Article Text

SAT0096 The Obesity Paradox in Rheumatoid Arthritis: Results from Uk Based Multi-Centre Observational Study
  1. S. Norton1,
  2. A. Young2,3,
  3. E. Nikiphorou2,3,
  4. P. Kiely4,
  5. D.A. Walsh5,
  6. J. Dixey6
  1. 1Department of Psychology, Institute of Psychiatry, Kings College London, London
  2. 2School of Life & Medical Sciences, University of Hertfordshire, Hatfield
  3. 3Eras&Eran, Rheumatology Department, St Albans City Hospital, St Albans
  4. 4Department of Rheumatology, St Georges Healthcare Trust, London
  5. 5Arthritis UK Pain Centre, University of Nottingham, Nottingham
  6. 6Department of Rheumatology, New Cross Hospital, Wolverhampton, United Kingdom


Background Several studies have noted that while being overweight or obese increases the risk of developing a chronic health problem it may provide some protective effect against subsequent mortality risk – the obesity paradox. Obesity is indicated as a risk factor for rheumatoid arthritis (RA), which carries and increased risk of cardiovascular mortality. Few studies have investigated whether the obesity paradox is present in RA.

Objectives To investigate whether overweight and obesity carries a protective effect in RA

Methods Data from two consecutive UK based multi-centre observational studies with near identical inclusion criteria were combined: Early RA Study and Early RA Network. Between 1986 and 2013, 2701 patients were recruited from a total of 32 outpatient rheumatology clinics. Sociodemographic and standard clinical variables were recorded at baseline and then annual follow-up visits until loss to follow-up or the end of study follow up in 2013. Baseline body mass index (BMI) data was available for 2360 patients (87.4%). BMI was categorised into the four groups according to WHO definitions: underweight (BMI less than 18.5), normal (BMI between 18.5 and 25), overweight (BMI between 25 and 30) and obese (BMI greater than 30). Cox regression models were employed to estimate the relative risk (hazard) of death between each of the BMI groups. The analysis was censored after 20 years of follow up, the 31st December 2014 or death. Estimates were adjusted for potentially confounding variables measured at baseline: age, sex, social deprivation, smoking status, Charlson comorbidity score, Health Assessment Questionnaire disability, and erythrocyte sedimentation rate.

Results Average BMI at baseline visit was 26.5 (SD 5.0). Of the 2360 patients in the analysis, 40.0% had BMI scores in the normal range, 1.8% were underweight, 37.1% were overweight and 21.1% were obese. A total of 718 deaths were observed during 29819 person-years of follow-up. Kaplan-Meier survival functions for each group are shown in the figure. Underweight carried the greatest mortality risk compared to patients with normal BMI (HR=3.06, 95%CI 1.16 to 8.08). Overweight and obese patients were observed to have a reduced mortality risk compared to normal BMI (HRover=0.82, 95%CI 0.70 to 0.98; HRobese=0.77, 95%CI 0.62 to 0.95). Adjusting for potentially confounding variables the protective effect of being overweight compared to normal BMI remained significant (HR=0.75, 95%CI 0.63 to 0.90). However, after adjustment for potentially confounding variables the relative difference in morality risk for the underweight and obese groups was reduced to non-significant compared those with normal BMI (HRover=2.51, 95%CI 0.96 to 6.55; HRobese=0.97, 95%CI 0.78 to 1.22).

Conclusions Some support for the obesity paradox in RA was observed. Being overweight or obese provided either a protective effect, or at least no increase in the risk of mortality. Further research is needed to investigate the role of weight and obesity on disease progression and outcomes in RA.

Acknowledgements With thanks to the nurses and rheumatologists from both cohorts for their participation and contribution.

Disclosure of Interest None declared

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.