Opioid dependence is associated with high rates of psychiatric and physical illness, with significant impacts on the wider community through health costs, crime, blood-borne viruses and family disruption. Reference Morrow and Costello1,Reference Walley, White, Kushel, Song and Tulsky2 There are approximately 140 000 people on opioid substitution treatment in the UK, with a similar number out of treatment. Buprenorphine and methadone are both evidenced pharmacotherapies that have been consistently shown to improve health outcomes and reduce mortality in those with opioid dependence. Reference Faggiano, Vigna-Taglianti, Versino and Lemma3,Reference Mattick, Kimber, Breen and Davoli4 Being in treatment is associated with significant improvements in health, psychological functioning, as well as reductions in heroin use, other drug use and crime. Reference Darke, Ross and Teesson5–Reference Ward, Mattick and Hall7 Longer retention in treatment and higher doses of medication are generally associated with better outcomes. Reference Faggiano, Vigna-Taglianti, Versino and Lemma3 Unfortunately, retention in treatment for opioid dependence tends to be poor (approximately 50% at 6 months). Reference Bell, Burrell, Indig and Gilmour8
A growing body of evidence now suggests that variables such as the knowledge, beliefs, perceptions and attitudes of opioid users towards opioid pharmacotherapy may influence therapeutic outcomes and early treatment drop out. Reference Treloar, Abelson, Cao, Brener, Kippax and Schultz9–Reference Pinto, Rumball and Holland13 A recent Australian study of opioid users’ knowledge and attitudes about opioid treatment identified significant gaps in knowledge about many aspects of treatment including medication, treatment benefits, risks and perceived side-effects. Reference Winstock, Lea, Maden and Bath14 These data corroborated previous findings that revealed opioid users’ dissatisfaction with the information provided by their doctors regarding medications’ pharmacological properties and treatment options. Reference Brown, Wright and Christensen15 Research conducted with other psychiatric populations has identified that patients’ lack of information about their treatment is not exclusive to the field of substance misuse. British studies carried out with individuals receiving antipsychotic pharmacotherapy showed that the knowledge that they possessed about their medication was insufficient, namely in terms of side-effects. Reference Brown, Wright and Christensen15–Reference Crossley and Withers17 These individuals appeared to be ambivalent towards their treatment and expressed a wish for more information on prescribed medications. Reference Brown, Wright and Christensen15–Reference Crossley and Withers17
Although adherence to and retention in treatment are important modifiers of health outcomes for many chronic conditions, Reference McLellan, Lewis, O'Brien and Kleber18 the potential role of patient education in optimising treatment outcomes has to date received little attention in the field of opioid dependence. In other areas of medicine, however, patient education has been shown to improve treatment adherence and knowledge. Reference Brown, Wright and Christensen15,Reference Hill, Bird and Johnson19
Our study aimed to assess the mean knowledge about opioid substitution treatment among patients at treatment centres in South London. Additionally, the level of information provision to these individuals was also estimated, aiming to comprehend to what extent these educational materials/interventions are associated with knowledge about treatment.
Method
Participants were recruited from four community drug and alcohol teams at South London and the Maudsley National Health Service (NHS) Trust. At the time of recruitment there were, altogether, 1159 people in treatment at these services. To collect data, the researcher attended each of the services for a period of five consecutive days, comprising a study period of approximately 1 month (from 2 November to 4 December 2009). During this period, the researcher approached all patients in the services’ common areas (e.g. waiting room, drop-in groups) and offered voluntary entry into the study if they were receiving either methadone or buprenorphine at these units and were over 18 years old. Overall, a total of 154 patients were approached (13% of those in treatment at the four services). Of these 154 individuals, 118 (76.3%) agreed to participate in the study. Ethical approval was received from the joint South London and the Maudsley and Institute of Psychiatry research ethics committee.
This cross-sectional study used a self-report knowledge scale adapted from an instrument developed in Australia. Reference Winstock, Lea and Fettell20,Reference Winstock, Lea, Maiden and Bath21 This scale contains 34 true/false items covering four core domains of opioid substitution medications and opioid dependence: knowledge about opioid substitution treatment, overdose, pregnancy and blood-borne viruses (see online supplement to this paper). The knowledge scale used in this study has a good overall internal consistency (Cronbach's α = 0.84), confirming its adequacy for the variables under study. Data regarding demographics, previous and current opioid treatment history and the range/nature of information provision about treatment were also collected.
