Article Text

THU0168 Patients' positive beliefs and certainty predict methotrexate adherence in a rheumatoid arthritis cohort: the rams study
  1. HF Hope1,
  2. K Hyrich2,
  3. F Varney3,
  4. S Wilkes3,
  5. SM Verstappen2,
  6. L Cordingley4
  1. 1NIHR Manchester Musculoskeletal Biomedical Research Unit, Centre for Musculoskeletal Research
  2. 2Arthritis Research UK Centre for Epidemiology
  3. 3Division of Psychology and Mental Health
  4. 4Centre for Musculoskeletal Research and Dermatological Science, University of Manchester, Manchester, United Kingdom


Background Living with RA confers substantial uncertainty about the long-term outcomes of the disease and the occurrence of possible medication associated adverse events (ADEs). This latter uncertainty may be particularly apparent when new therapies are initiated and reduce patient adherence.

Objectives The aim of this study was to evaluate whether patients' negative and/or uncertain illness beliefs during methotrexate (MTX) therapy are associated with adherence.

Methods This was a sequential mixed-methods design study using data from the Rheumatoid Arthritis Medications Study (RAMS). RAMS is a 12 month observational study in the UK recruiting patients with RA commencing MTX for the first time. Clinical and demographic data were collected at baseline and patients were asked to complete a weekly diary recording MTX intake (adherence) and reasons for not taking MTX. In addition there was a free text section in the diary where patients could comment about any aspect of their disease or care. Six month diary data were used for the purpose of this study.Patients were categorised as non-adherent if the proportion of adherent weeks was <90%.

Phase 1:Using a random sample (n=50/417) of patient diaries with free text comments a coding system was developed to categorise illness events and beliefs contained in these data. Inter-rater reliabilities for codes rated by three judges were calculated using intraclass correlation coefficients (ICC) and unreliable codes (ICC<0.6) dropped.

Phase 2:179 of 200 diaries from adherent and non-adherent individuals were randomly selected, coded and categorised into illness belief profiles (IBPs) by three researchers blind to adherence data.Univariate logistic regression analyses adjusted for age and gender were used to investigate the association between IBPs and MTX non-adherence behaviour (<90% adherence).

Results Phase 1: ten codes with ICCs ranging from 0.6–1.0 were used to create three IBPs (Table):“Positive & Certain” (PC), “Negative & Certain” (NC) and “Negative and Uncertain” (NU) (Fig.).

Phase II, the median age of the sample was 62 [51.8–65.6] years, 67% were women and the median disease activity score was 4.3 [3.4–5.2]. Being PC lowered the odds of non-adherence (OR 0.32, 95% CI 0.12–0.85), being NU increased the odds of non-adherence (OR 2.7, 95% CI 1.0–7.0), but being NC didn't associate with non-adherence during the first six months of therapy (OR 0.98, 95% CI 0.31–3.2).

Table 1.

The reliability (ICC) of codes that contribute to the IBPs

Conclusions People who are uncertain about how to attribute illness events are less likely to adhere within the first six months of starting MTX therapy. Encouraging patients to actively monitor their progress with therapy and providing them with support to understand likely effects of MTX may help optimise DMARD use.

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.