Background/Purpose:  Methotrexate (MTX) is a first line therapy for Rheumatoid Arthritis (RA). Treatment response to MTX is not universal, and nonadherence may partially explain poor treatment response to MTX, therefore it is imperative to investigate adherence to MTX specifically.  Previous systematic reviews have evaluated adherence to all DMARDs. The aims of this systematic review were:  1) to summarise existing rates of adherence to MTX, 2) to identify predictors of adherence and 3) assess the association between non-adherence with patient outcomes.

Methods: A systematic search of papers published between January 1980 to 2014 was conducted in PubMed, PsycINFO, EMbase and CINAHL databases. Studies were eligible for inclusion if; 1) MTX was used as monotherapy or in combination with other therapies, 2) MTX was used in an RA or inflammatory polyarthritis population, 3) adherence was defined and measured as the extent to which patients followed their MTX regimen during the period of prescription, and 4) it was an original piece of research. Papers were reviewed by two researchers and consensus agreed with a third. A quality assessment (QA) tool formally assessed each included article.

Results:  In total, ten studies met the inclusion criteria and eight were evaluated high quality using the QA tool. Rates of adherence ranged widely from 59% to 107%, and exposed differences in definitions of adherence, study methodologies and sample heterogeneity. The methods used to assess adherence included; Medication Event Monitoring Systems (n=2), Pharmacy Refill (n=5), validated self-report questionnaire (n=2) and 7 day diary (n=1) (Table 1). Twenty-one potential predictors of MTX adherence were identified; the strongest evidence occurred for beliefs in the necessity and efficacy of MTX, absence of low mood, mild disease and MTX as a monotherapy. Disease activity was the only patient outcome where the effect of nonadherence was assessed. Three studies tested this association and each found nonadherence associated with poor treatment response.

Conclusion:  Psychological factors were the strongest predictors of adherence rates in this first systematic review specific to MTX. It was the first to include synthesis of evidence on patient outcomes and show nonadherence to MTX affects treatment response in RA.   

Table 1. Comparison of MTX rates of adherence across studies

Author	QA score	High/ low quality	Adherence method used	Definition of MTX adherence	n	MTX adherence	95% CI/SD
MEMS
Waimann et al. (2013) [19]	15	High	MEMS	%  of correctly taken doses	76	63%	20%
de Klerk et al. (2003) [22]	12	Low	MEMS	%  of correctly taken doses	23	107%	98-117
Pharmacy refill
Cannon et al.  (2010) [18]	15	High	MPR	=>80% prescriptions filled	85	80%	NP
	15	High	MPR	=> 80% prescriptions filled	370	85%	NP
de Thurah et al. (2010) [20]	14	High	CMG	% of treatment gaps (reversed)	941	87.7%	86.8-88.5
Grijalva et al. (2010) [17]	9	High	MPR	%  of  prescription filled	NP	59%	31-82
Harley, Frytak & Tandon (2003) [16]	8	High	MPR	≥ 80% prescriptions filled	1668	63.70%	23.8 – 102
Grijalva et al. (2007) [15]	8	High	MPR	% of prescriptions filled excluding the last prescription	2933	80%	NP
Self report
de Thurah et al (2010)a [21]	14	High	CQ-R	CQ-R score >25th percentile	85	BL 23.5%	NP
					65	9 mo. 23.1%	NP
Contreras-Yanezet et al. (2010) [23]	11	Low	7 day DRR	% of correct doses taken across 3 time points	10	78%	NP
Salt & Frazier (2011) [24]	9	Low	MARS	MARS score >39	77	92.10%	NP
(CQ-R = Compliance Questionnaire-Rheumatology, DRR = Drug Record Registry, MARS = Medication Adherence Revised Scale, MPR = Medication Possession Ratio, CMG = Continuous Medication Gap, NP= information not presented)

Table 2. Associations with patient outcomes

Author	Sample size	Predictor	Outcome	Statistical test	Unadjusted Strength of effect (95% CI)	P	Adjusted Strength of effect (95% CI)	P
Cannon et al. (2011) [17]:
Full cohort	455 (71 non –adherent, 384 adherent)	MPR ≥ 80%	Mean difference in DAS28	Linear regression	-0.34 (-0.68, -0.06)	< 0.05	-0.37 (-0.67, -0.07)	< 0.05
First time user cohort	85 (17 non adherent 68 adherent)	MPR ≥ 80%	Mean difference in DAS28	Linear regression	-0.54 (-1.18, 0.11)	NS	-0.40( -1.11, 0.30)	NS
Established cohort	370 (54 non adherent 316 adherent)	MPR ≥ 80%	Mean difference in DAS28	Linear regression	-0.38 (-0.67, -0.05)	< 0.05	-0.37 (-0.72, -0.02)	< 0.05
Waimann,C.A.   et al. (2013) [19]	102	% of prescribed DMARD doses takena	DAS28 at 2 years	Linear regression	-0.02 (-0.03, -0.001)	NP	-0.2	0.03
Contreras-Yanez,I. Et al. (2010) [23]	68 (54 maintained remission 14 disease flare)	MPR ≥ 80%b	Maintained remission (DAS28 < 2.4)	Logistic regression	NP	NP	NP	< 0.001
a includes sDMARDS and bDMARDS. b includes other sDMARDS. NP = not presented. RF = rheumatoid factor. CCP = cyclic citrullinated peptide

---

« Back to 2014 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/psychological-factors-predict-adherence-to-methotrexate-mtx-in-rheumatoid-arthritis-ra-findings-from-a-systematic-review-of-rates-predictors-and-associations-with-patient-outcomes/