Statistical analyses
The primary outcome for this study was the mean knowledge score obtained in the 34-item scale, which represented the arithmetic mean of correct answers provided by each participant. As a secondary outcome, we asked participants the mean number of correct answers they thought they had achieved out of 34. Data analysis was performed using SPSS 17.0 in MS Windows Vista, utilising parametric methods, namely, independent t-test (knowledge about treatment in participants receiving methadone and buprenorphine; knowledge about treatment in males and females) and one-way ANOVA (knowledge about treatment and previous treatment experiences). Non-parametric methods (e.g. chi-squared) were also used to compare the differences between participants in both treatment options regarding sociodemographics (e.g. employment status of those receiving methadone v. buprenorphine) and treatment variables (e.g. experience of previous treatments in participants receiving methadone v. buprenorphine). Linear regression was also performed in order to explore the potential of relevant variables in predicting knowledge about treatment.
Results
Of 118 participants, 89 (67.8%) were White British and 96 (81.6%) were male. The mean age of the participants was 40.7 years (s.d. = 7.6, range 24–61). Thirty-two (27.6%) were unemployed and 65 (56%) were receiving pension/benefits. Forty-seven (40.2%) reported not having qualifications.
Treatment characteristics
In total 101 (85.6%) participants were receiving treatment with methadone and 17 (14.4%) with buprenorphine, the majority at community pharmacies under supervision. Seventy-three (64.4%) individuals had experience of previous treatment with methadone, buprenorphine or both (Table 1). There were no significant differences between demographic and treatment variables across participants prescribed with methadone and buprenorphine.
Table 1 Current and previous treatment history (n = 118)
| n (%) | |
|---|---|
| Previous treatment episodes | |
| None | 45 (35.6) |
| With methadone | 46 (39) |
| With buprenorphine | 7 (7.6) |
| With methadone and buprenorphine | 20 (17.8) |
| Current treatment episode, months | |
| 0–5 | 35 (30.4) |
| 6–24 | 28 (24.3) |
| 25–48 | 25 (21.7) |
| 49+ | 27 (23.5) |
Overall knowledge about treatment for opioid dependence
Each correct answer on the 34-item knowledge scale resulted in a score of one point. Out of a maximum score of 34, the mean knowledge score obtained was 14.74 (s.d. = 4.7, range 2–25). There was no significant difference in the mean score between those receiving treatment with methadone (14.89) and buprenorphine (13.82, t = 0.84, P = 0.41). Higher knowledge scores were significantly associated with more experience of treatment, with those in the first treatment episode scoring a mean of 13.07 and those with previous treatments with methadone and buprenorphine scoring, on average, 16.25 (F = 3.140, P = 0.03). There was also a significant difference between the mean knowledge score in both genders, with the female group more likely to score higher than their male counterparts, respectively 16.59 v. 14.31 (t = –2.01, P = 0.04). At the end of the questionnaire participants were asked to estimate the number of correct answers they thought they had achieved out of 34. The mean predicted score among participants was 23.12 (s.d. = 6.27, range 2–34), which represents expectations of obtaining the correct answers to more than 60% of the items. This compares with an average correct response rate of 43.4%, representing an overestimation of their true knowledge by a mean of 38.2%.
Knowledge about treatment and information provision by topic
Items in this questionnaire were clustered into six thematic subscales, each representing different topics associated with treatment (Table 2). Most participants (79.7%) reported being informed about at least one of these topics, with 17.8% reporting information provision on all the aspects covered by this questionnaire.
Table 2 Mean knowledge by information provision in the respective topic
| Knowledge subscales | Knowledge score Mean (s.d.) |
Range | Proportion of correct answers, % |
Patients provided with information on the topic, % |
|---|---|---|---|---|
| How does medication work | 3.76 (1.80) | 0–8 | 47.0 | 76.1 |
| Starting treatment/side-effects | 3.26 (1.70) | 0–8 | 40.8 | 55.6 |
| Hepatitis | 3.17 (1.51) | 0–6 | 52.8 | 57.3 |
| Overdose | 2.67 (1.44) | 0–6 | 44.5 | 56.1 |
| Risks of treatment drop out | 2.07 (0.96) | 0–3 | 69.0 | 58.1 |
| Pregnancy and contraception | 1.20 (0.97) | 0–4 | 30.0 | 42.4 |
Variables potentially associated with knowledge about treatment
A linear regression analysis was conducted to explore the role of some variables that might predict knowledge about treatment (Table 3). Information provision on the topics ‘Overdose’ and ‘How does medication work’ was associated with an increase in knowledge about these topics. Furthermore, those receiving information in one-to-one sessions with doctors were also more likely to score higher in the knowledge scale.
