Background Improved functional ability is among the most relevant outcomes for patients with psoriatic arthritis (PsA).1 The modified health assessment questionnaire (mHAQ) is one of the most commonly used patient reported outcome measures used in PsA and addresses the domains of disability and physical function.
Objectives We examined the change in mHAQ over one year among patients with PsA initiating a TNFi and examined predictors of a clinically meaningful improvement in disability and physical function, as measured by the mHAQ.
Methods Patients with PsA enrolled in the Corrona Registry between 2005–2013 were followed from initiation of a TNFi (etanercept, adalimumab, infliximab, certolizumab, or golimumab) to the visit closest to 12 months. Patients were required to have at least three tender or swollen joints for inclusion for this study (this is not an inclusion criteria for Corrona), mHAQ within six months prior to TNFi initiation date, and at least one follow up visit 5–13 months after TNFi initiation with mHAQ. The primary outcome was the proportion of patients with a decrease of 0.35 or more in the mHAQ score, the minimal clinically important improvement in PsA2. Predictors of mHAQ response were measured in the six months before TNFi initiation. Covariates with p-value ≤0.10 in univariable logistic regression models and ≤5% missing values were included in a multivariable model and removed individually until all remaining variables were significant (p<0.05).
Results Among 1742 TNFi initiations, 721 initiations (623 patients) met inclusion criteria. Mean age at initiation was 51.5 years (SD 12.3), 56% were female, median PsA duration was 5 years (IQR 2–11), and median baseline mHAQ was 0.375 (IQR 0.125–0.875). The mean change in mHAQ was -0.098 (SD 0.38) and median change in mHAQ was 0 (IQR: -0.25 to 0.125); 23% had a mHAQ decrease of 0.35 of more. In univariable models, college education and being married or partnered were inversely associated with mHAQ response (baseline scores were lower in patients with these characteristics) while patient global, patient pain, baseline clinical disease activity index (CDAI) and swollen joint count were positively associated with the outcome. However, in a multivariable model, only age (0.96 per year, 0.94–0.98) and baseline mHAQ score (17.7 per one point, 10.6–29.4) were associated with a decrease of 0.35 or more in the mHAQ.
Conclusions A clinically meaningful change in mHAQ occurred in 23% of patients and was strongly associated with age and baseline mHAQ score. In this real world population, there was little change in mHAQ scores over one year, despite treatment. These analyses are limited by the floor effect of the mHAQ among patients with established PsA. Despite having at least 3 tender and 3 swollen joints, many patients in this cohort had a low mHAQ at baseline which did not allow for sufficient change to meet the MCII.
Mease PJ et al. J Rheumatol 2011; 38:2461–5.
Orbai AM et al. Ann Rheum Dis 2016; Epub.
Acknowledgements This work was funded by the Corrona Research Foundation.
Disclosure of Interest A. Ogdie Grant/research support from: Pfizer (co-investigator), Consultant for: Pfizer, Novartis, L. Palmer: None declared, D. Solomon Grant/research support from: Grants to BWH from amgen, lilly, bristol myers squibb, and pfizer, Consultant for: Corrona, LLC, A. Kavanaugh: None declared, J. Greenberg Shareholder of: Corrona, LLC, Consultant for: Genentech, Janssen, Novartis and Pfizer, Eli Lilly, Employee of: Corrona, LLC, J. Curtis: None declared, L. Harrold Shareholder of: Corrona LLC, Grant/research support from: Pfizer (to UMass), Consultant for: Roche, Employee of: Corrona LLC, UMass Medical School, J. Kremer Shareholder of: Corrona, LLC, Employee of: Corrona, LLC, P. Mease: None declared