Abstract
Objective. The objectives of this study were to examine the reasons patients give for nonadherence to allopurinol and to examine differences in intentional nonadherence for patients who did and did not achieve serum urate (SU) levels at treatment target.
Methods. Sixty-nine men with gout attending rheumatology clinics, all prescribed allopurinol for ≥ 6 months, completed the Intentional Non-Adherence Scale (INAS). Differences in the types of intentional nonadherence were analyzed between those who did and did not achieve SU at treatment target (< 0.36 mmol/L, 6 mg/dL).
Results. The most frequently endorsed reasons for not taking their urate-lowering therapies (ULT) were because participants wanted to lead a normal life (23%) or think of themselves as a healthy person again (20%). Patients also reported not taking allopurinol as a way of testing if they really needed it (22%). Participants with SU above target endorsed significantly more INAS items as reasons for not taking their medication, had more medicine-related concerns, and were more likely to give Testing treatment as a reason for nonadherence. Participants who were younger, single, and non–New Zealand European also endorsed more reasons for not taking their allopurinol.
Conclusion. The major reasons behind the patient’s decision not to take allopurinol relate to the desire to lead a normal life and the strategy of testing the treatment to see if they could reduce the dose without getting symptoms. These results provide some potentially modifiable targets for adherence interventions and some recommendations to clinicians about how to reframe ULT for patients in order to improve adherence.
Gout is a chronic disease of monosodium urate crystal deposition. Urate-lowering therapy (ULT) is indicated for patients with recurrent gout flares, joint damage due to gout, and tophaceous gout.1 Allopurinol is recommended as first-line ULT, and is the most widely used treatment, accounting for > 90% of all ULT used in the US.2 The benefits of ULT are realized with long-term, continuous therapy to maintain serum urate (SU) < 0.36 mmol/L (6 mg/dL). However, adherence rates for ULT are very low3 even in comparison to other chronic illnesses.4 Typically, adherence steadily declines in patients continuing treatment after their initial prescription and less than half of patients take allopurinol as prescribed at 12 months.5,6 The low rates of adherence to ULT have led to a call to better understand patient-related factors that lead to intentional nonadherence to gout therapy.7
Early research in medication adherence largely focused on unintentional factors, such as forgetting or uncertainty about the treatment regimen.8 Interventions based on this approach, including many apps, have concentrated on reminders, but because most nonadherence is intentional,9,10 these often have little effect on improving adherence.11 Intentional nonadherence describes the process by which patients decide not to take their medication based on specific perceptions about their condition or treatment. Intentional nonadherence is a new area in adherence research and a new measure, the Intentional Non-Adherence Scale (INAS), has been developed as a way of ascertaining patients’ reasons for nonadherence behaviour.12
Currently there is a lack of research on what beliefs and perceptions drive people with gout to not adhere to ULT. It is frequently assumed that nonadherence is driven by side effects.7 Patients are often asked about medication side effects and clinicians often target these beliefs to improve adherence; however, there is a wide range of reasons that could influence nonadherence. A better understanding of the drivers of intentional nonadherence with ULT may allow the development of interventions designed to change these beliefs in order to improve adherence behavior.
This study had 3 aims designed to improve our current understanding of intentional nonadherence in patients with gout who take allopurinol. First, we utilized the INAS to find out what reasons patients with gout give for not taking allopurinol. Second, we investigated the differences in the types of intentional nonadherence between those who did and did not achieve SU levels at treatment target (< 0.36 mmol/L, 6mg/dL). SU level is an important biomarker of gout disease and an acceptable objective measure of adherence.13 Last, we explored the relationships between intentional nonadherence and demographic factors in patients with gout.
METHODS
Patients and methods. This was a cross sectional study of 69 men with gout. The participants were recruited from rheumatology clinics between September 2019 and March 2020. Inclusion criteria were as follows: (1) age ≥ 18 years; (2) a rheumatologist-confirmed diagnosis of gout; (3) allopurinol prescription ≥ 6 months; and (4) English-speaking. The New Zealand (NZ) Health and Disability Ethics Committee approved this study (ref. HDEC19/CEN/148) and all patients provided written informed consent. Participants completed demographic and clinical data forms and study questionnaires. SU result was obtained through medical record review.
INAS. This 22-item scale assesses the potential reasons behind intentional nonadherence behavior.12 The scale is prefaced with the following instructions: “People have different experiences when taking medication and use their medications in ways that suit them. Sometimes people forget or decide not to take their medication for various reasons. We are interested in your personal views and experiences of your prescribed medication regime and the way you use your medications. Listed below are some of the reasons why people sometimes stop taking their medications. We would like to know how often each of the following statements is true for you in the past 6 months”. The scale asks patients whether they have not taken their medicine due to a list of 22 reasons scored on a 5-point Likert scale from strongly agree (5) to strongly disagree (1).
