27%) | 82·9 (82·8–83·1) | 82·6 (82·2–82·9) | −0·4 (−0·7 to 0·0) | −0·58 (−0·93 to −0·22) | 0·0014 | |
| ≥75 | 42 541 (20%) | 6229 (14%) | 77·2 (77·1–77·4) | 76·2 (75·8–76·6) | −1·0 (−1·4 to −0·6) | −1·19 (−1·68 to −0·69) | <0·0001 | |
| Sex | ||||||||
| Female | 114 142 (55%) | 25 776 (56%) | 79·0 (78·9–79·1) | 82·7 (82·4–82·9) | 3·7 (3·4 to 4·0) | 1·54 (1·28 to 1·79) | <0·0001 | |
| Male | 93 464 (45%) | 20 253 (44%) | 91·0 (90·9–91·1) | 92·1 (91·8–92·4) | 1·1 (0·8 to 1·4) | −0·40 (−0·69 to −0·12) | 0·0058 | |
| Unknown | 448 (<1%) | 40 (<1%) | 82·6 (81–84·3) | 82·2 (75·3–89·1) | −0·4 (−6·3 to 5·4) | NA | NA | |
| Ethnicity | ||||||||
| Asian | 23 228 (11%) | 5752 (12%) | 76·2 (76·0–76·4) | 77·9 (77·5–78·4) | 1·7 (1·3 to 2·2) | −0·03 (−0·56 to 0·50) | 0·91 | |
| Black | 12 073 (6%) | 3163 (7%) | 87·0 (86·7–87·3) | 89·2 (88·5–89·9) | 2·2 (1·5 to 3·0) | 0·62 (−0·10 to 1·34) | 0·092 | |
| Mixed | 3422 (2%) | 883 (2%) | 85·1 (84·5–85·8) | 88·4 (86·8–90·0) | 3·3 (1·7 to 4·9) | 0·82 (−0·54 to 2·18) | 0·24 | |
| Other | 3071 (1%) | 576 (1%) | 80·6 (80·0–81·3) | 82·0 (80·2–83·7) | 1·3 (−0·4 to 3·0) | −0·65 (−2·28 to 0·99) | 0·44 | |
| White | 152 775 (73%) | 34 232 (74%) | 85·5 (85·4–85·6) | 88·2 (87·9–88·4) | 2·7 (2·4 to 2·9) | 0·82 (0·60 to 1·04) | <0·0001 | |
| Unknown | 13 485 (6%) | 1463 (3%) | 83·9 (83·5–84·2) | 85·4 (84·3–86·5) | 1·6 (0·5 to 2·6) | NA | NA | |
| Deprivation quintile | ||||||||
| IMD 1 (most deprived) | 34 899 (17%) | 7615 (17%) | 86·5 (86·2–86·7) | 89·6 (89·0–90·1) | 3·1 (2·6 to 3·6) | 1·53 (1·07 to 2·00) | <0·0001 | |
| IMD 2 | 38 295 (18%) | 8494 (18%) | 85·1 (84·9–85·3) | 88·4 (87·9–88·9) | 3·2 (2·8 to 3·7) | 1·45 (1·01 to 1·89) | <0·0001 | |
| IMD 3 | 41 715 (20%) | 9494 (21%) | 84·3 (84·1–84·5) | 86·4 (85·9–86·8) | 2·1 (1·6 to 2·5) | 0·32 (−0·10 to 0·74) | 0·13 | |
| IMD 4 | 44 176 (21%) | 9997 (22%) | 83·8 (83·6–84·0) | 86·0 (85·6–86·5) | 2·2 (1·8 to 2·7) | 0·37 (−0·04 to 0·78) | 0·074 | |
| IMD 5 (least deprived) | 48 670 (23%) | 10 428 (23%) | 82·9 (82·8–83·1) | 84·7 (84·3–85·1) | 1·8 (1·4 to 2·2) | 0·06 (−0·34 to 0·45) | 0·77 | |
| Unknown | 299 (<1%) | 41 (<1%) | 85·2 (83·0–87·5) | 85·5 (77·9–93·1) | 0·3 (−6·4 to 7·0) | NA | NA | |
| Total | 208 054 (100%) | 46 069 (100%) | 84·4 (84·3–84·5) | 86·8 (86·6–87·0) | 2·4 (2·2 to 2·6) | 0·68 (0·49 to 0·87)† | <0·0001 | |
Data are n (%), mean (95% CI), or p values. Adjusted mean difference was not calculated for unknown categories. NA=not applicable. IMD=Index of Multiple Deprivation.
