G Model DAD-5176; No. of Pages 7
ARTICLE IN PRESS Drug and Alcohol Dependence xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Drug and Alcohol Dependence journal homepage: www.elsevier.com/locate/drugalcdep
Development and validation of the scale to assess satisfaction with medications for addiction treatment-methadone for heroin addiction (SASMAT-METHER)夽 ˜ a , Francesca Batlle a José Pérez de los Cobos a,∗ , Joan Trujols a,b , Núria Sinol a Addictive Behaviours Unit, Department of Psychiatry, Hospital de la Santa Creu i Sant Pau, Sant Pau Biomedical Research Institute (IIB Sant Pau), Autonomous University of Barcelona School of Medicine, Sant Antoni Maria Claret 167, 08025 Barcelona, Spain b Centro de Investigación Biomédica en Red de Salud Mental (CIBERSAM), Madrid, Spain
a r t i c l e
i n f o
Article history: Received 30 October 2013 Received in revised form 24 April 2014 Accepted 31 May 2014 Available online xxx Keywords: Patient satisfaction Methadone Heroin addiction Factor structure Validation study
a b s t r a c t Objective: To develop and examine the psychometric properties of a scale to specifically assess satisfaction with methadone in heroin-dependent patients. Methods: The 44-item preliminary version of the scale to assess satisfaction with medications for addiction treatment–methadone for heroin addiction (SASMAT–METHER) was obtained from a pool of items designed to assess satisfaction with any medication-addiction combination. Theoretical domains of the initial SASMAT–METHER were overall satisfaction, pharmacotherapy, initiation, anti-addictive effect on heroin, mental state, physical state, personal functioning, acceptability, and anti-addictive effect on secondary substances. The Treatment Satisfaction Questionnaire for Medication 1.4 version (TSQM 1.4) and the Verona Service Satisfaction Scale for Methadone Treatment (VSSS-MT) were used for concurrent validation. Participants included heroin-dependent patients receiving methadone treatment for at least the last 3 months. Results: The preliminary version of the SASMAT–METHER scale was completed by 241 patients, with 180 surveys considered suitable for factor analysis. Principal component analysis of these SASMAT–METHER surveys revealed a 3-factor structure that accounted for 40.4% of total variance. Based on similarities between empirically-obtained factors and theoretical domains, factors 1 through 3 were named ‘Personal Functioning and Well-Being’ (7 items), ‘Anti-Addictive Effect on Heroin’ (5 items), and ‘AntiAddictive Effect on Other Substances’ (5 items). All factors showed good to excellent internal consistency (Cronbach’s ˛: 0.83–0.92) and test–retest reliability (intraclass correlation coefficients: 0.66–0.89). Correlations between overall SASMAT–METHER and TSQM 1.4 scores were stronger (Pearson r = 0.69) than correlations between overall SASMAT–METHER and VSSS-MT scores (Pearson r = 0.26). Conclusion: These results present evidence for the validity and reliability of SASMAT–METHER. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction In accordance with the fundamental role of expectations in treatment satisfaction (Sitzia and Wood, 1997), patient satisfaction with a medication can be defined as a subjective condition that is determined by perceived differences between expectations and the
夽 Supplementary material can be found by accessing the online version of this paper at http://dx.doi.org and by entering doi:... ∗ Corresponding author. Tel.: +34 93 5537665; fax: +34 93 5537666. E-mail address: [email protected] (J.P.d.l. Cobos).
actual experience of taking the medication. Research in other areas of psychiatry has shown that patient satisfaction with medications is positively associated with adherence and/or clinical outcomes (Gasquet et al., 2006; Gharabawi et al., 2006; Sweileh et al., 2012). Given this positive association, it may be possible to increase the effectiveness of medications currently used to treat addictions by assessing patient satisfaction with those medications. It appears highly probable that patient satisfaction with methadone is a clinically relevant factor in the treatment of heroin addiction. One report found that nearly 30% of patients were either dissatisfied or had mixed feelings about methadone as a medication (Pérez de los Cobos et al., 2005). Consistent with this finding, some
http://dx.doi.org/10.1016/j.drugalcdep.2014.05.024 0376-8716/© 2014 Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: Cobos, J.P.d.l., et al., Development and validation of the scale to assess satisfaction with medications for addiction treatment-methadone for heroin addiction (SASMAT-METHER). Drug Alcohol Depend. (2014), http://dx.doi.org/10.1016/j.drugalcdep.2014.05.024
G Model DAD-5176; No. of Pages 7 2
ARTICLE IN PRESS J.P.d.l. Cobos et al. / Drug and Alcohol Dependence xxx (2014) xxx–xxx
heroin-dependent patients have negative views about the effects of methadone on their physical or psychological state, an attitude that has been associated with worse engagement with this treatment (Fischer et al., 2002; Kayman et al., 2006; Schwartz et al., 2011). Dissatisfied patients are highly unlikely to adhere to methadone treatment for the extended period of time needed to successfully treat heroin dependence, especially given the chronic nature and high relapse rate associated with this disorder. To our knowledge, no instruments have been developed to specifically evaluate satisfaction with medications with a proven efficacy in addiction treatment. Instead, researchers must rely on generic instruments designed to assess satisfaction with all types of medications, an example of which is the Treatment Satisfaction Questionnaire for Medication 1.4 version (TSQM 1.4; Atkinson et al., 2004). The TSQM 1.4 contains various domains and this feature is important given that a medication could produce desirable (e.g., symptom relief) or undesirable results (e.g., adverse effects). Although the TSQM 1.4 is a reliable and valid measure of satisfaction with methadone to treat heroin addiction (Trujols et al., 2012), it is a generic instrument that does not explicitly evaluate certain patient experiences that can strongly influence satisfaction with the medication. The TSQM 1.4 includes questions about side effects, but it does not mention interactions between the medication and substances of abuse. Importantly, some of these interactions are associated with very disturbing experiences for patients such as unintentional overdoses due to