Patient preferences for adherence to treatment for osteoarthritis: the MEdication Decisions in Osteoarthritis Study (MEDOS)

Osteoarthritis (OA) is a musculoskeletal disease that causes chronic joint pain and reduced physical functioning. Often affecting knee joints, OA is the most common type of arthritis. By 2020, OA is predicted to become the fourth leading cause of disability globally [1].

Currently there is no known cure for OA, nor are there effective interventions to slow disease progression [2‐4]. Medication management is symptomatic, mostly with simple analgesics such as acetaminophen and non-steroidal anti-inflammatory drugs (NSAIDs) [5, 6]. Increasingly, glucosamine sulfate (GS) [7], is being used as a potential analgesic and disease-modifying agent [3, 8‐11]. In Australia, GS is considered a dietary supplement and is purchased without prescription. Unlike other OA medications, the cost of GS is not subsidised by the Australian government [12].

As occurs with most chronic conditions, adherence to arthritis medications is low [4, 13‐16]. Factors implicated in adherence to OA and other anti-rheumatic medications include dosing frequency [16], pain and self-efficacy levels [13], and physician trust [4, 17, 18]. Intentional non-adherence [19], that is deliberate decision-making about the use of OA medications, is reported in the literature. In particular, intentional under-dosing and rationing of analgesics occurs [20‐22]. Such decisions appear to be driven by factors including the fear of addiction [20], previous medication effectiveness, and the burden and illness stigma represented by increased pill loads [21]. For NSAIDs specifically, a high level of trust in the prescribing physician influences decisions [22].

Primarily, qualitative methods have been used to investigate medication decisions in OA. Although a limited number of studies have used quantitative techniques, the extent to which individuals’ trade off one treatment factor for another in decision-making about medication adherence has not been extensively studied [23‐28]. Physicians and policy makers could use such information to tailor adherence support to match the preferences of OA patients.

Discrete choice experiment (DCE) is a survey methodology that can be used to elicit preferences to quantitatively determine the relative influence of factors on decision-making with regard to medication adherence [29, 30]. Developed initially in marketing research, DCEs are used increasingly in health economics and are considered state-of-the-art in this field to elicit preferences for health services [31‐35]. To date, three DCE studies have explored patient preferences for treatment factors associated with knee and/or hip osteoarthritis [23, 24, 28]. In these studies, both efficacy and the gastrointestinal side effects of treatment significantly impacted patient choice. However, other factors potentially relevant to OA medication adherence were not consistently included. In particular, neither the cardiovascular, hepatic or renal side effects nor the chronic or intermittent treatment scheduling of OA therapy were incorporated. Additionally, preferences about acetaminophen were omitted from one study [24].

In Australia, the LEGS (L ong-term E valuation of G lucosamine S ulfate) study was a two-year, double-blind, placebo-controlled randomised clinical trial aiming to evaluate whether the dietary supplements, GS and/or chondroitin can limit or reduce structural disease progression (cartilage loss), whilst providing pain relief, in people with osteoarthritis of the knee (ClinicalTrials.gov Identifier NCT00513422). Throughout the LEGS study, as is typically the case in clinical trials, a number of trial-related factors could potentially affect treatment decisions and adherence outside of the trial setting. Firstly, as a part of the study protocol, participants were regularly encouraged to persist with the study treatments, even in the absence of knee pain. Furthermore, study treatments were mailed to participants and provided free of charge. The use of DCEs potentially helps understand the effects of such factors, including out-of-pocket costs, beyond the clinical trial setting.

The Me dication D ecisions in O steoarthritis S tudy (MEDOS) aimed to estimate the relative influence of different medication-related factors and respondent characteristics on decisions to continue medications among people with symptomatic OA.

A paper-based survey was given to all LEGS participants attending their end-of-study visit by a member of the LEGS research team; surveys were mailed to participants who had already completed end-of-study visits at MEDOS commencement. The survey was self-administered and completed either during the end-of-study visit or at a later date and returned via mail.

503 LEGS participants attended the end-of-study visits; 59 participants had already attended the end-of-study visit at MEDOS commencement and were mailed a survey. The remaining participants were given a survey at the end-of-study visit.

188 (response rate 37%) completed surveys were returned. Table 3 displays the background characteristics for all LEGS participants attending their end-of-study visit, and the subset of these participants completing the MEDOS study. With the exception of a lower proportion of people taking “when required” medications in the MEDOS group, there does not appear to be evidence for difference.

For the MEDOS participants, self-reported adherence to the study treatment throughout the LEGS study was generally high, (47% reporting 100% adherence). The lowest adherence rate reported was 75%. For other medications, 24% of participants reported intentionally stopping or altering the dose during the previous 12 months. In general, MEDOS study participants had a high level of trust in their primary care physician (median PCAS score 75 out of 100).

Using DCE, this study has assessed the factors that influence the decision to continue with a medication among a group of people with symptomatic OA. To the best of our knowledge, this is one of few DCEs assessing medication preferences nested within a clinical trial [46], and the first in an OA clinical-trial population. Such decisions underscore the concept of intentional medication non-adherence, which may influence the translation of clinical-trial results into practice.