Table 3 Linear regression analysis of independent predictors associated with knowledge about treatment (R 2 = 0.46)
| B (95% CI) | s.e. | t-test | P | |
|---|---|---|---|---|
| Information provision modality | ||||
| 1:1 session with key worker | 1.27 (–0.78 to 3.33) | 1.03 | 1.23 | 0.22 |
| 1:1 session with doctorFootnote * | 2.44 (0.52 to 4.37) | 0.97 | 2.53 | 0.01 |
| Group sessions | 0.57 (–1.57 to 2.72) | 1.08 | 0.53 | 0.59 |
| Leaflets | –0.29 (–2.39 to 1.81) | 1.06 | –0.28 | 0.78 |
| Information provision topic | ||||
| Starting treatment | 1.55 (–0.84 to 3.93) | 1.20 | 1.29 | 0.20 |
| How medication worksFootnote * | –3.42 (–5.79 to –1.04) | 1.19 | –2.86 | 0.01 |
| Side-effects | –1.44 (–3.63 to 0.76) | 1.10 | –1.30 | 0.19 |
| Risks of treatment drop out | 2.14 (0.07 to 4.35) | 1.11 | 1.92 | 0.06 |
| Contraception | 0.44 (–1.48 to 2.36) | 0.96 | 0.46 | 0.65 |
| Hepatitis | –2.65 (–5.36 to 0.05) | 1.36 | –1.95 | 0.06 |
| OverdoseFootnote * | 3.42 (0.76 to 6.08) | 1.34 | 2.56 | 0.01Footnote * |
* Significant at α = 0.05 level.
Discussion
This study aimed to ascertain knowledge about opioid substitution treatment and relevant areas among patients receiving treatment with methadone and buprenorphine. Despite the majority of participants in this sample reporting receiving information about at least some aspects of treatment, the mean score achieved in this sample suggests significant gaps in knowledge across all core domains of treatment, as in line with previous studies. Reference Pinto, Rumball and Holland13,Reference Winstock, Lea, Maiden and Bath21
At the policy level, the importance of patient education in the addictions field has been acknowledged in the UK. Thus, according to the National Institute for Health and Clinical Excellence (NICE) guidelines, 22 the benefits and risks associated with methadone and buprenorphine should be discussed by the clinician and the patient, so that the latter may participate in the decision-making process. However, the findings from this small study suggest that in spite of current guidelines and the wide provision of information, the current approaches and resources being utilised to inform patients of their condition and its treatment in routine practice across the NHS may be limited in their effectiveness.
Increased knowledge about one's healthcare problems and available treatments can be associated with better clinical outcomes. Reference Cameron23–Reference Lutfey and Wishner26 Although some authors suggest that provision of adequate verbal and written information to motivated patients is advantageous and reinforces adherence, Reference Brown, Wright and Christensen15,Reference Cameron23 future studies should explore the mechanisms by which improved knowledge about treatment leads to better outcomes and adherence to interventions targeting opioid dependence.
Zaller and colleagues Reference Zaller, Bazazi, Velazquez and Rich27 stress that overlooking education about treatment may lead to undesirable results including increased drug-related morbidity and mortality. Reference Zanis and Woody28 Conversely, recent studies have shown that education can reduce potential harm. For example, research has demonstrated that overdose training increases knowledge about risk factors, preventive strategies and emergency techniques, which could be utilised to avoid preventable deaths. Reference Dietze, Jolley, Fry, Bammer and Moore29–Reference Strang, Manning, Mayet, Best, Titherington and Santana31 Moreover, as White et al Reference White, Schelten, Kelly and Casey32 recently pointed out, the provision of health-related information is likely to reinforce patient-clinician communication and empower individuals to take a more active role in their treatment decision–making, as advocated by the Royal College of Psychiatrists 33 and NICE. 22 Nevertheless, the fact that participants tended to significantly overestimate their knowledge suggests that many may not be in a position to make truly informed decisions regarding their treatment or be aware of the risky nature of their behaviours. If, in addition to poor knowledge, negative attitudes to treatment are also held, these may contribute to drop out.