Adjusted mean weight differences for a given characteristic were calculated using a multivariable linear regression model incorporating an interaction term between time period and the given characteristic.
This value relates to a regression model incorporating all characteristics.
After adjusting for age, sex, ethnicity, and socioeconomic deprivation, the mean body weight of people entering the programme during the pandemic was 0·68 kg (95% CI 0·49–0·87; p<0·0001) greater than that of people entering the programme in the 3 prepandemic years (table). There were significant interactions between participant characteristics and the observed differences in weight between the two time periods. Higher weights were recorded during the pandemic for people who were younger, female, and in the two quintiles of greatest deprivation compared with those who were older, male, and in the quintiles of lesser deprivation respectively. Asian ethnicity was associated with a smaller difference in body weight than White ethnicity; there were no significant differences for the other ethnic groups compared with White ethnicity, though the participant numbers in some population subgroups were small.
There has been much speculation about the effect of the COVID-19 pandemic and the resulting lockdown restrictions on body weight, especially given the known association of COVID-19 severity with high BMI.2 This large analysis of data from more than 46 000 people at high risk of type 2 diabetes in England over the past year, compared with more than 200 000 people in the preceding 3 years, provides new insights into differential effects of the COVID-19 pandemic on body weight across the population. Participants entering the NHS DPP in 2020–21 tended to be younger than those entering in preceding years, which we have previously shown when the programme is delivered digitally.3 However, even after adjustment for this, body weights at entry recorded during the pandemic remained significantly greater than those recorded in the previous 3 years.
Among people who were younger than 65 years, female, and in the two quintiles of greatest deprivation, the increase in baseline weights between the two cohorts assessed before and after pandemic onset was more than twice as large as the difference in the total sample. This greater weight increase is likely to have a significant effect on future type 2 diabetes incidence in these groups. After adjustment for initial BMI, the odds ratio of developing diabetes following a 1 kg longitudinal increase in body weight has been reported to be 1·08 (95% CI 1·04–1·13).4 Future incidence might also be influenced by delays in presentation with type 2 diabetes because of health service disruption due to COVID-19, and a possible direct effect of SARS-CoV-2 on pancreatic β cell function.5 In our companion Article,6 we show an increase in hospital admissions coded with diabetic ketoacidosis during the pandemic in England in people with newly diagnosed diabetes; the underlying aetiology is unclear, although weight gain during home confinement might have been contributory.
This study has some substantial limitations. These data are not longitudinal, and we assessed different cohorts before and after pandemic onset. The reliance on self-reported weights in 2020–21 might have led to an underestimation of the true difference, as there is a tendency to under-report weight with self-reported measurements.7 In addition, there were substantial missing weight data in 2020–21, which might have had an effect on the estimation of weight difference (appendix p 2). Given that the missing weight data were unlikely to be missing at random and that the distribution of missing weight values was not known, it was not possible to determine the effect of this missing data.
Two longitudinal studies in the USA have reported broadly similar effects on body weight before and after lockdown. In the first study, a convenience sample of 727 adults from across the country self-reported body weight during peak lockdown and 5 months later. There was a mean increase in weight of 0·62 kg; around 40% of participants reported gaining weight, and 18% reported losing weight.8 In the second, much larger study, of 11 534 attendees at a medical facility in Massachusetts who had body weight recorded before and after lockdown (on average 3 months later), there was an increase in weight among women (0·51 kg [95% CI 0·31 to 0·72]), but a decrease among men (−0·81 kg [–1·14 to −0·46]), with the highest risk of weight gain among people younger than 40 years.9 A systematic scoping review assessing the effect of COVID-19 on body weight and weight-related behaviours included 19 studies all using online self-report surveys and found that almost half of respondents reported that they had gained weight during the pandemic.10
The stay-at-home policies in England as a result of the COVID-19 pandemic have coincided with a small but clinically significant increase in mean body weight in people at high risk of type 2 diabetes, and differences were greater among women, younger people, and those from more deprived areas. This increase in body weight is likely to lead to an increase in the incidence of type 2 diabetes, and highlights the importance of ongoing access to the NHS DPP in England.
