Background/Purpose:

Biologic therapy is now a cornerstone of RA management, but treatment response is variable, failing in up to 40% of patients. To direct appropriate treatments towards those patients most likely to benefit, reliable predictors of treatment response must be identified. Response is usually measured using DAS28, using clinical, biochemical, and patient-reported data. It has recently been shown that psychological factors influence individual DAS28 components, notably tender joint count and patient visual analogue scale (VAS) (Arthritis Care Res 2014, 66:861-868).

We hypothesised that psychological factors also influence change in DAS28 after biologic therapy, and that individual components of DAS28 would be affected disproportionately.

Methods:

Data from the BRAGGSS cohort was used: a multicentre, observational study of RA patients with severe active disease according to NICE criteria and undergoing biologic therapy.

Clinical and demographic data were collected from patients about to commence a biologic drug, then at three and six months. Illness beliefs, treatment beliefs, and mood were assessed at the same timepoints using validated measures: the Brief Illness Perceptions Questionnaire (B-IPQ), the Beliefs about Medications Questionnaire (BMQ), and the Hospital Anxiety and Depression Scale (HADS), respectively.

The relationships between psychological factors at baseline and 3- and 6-month change in DAS28 components were analysed using linear regression. 

Results:

At baseline, 75.5% of 1847 patients were female, with a mean age of 58.4, DAS28 of 5.74 and HAQ of 1.74. 73.1% patients received anti-TNF drugs. There were significant associations (p<0.05) between nine of the 12 psychological factors with patient VAS, and six psychological factors with tender joint count. Three B-IPQ items and HADS depression were associated with CRP (p<0.05). At six months, no associations between baseline B-IPQ scores and changes in composite DAS28 were observed but BMQ and HADS anxiety scores showed a significant association (p<0.05). Interestingly, an association was found between baseline illness beliefs and changes in CRP, indicating that with lower baseline psychological scores inflammation was more reduced; for example, personal control (β=-0.786, p=0.05), treatment control (β=-1.48, p=0.005) and coherence (β=-1.623, p=<0.001). This was true at both three and six months follow-up. 

Conclusion:

This study confirms earlier findings suggesting that baseline psychological factors affect baseline DAS28, especially patient-reported measures. Interestingly, psychological factors were significantly correlated with change in CRP at both 3 and 6 months, suggesting psychological state may be a driver of inflammation in RA and that the most appropriate approach for some patients may include addressing psychological wellbeing.

Overall, psychological factors were associated disproportionately with individual components of DAS28, but not with change in composite DAS28 score. This illustrates that reporting of individual DAS28 components may highlight individual patients’ differing needs despite superficially similar composite scores.

« Back to 2015 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/investigating-psychological-predictors-of-biologic-treatment-response-in-patients-with-severe-active-rheumatoid-arthritis/