Background Obesity is a known risk factor for knee OA and to a lesser extent for hip OA. There is moderate evidence that obesity is a risk factor for radiographic hand OA. Whether high BMI may be a risk factor for symptomatic hand OA, and whether BMI over time is different for persons with hand OA compared to healthy, is to our knowledge, unknown.
Objectives The aim of this prospective cohort study was to investigate whether BMI is a risk factor for hand OA.
Methods The project is a part of the “Musculoskeletal pain in Ullensaker Study” (MUSt). Participants aged 20-59 years in 1990 received postal questionnaires in 1990, 1994, 2004 and 2010 (N=1039). Persons with self-reported knee, hip and/or hand OA in 2010 were invited to participate at a clinical examination. On the basis of a clinical evaluation by a physician, a classification of hand OA was made according to the ACR-criteria (N=59). BMI was calculated based on self-reported height and weight at the four different measurement times. BMI over time for the persons with hand OA was compared to the BMI over time for the persons with no self-reported OA (N=980), with the use of a Generalized Estimating Equation analysis, adjusted for age, sex and education level (primary/upper secondary school vs. >1 year college/university).
Results The mean age was 37 (10.4) years of age in 1990 and 54% were females. Persons who were diagnosed with hand OA in 2010 had higher BMI in 1990 than persons who were not diagnosed (B, 1.12, 95% CI, 0.16 to 2.07). Although BMI increased consistently for the whole cohort the next 20 years (B, 0.92, 95% CI, 0.84 to 0.99 for 1 unit = 6.7 years increase), there was a significant interaction between OA status and time (p=0.007). In 1994, 2004 and 2010, persons with hand OA had lower BMI than those without, and the difference between groups increased for each year (B, -0.09, 95% CI, -0.64 to 0.45 in 1994: B, -3.12, 95% CI, -5.12 to -1.13 in 2004: B, -4.94, 95% CI, -7.97 to -1.92 in 2010). These differences could not be explained by age at baseline. Sex was a confounder of the relationship (for females: B, -1.28, 95% CI, -1.70 to -0.85) but education had no influence on the estimates and was taken out of the model.
Conclusions Having symptomatic hand OA was related to higher BMI early in life, and lower BMI later in life, as compared to healthy, independent of age and education level. The study indicates that the relationship between BMI and hand OA is complex. However, high BMI related to poor lifestyle was relatively uncommon in Norway at the two first measurement points, which may indicate that heredity may play a greater role for high BMI and hand OA than the role of later lifestyle-related BMI alone.
Disclosure of Interest None Declared