JV is the National Clinical Director for Diabetes and Obesity at NHS England and NHS Improvement. EB is an analyst at NHS England and NHS Improvement and leads analysis of the NHS DPP. DB is an analyst at NHS England and NHS Improvement and is involved in analysis of the NHS DPP. CB is the Primary Care Advisor to the NHS DPP. KK has been a consultant and speaker for Novartis, Novo Nordisk, Sanofi-Aventis, Lilly, and Merck Sharp & Dohme, has received grants in support of investigator and investigator-initiated trials from Novartis, Novo Nordisk, Sanofi-Aventis, Lilly, Merck Sharp & Dohme, Pfizer, and Boehringer Ingelheim, and has served on advisory boards for Novo Nordisk, Sanofi-Aventis, Lilly, and Merck Sharp & Dohme. KK is supported by the National Institute for Health Research (NIHR) Applied Research Collaboration East Midlands and the NIHR Leicester Biomedical Research Centre. SJ is funded by the NIHR Oxford Biomedical Research Centre and Oxford NIHR Collaboration and Leadership in Applied Health Research. We thank Eszter Vamos (Imperial College London, London, UK) for providing statistical advice.
Supplementary Material
References
- 1.Valabhji J, Barron E, Bradley D, et al. Early outcomes from the English National Health Service Diabetes Prevention Programme. Diabetes Care. 2020;43:152–160. doi: 10.2337/dc19-1425. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Gao M, Piernas C, Astbury NM, et al. Associations between body-mass index and COVID-19 severity in 6·9 million people in England: a prospective, community-based, cohort study. Lancet Diabetes Endocrinol. 2021;9:350–359. doi: 10.1016/S2213-8587(21)00089-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.McGough B, Murray E, Brownlee L, Barron E, Smith J, Valabhji J. The healthier you: NHS Diabetes Prevention Programme: digital modes of delivery engage younger people. Diabet Med. 2019;36:1510–1511. doi: 10.1111/dme.14083. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Jacobs-van der Bruggen MAM, Spijkerman A, van Baal PH, et al. Weight change and incident diabetes: addressing an unresolved issue. Am J Epidemiol. 2010;172:263–270. doi: 10.1093/aje/kwq134. [DOI] [PubMed] [Google Scholar]
- 5.Rubino F, Amiel SA, Zimmet P, et al. New-onset diabetes in COVID-19. N Engl J Med. 2020;383:789–790. doi: 10.1056/NEJMc2018688. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Misra S, Barron E, Vamos E, et al. Temporal trends in emergency admissions for diabetic ketoacidosis in people with diabetes in England before and during the COVID-19 pandemic: a population-based study. Lancet Diabetes Endocrinol. 2021 doi: 10.1016/S2213-8587(21)00208-4. published online Sept 2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hodge JM, Shah R, McCullough ML, Gapstur SM, Patel AV. Validation of self-reported height and weight in a large, nationwide cohort of U.S. adults. PLoS One. 2020;15 doi: 10.1371/journal.pone.0231229. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Bhutani S, vanDellen MR, Cooper JA. Longitudinal weight gain and related risk behaviors during the COVID-19 pandemic in adults in the US. Nutrients. 2021;13:671. doi: 10.3390/nu13020671. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Mulugeta W, Desalegn H, Solomon S. Impact of the COVID-19 pandemic lockdown on weight status and factors associated with weight gain among adults in Massachusetts. Clin Obes. 2021;11 doi: 10.1111/cob.12453. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Chew HSJ, Lopez V. Global impact of COVID-19 on weight and weight-related behaviours in the adult population: a scoping review. Int J Environ Res Public Health. 2021;18 doi: 10.3390/ijerph18041876. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.