This study has found that treatment factors, as opposed to respondent characteristics including self-reported pain levels and physical functioning, were driving adherence decisions. Preferences to continue with OA treatments were influenced by, in order of importance: the possibility of high blood pressure, heart/liver/kidney problems as side effects, out-of-pocket costs, the possibility of heartburn/reflux, or stomach ulcers as side effects, treatment schedule (i.e.: daily versus when required), mode of action (slowing OA versus symptomatic pain relief) and the possibility of drowsiness or constipation as a side effect.

Perhaps surprisingly, treatment efficacy did not significantly influence patient choices in this study. However in contrast to previous DCE studies conducted among OA patients in observational settings [23, 24, 28], this study included additional factors related to treatment schedule as well as cardiovascular, liver and renal side effects. When treatment schedule and cardiovascular, liver, and renal side effects were taken into account, as in the present study, their influence over patients’ treatment preferences then seem to dominate over considerations of treatment efficacy.

Assuming equal efficacy and GS as a disease-modifying agent, this study has found that the relative likelihood of continuing sustained-release acetaminophen and GS are positive and in contrast to NSAIDs. This disparity in predicted adherence was primarily driven by negative preferences expressed for cardiovascular, liver and renal side effects. This result may reflect an increased awareness of, or general aversion to, NSAID-related side effects through the recent and widely publicised removal of two NSAIDs, rofecoxib (2004) and lumaracoxib (2007), from the Australian market due to cardiovascular and hepatic toxicity, respectively [47, 48].

However, the reduced likelihood of continuing NSAIDs compared to acetaminophen predicted from this study seems to be at odds with the high levels of self-reported use of NSAIDs in Australia by people with OA [3]. In light of the efficacy and improved safety of acetaminophen in OA compared to NSAIDs [3, 5], our findings therefore reinforce the message that the uptake of guidelines-recommended acetaminophen in practice would benefit from ensuring that patient medication decision-making is supported through the provision of explicit risk/benefit information.

Medication adherence within a clinical trial is typically higher than in observational settings [49]. For the LEGS study, as the treatment was provided free of charge, the level of medication adherence observed in the trial was predicted to be higher than would exist outside of the trial setting. This was demonstrated in the findings of the cost-sensitivity analysis: when provided free of charge as per study protocol, the predicted continuation of GS is around 91%; however at the average current monthly price for GS ($20), this figure dropped to below 80%. Such rates of long-term adherence and their sensitivity to out-of-pocket costs will need to be factored into translation of the findings into policy and future economic evaluations.

The findings of this study must be viewed in light of its limitations. First, this study was conducted within a clinical-trial population, which may affect the generalisability of results. In particular, the self-reported adherence to medications and the proportion of participants using “when required” medications in this study was high. However, current medication use and self-reported adherence did not improve the model fit, suggesting that such preferences are formed independent of self-reported medication-taking behaviour.

Second, while discrete choice methods are widely used in health economics, an inherent limitation is that respondents are evaluating hypothetical medications; what respondents declare they will do may be quite different to what they would actually do if faced with the consequences of a choice. Forcing respondents to choose between medications may also be contrary to actual behaviour, particularly considering the over-riding influence of a prescriber’s recommendations upon patient preventive treatment decisions [22, 50]. To minimise such potential differences, measures were taken to design the hypothetical tasks to be realistic, for instance by centring levels of cost on current treatment costs and describing pain on the same scale used throughout the clinical trial. Encouragingly, in this study, trust in prescribers and the actual self-reported adherence did not influence the model results. Nonetheless, to investigate the relationship between relative preferences captured in this study and one’s absolute preference to adhere to medications, future research incorporating the influence of the prescribers recommendation, for instance by allowing respondents to opt out of the non-adherent choice [31, 50], is warranted.

Finally, as the constant term (α) was significant, the factors included in this study do not explain all of the behaviour modelled. Further work is therefore needed to clarify which other factors are being considered in adherence decisions.

Osteoarthritis is a chronic condition incurring considerable costs to most health care systems. As with any chronic condition, non-adherence to the available pharmacological treatments is a problem that has the potential to impact on population health and expenditure. In the context of a clinical trial assessing therapy effectiveness, non-adherence has the potential to derail translation into clinical practice. By recognising that a component of this health behaviour is intentional and subject to rational choices, this study has characterised the complexity of medication-taking decisions for people with symptomatic OA that may lead to intentionally non-adherent behaviour, identifying the treatment factors driving such decisions. Such factors may be amenable to intervention such as strategic pricing. The salience of medication risks in such choices highlights the importance of providing appropriate risk/benefit information upon prescription. Cost was also a strong consideration in medication-taking decisions, a finding that ought to be acknowledged when interpreting clinical trial evidence for practice. Ultimately addressing these factors may be the way forward to realising the full potential health and economic benefits from the efficacious and safe use of osteoarthritis medications.