This study indicates that the source and format of information provision can be important. Despite the existence of potential confounding factors (e.g. inaccurate recalling of events), this study suggests that psychoeducation provided by doctors may be more effective than that provided by other healthcare staff. The explanation for this finding is unclear, but it is noteworthy that previous studies have suggested that variables such as staff members’ knowledge, attitudes and personal experiences may influence the information they provide to others. Reference Andrews, Sorensen, Guydish, Delucchi and Greenberg34,Reference Knudsen, Ducharme and Roman35 Future studies should focus on evaluating the level of knowledge about treatment among staff members, as well as the efficacy of the psychoeducation strategies adopted by different professionals and their perceived credibility by those in treatment. Ultimately, this would inform treatment providers and aid the development of clinical guidelines about the most efficient language, modality and information sources when it comes to increasing individuals’ understanding about treatment.
Consideration should also be given as to when information can most effectively be provided. Clinical experience would indicate that written information is usually provided to patients at the start of treatment when patients may have other priorities or may be cycling in and out of withdrawal and intoxication while they stabilise on treatment. Reliance on written information may be misplaced given the low levels of literacy among this population. Reference Davis, Jackson, George, Long, Talley and Murphy36 That those in their first episode of treatment scored more poorly in this study suggested repeated exposure to information provision may also be important. We suggest that as part of care planning, key workers and prescribers should evaluate the level of knowledge that their patients have about their condition, its treatment and associated risks. This should not be seen as a single event but a continuous process of information exchange and evaluation. Such information exchange may reinforce gains made in treatment and dispel myths about prescribed medications and reduce treatment drop out. Reference Sokya, Zingg, Koller and Kuefner37
The study is not without limitations. First, the sample size was relatively small, with an uneven number of participants in the methadone and buprenorphine groups. Although this disproportion mirrors what occurs in real circumstances (approximate ratio of 3:1 individuals receiving methadone v. buprenorphine), there is the possibility of a type II error when comparing the level of knowledge between the two treatment groups. Similarly, the significant difference in knowledge between males and females might be related to the unevenness of male and female participants, possibly resulting in a type I error. Although the distribution of gender and the split between methadone and buprenorphine is broadly representative of the sample from which the participants were drawn from (E. Finch, personal communication, 2011), we believe strategies to increase sample size (e.g. use of vouchers, larger data collection period) should be considered in future studies to ensure higher statistical power. The questionnaire in this survey has been previously used in a similar study. Reference Winstock, Lea, Maden and Bath14 Nevertheless, the psychometric properties of this instrument, such as its test–retest reliability or concurrent validity with similar measures, need to be addressed in future validation studies. Another limitation also relates to the questionnaire properties, even though this measure was piloted for its overall understandability, the literacy level of participants in this sample was not determined, hence, it is possible that some participants did not fully comprehend all of the questions. Finally, future research should also focus on the resources (e.g. peers, health professionals, media) used by patients to gain knowledge about drug use, its risks and treatment-related areas. Without knowing how information was presented to patients, it is unclear how to develop strategies to improve patients’ knowledge and to avoid the perpetuation of myths regarding drug use and its treatment.
Implications
Opioid users prescribed with methadone and buprenorphine in the UK have low levels of knowledge regarding their condition, the medications used to treat it and related health problems. Increasing knowledge about treatment to improve its effectiveness is likely to be beneficial to both patients and treatment providers. Reference Stancliff, Myers, Steiner and Drucker38 The combination of overestimation and low levels of knowledge regarding treatment may have significant implications for health outcomes and participation in risky behaviours. Poorly informed patients make less well-informed choices regarding their treatment and this may contribute to a sense of disempowerment, common to many in drug treatment. Information provision should be the responsibility of all clinical staff. Future studies must focus on the development of strategies to improve patients’ knowledge, as well as to explore the impact that variables such as beliefs, perceptions and attitudes about treatment may have on the clinical outcomes.
Open access
eLetters
No eLetters have been published for this article.