The INAS comprises 4 subscales: (1) Resisting illness, which links the decision not to take treatment with not wanting to be reminded of one’s illness and the desire to feel normal and healthy (eg, “Because it reminds me I have an illness”); (2) Testing treatment, which assesses the individual’s reasons for not taking treatment based on the person’s attempts to see if they can get away with taking less or no treatment at all (eg, “To see if I really need it”); (3) Drug-specific concerns, such as side effects and becoming dependent on the medicine (eg, “Because I don’t like the side effects”); and (4) General sensitivity to medicines, which consists of a set of beliefs about how they are personally affected by medicine and need to control the medicine intake to minimize harm (eg, “To give my body a rest from the medicine,” and “I don’t like medicines accumulating in my body”). All items and the subscales are shown in Figure 1. Scores on the total INAS range from 22 to 110, with higher scores indicating more motives for intentional nonadherence behavior. Each of these subscales shows acceptable internal consistency, with Cronbach α ranging from 0.91 to 0.93 in the current sample; the INAS total was 0.95.
Percentage of respondents agreeing or strongly agreeing with INAS items about why they did not take their allopurinol. INAS: Intentional Non-Adherence Scale.
Data analysis. All statistical analyses were performed using SPSS version 25.0 (IBM Corp.). Medians with ranges and percentages were used to describe the clinical characteristics of participants. As the INAS scores were not normally distributed, Mann-Whitney U tests were conducted to investigate differences in INAS scores for participants categorized as adherent vs nonadherent on the basis of achieving SU target. Spearman correlation was also used to describe the associations between variables. All tests were 2-tailed and a significance level of 0.05 was used to determine significance for all analyses.
RESULTS
Characteristics of the study population. A total of 69 men with gout with an average age of 63.5 years were included in the study. Most patients were NZ European (66%), married (67%), and had university education or higher (67%). On average, the participants had been taking allopurinol for 8 years (Table 1). Using the SU level criterion for treatment adherence, 46 patients were classified as adherent and 23 as nonadherent.
Characteristics of the study population.
Reasons behind nonadherence. To look at the main reasons participants with gout gave for not taking their medicine, we ranked each INAS item by the percentage of patients agreeing (agree, strongly agree) with each statement, and these are shown in Figure 1. The graph indicates that the top 4 reasons why patients with gout do not take their allopurinol are comprising 2 Resisting illness items including “because I want to lead a normal life again” (23%) and “because I want to think of myself as a healthy person again” (20%), and 2 Testing treatment items including “to see if I really need it” (22%) and “to see if I can do without it” (22%). The next 3 items were “because I don’t like the side effects” (17%), “because I think the drug might become less effective over time” (16%), and “because I worry about becoming dependent on my medicine” (16%). The Medicine sensitivity items formed most of the middle-ranked items such as “because the medicine is harsh on my body” (13%). Other Drug-specific concerns made up the less-endorsed items, such as “because I think I am on too high a dose” (6%) and “because I don’t think the treatment is worth it” (1%).
Differences between participants who did and did not achieve SU levels at target. Next, we looked at the differences on overall INAS scores and INAS subscales according to SU target. We found that, as expected, respondents who did not achieve target SU had significantly higher total INAS scores, as well as significantly higher Testing treatment and Drug-specific concern scores (Table 2).
INAS (total and subscale) scores for participants who did and did not achieve target SU (< 0.3 6mmol/L, 6 mg/dL)a.
Associations between INAS scores and demographic features. The third aim was to explore any demographic differences in terms of INAS scores. The data showed that age was significantly correlated with the INAS Testing treatment subscale, indicating that younger patients were more likely to test their treatment and see if they can get away with taking less or no treatment (rs = –0.27, P = 0.02). However, there were no significant correlations between age and the INAS total score (rs = –0.20, P = 0.10) or the other INAS subscales. There were also no significant correlations between time on allopurinol and the total INAS score or any of the INAS subscales (data not shown). NZ Europeans had significantly lower total INAS scores and all 4 subscale scores than non-NZ European ethnicities. We found that participants who were single had significantly higher total INAS scores and Testing treatment subscale scores compared to those in a relationship (Table 3). Education or employment status showed no effects for the total INAS score or INAS subscales.
Differences between INAS total and INAS subscales in demographic subsamples.
