Women who made at least four healthy lifestyle choices saw their risk of developing rheumatoid arthritis reduced significantly, an analysis of Nurses’ Health Study (NHS) data found.
Adopting such habits as moderate drinking, never smoking, regular exercise, and a good diet was associated with a population-attributable risk reduction of 34% (95% CI 20%-47%), reported Jill Hahn, ScD, MS, of the Harvard T.H. Chan School of Public Health in Boston, and colleagues.
With data on nearly 250,000 women, the study — published in Arthritis Care & Research — is one of the largest and longest ever to examine the relationship between lifestyle and incident rheumatoid arthritis.
Hahn and colleagues drew on the original NHS, which began in 1976, and its second wave initiated in 1989. Both recruited female professional nurses in the U.S. and are still underway. At baseline and repeatedly during follow-up, participants completed questionnaires about their health status and related behaviors, and they allowed access to their medical records as well.
Diet was assessed through questions about their intake of more than 130 types of foods, allowing the investigators to gauge dietary healthiness via the Alternate Healthy Eating Index (AHEI). Exercise habits were summarized into metabolic equivalent of task (MET) hours per week. Other lifestyle factors examined in the study included self-reported average alcohol intake and smoking history. Participants’ body mass index (BMI) was also used as an indicator of lifestyle healthiness. All factors together were entered into a categorical “healthy lifestyle index score” or HLIS.
For statistical purposes, Hahn and colleagues analyzed these factors as binary variables, with the following defined as “healthy”:
- ≥19 MET hours/week
- BMI 18.5-24.9
- Upper 40th percentile of AHEI
- Never smoking
- Daily alcohol intake of 5-15 g (equivalent to about one drink)
Choice of this last item as the healthiest approach to drinking was based on previous studies indicating that total abstinence, as well as greater daily intake, were both associated with adverse health status relative to this moderate level.
Over a mean 24 years of follow-up, rheumatoid arthritis developed in 1,219 participants, with roughly two-thirds seropositive for rheumatoid factor (RF).
Each HLIS increment was associated with a 14% decrease in incident rheumatoid arthritis risk (95% CI 10%-18%), which varied little by whether participants were RF-positive. Women in the highest categories for each of the five HLIS components had just 40% of the risk seen in those in the lowest category for each.
Taken individually, the biggest contributors to the apparent protective effect, as reflected in population-attributable risk, were moderate drinking, smoking abstinence, and normal BMI. By the same token, the least important for rheumatoid arthritis were diet and exercise.
Hahn and colleagues acknowledged the potential for reverse causation “if preclinical rheumatoid arthritis caused less physical activity or led to an increase in BMI.” But a sub-analysis in which they “lagged” physical activity by 4 years did not yield notably different results, arguing against reverse causation.
Limitations to the analysis included the predominantly white NHS sample and reliance on self-report for much of the lifestyle data.
Still, the investigators concluded that “the confluence of modifiable lifestyle factors represents something of a paradigm shift in thinking about rheumatoid arthritis and autoimmune disease risk.”
Moreover, “promotion of multiple healthy behaviors to minimize risk is an important message for the general population and in particular those at risk by virtue of family history,” they wrote.