DISCUSSION
The INAS allowed us to investigate in more detail the reasons from the patients’ perspective that lie behind not taking their urate-lowering medication. Looking at the items with the highest level of endorsement suggests that the major reasons behind the decision not to take allopurinol relate to wanting to lead a normal life and patients wishing to think of themselves as healthy again. Another important motivation is the strategy of testing treatment. Here, nonadherence represents a deliberate effort to see if the patient can get away with taking less or none of their allopurinol medication without their painful symptoms returning. Two Drug-specific concerns, side effects and the belief that allopurinol might become less effective over time, were the next frequently endorsed items. As expected, INAS scores differed significantly between those who did and did not achieve target SU, with those not at target endorsing more INAS items. Those not at target also scored significantly higher in Testing treatment and Drug-specific concerns. INAS scores were also higher in patients who were younger, of non-NZ European ethnicities, and not in a current relationship.
The results demonstrate the importance of the patients’ view of their illness and treatment in the long-term management of gout.14,15 Intentional nonadherence generally decreased with age and the strategy of testing treatment was more common in younger age groups, where the disease diagnosis may not fit comfortably with the common illness model of gout being typically that of older men.16 Development of gout later in life is more normative and has been shown to be less stigmatizing at an older age.16 This mismatch between illness perception and treatment can make the patient feel uncertain about the diagnosis and long-term ULT.17 Previous studies in other illnesses show that illness beliefs can be modifiable by targeted interventions and this can lead to improved outcomes.18,19,20
Our results provide some indication to clinicians on how ULT may be framed for patients in order to improve adherence. The data from this study suggest that framing ULT as a way of correcting an unhealthy imbalance that will allow a return to normal activity and lifestyle may be a useful strategy when initiating treatment. Further, an early discussion about the drawbacks of testing treatment by reducing medication or the problem of basing medication-taking on symptoms rather than SU level may also be helpful. Additionally, when patients have doubts about effectiveness and worries about their prescribed medication, they may become nonadherent when the perceived risk of taking medication outweighs its perceived benefits over time.21 Therefore, it may be beneficial to correct the concern that allopurinol will become less effective over time.
The study has a number of strengths including the use of a new measure that allowed closer examination of the reasons behind patients’ nonadherence to allopurinol and the use of an objective marker of adherence. Whereas previous work on nonadherence has concentrated on demographic and clinical factors associated with it,3,4,5,6 the current study focus is on psychological factors that could be potentially modifiable in an intervention. However, the study sample size was modest and consisted only of men who were recruited from a hospital clinic; these may limit generalizability and should be considered when interpreting findings. While there is a strong relationship between allopurinol adherence and achieving target SU levels, it should be acknowledged that there may be reasons for low levels other than nonadherence, such as inadequate dosage.22 Although self-report, pill counts, and interviews typically provide other measures of nonadherence in gout,23 SU level does have the advantage of assessing a widely recommended clinical outcome.
The patients in the study were also more adherent than the nonadherence rates identified in previous research.3,4,5,6 This may be due to nonadherent patients being less likely to participate in such research or a Hawthorne effect in reaction to participation in the study. Nonadherent patients may include both patients who do not fill a prescription and those who collect allopurinol but do not take it regularly. It should also be noted that SU levels not reaching target could be due to inadequate dosing as well as patient nonadherence.24 This possibility should be addressed in future research with the INAS. It would be also valuable to examine the relationship between intentional nonadherence and other clinical data, such as comorbidities, the presence of tophi, and the prescription of other medications.
In conclusion, our study highlights a potentially effective new approach to decrease ULT nonadherence, which is a prevalent problem for long-term management of gout. Examining the motivations that patients provide for not taking their medication has identified some potential targets for interventions in patients with gout who are finding it difficult to adhere to ULT. These include reframing the treatment as a way of returning to feeling normal again and identifying for patients the potential difficulties of using symptoms as a way of testing the dose of ULT.25 Further research is needed to turn these insights into workable and scalable interventions that could provide improvements in reducing nonadherence.
ACKNOWLEDGMENT
We acknowledge the participants and staff of Greenlane Clinical Centre, Auckland for their help with the study.
Footnotes
This research was funded by The University of Auckland. The sources of funding for this study played no role in the study’s design, conduct, or reporting.
ND reports grants and personal fees from AstraZeneca; grants from Amgen; and personal fees from Dyve BioSciences, Hengrui, Selecta, Arthrosi, Horizon, AbbVie, Janssen, PK Med, and JW Pharmaceuticals, outside the submitted work. The remaining authors declare no conflicts of interest relevant to this article.
- Accepted for publication February 3, 2022.
- Copyright © 2022 by the Journal of Rheumatology