synergic effects of medications and substances of abuse (Neira-León et al., 2006), withdrawal from the main substance precipitated by medications (Johnson et al., 2003), and alcohol–disulfiram reaction (Christensen et al., 1991). Likewise, the TSQM 1.4 does not assess patient experiences in terms of the compatibility of medications with social, occupational, and other key activities involved in addiction rehabilitation. Moreover, it seems likely that the generic questions regarding symptom relief has important limitations in the case of substance-dependent patients given that such patients frequently present impaired insight into addiction symptoms (Goldstein et al., 2009). This deficit could account for the absence of spontaneous substance craving reported by more than 20% of patients with severe addictions during a 14-day detoxification treatment program (Pérez de los Cobos et al., 2011). We assume that the specific questioning strategy is more efficacious than the generic one to reduce impaired insight on relief of addiction symptoms and, in general, for eliciting patient opinions about certain experiences that may be associated with dissatisfaction with medications. The objective of the present study was to develop an instrument to specifically and multidimensionally assess satisfaction with methadone in heroin-dependent patients. This new instrument was developed from a pool of items applicable to any medication–addiction combination. We denominated this pool of items the ‘Scale to Assess Satisfaction with Medications for Addiction Treatment’ (SASMAT) (see the Supplementary material). In this study we examine the factor structure, reliability, and concurrent validity of the SASMAT for methadone treatment of heroin addiction (SASMAT–METHER) (see the Supplementary material). The TSQM 1.4 and the Verona Service Satisfaction Scale for Methadone Treatment (VSSS-MT; Pérez de los Cobos et al., 2002) are used for concurrent validation. We expect the SASMAT–METHER to have a stronger positive relationship to the TSQM 1.4 than to the VSSS-MT given that the TSQM 1.4 assesses satisfaction with methadone whereas the VSSS-MT is focused on satisfaction with services delivered by treatment centres. Some association between SASMAT–METHER and VSSS-MT is expected because we have found a correlation between the answer to a single question about satisfaction with methadone as a medicine and overall VSSS-MT scores (Pérez de los Cobos et al., 2005).
2. Methods 2.1. Development of SASMAT and the preliminary version of SASMAT–METHER The SASMAT was developed by a team consisting of a psychiatrist (J.P.) and a psychologist (J.T.). Development of the SASMAT was based on the researchers’ 20+ years of professional experience in addiction treatment combined with a review of the relevant literature. The authors agree with Shikiar and Rentz (2004), who concluded that the degree of patient satisfaction with a medication is a function of both the process of taking the medication and the results thereof. Moreover, we agree with the aforementioned authors that the main domains of patient satisfaction with a medication are effectiveness of symptom relief, the pharmacologicrelated features of the medication, and the impact of the medication on health-related quality of life. In addition, we believe that the acceptability of the medication is a key component in patient satisfaction with the pharmacological treatment of addictions. The SASMAT is an overinclusive pool of 44 items, all of which may be of value to assess patient satisfaction with any medication–addiction combination. The SASMAT was divided into nine theoretical domains, with each domain composed of a number of items directly proportional to their conceptual importance. The domains are as follows (the number of items in each domain is given in brackets): overall satisfaction (3), pharmacotherapy (8), initiation (3), anti-addictive effect on main substance (7), mental state (5), physical state (5), personal functioning (4), acceptability (4), and anti-addictive effect on secondary substances (5). The English version of SASMAT and the rationale for the contents of each domain are included in the Supplementary materials. The SASMAT instructions clearly indicate that the questions are specifically designed only to assess satisfaction with the medication assessed but not with psychosocial interventions or other factors (e.g., the attitude of physicians) that accompany (or may accompany) pharmacological treatment. The minimum assessment period for the SASMAT is 3 months, a period of time that should allow patients to have acquired sufficient experience with the medication. However, it should be noted that SASMAT has been designed to assess the overall experience (current or previous) with the medication. The response options on all the SASMAT items, except those included in the acceptability domain, use a five-point Likert scale (1 = terrible, 2 = generally unsatisfactory, 3 = mixed, 4 = generally satisfactory, 5 = excellent). Questions that evaluate satisfaction with the efficacy of the medication to interfere with the effects of the main addiction substance or which assess compatibility between the medication under evaluation with other medications or with work/study activities also include an additional response option: ‘not applicable’. This extra option was included in these cases to account for situations in which the patient did not take the main substance of abuse, took other medications, or performed work/study activities while receiving treatment with the medication. Likewise, the five items that assess the anti-addictive effect of methadone on secondary substances of abuse (e.g., cocaine, nicotine, etc.) may not be applicable if these secondary substances were not consumed at least twenty times during the treatment period. On the acceptability domain, two items ask patients if they would take the medication again or recommend it to a friend; the response options for these questions range from 1 (‘no, definitely not’) to 5 (‘yes, definitely yes’). Similarly, on the same acceptability domain, the two questions on thoughts or behaviours of abandoning the medication range from ‘often’ (1) to ‘never’ (5). All response options of the SASMAT are presented with alternate directionality. SASMAT scores are obtained by averaging the applicable items. Thus, the scores for each domain and overall for the SASMAT can range from 1 to 5.
Please cite this article in press as: Cobos, J.P.d.l., et al., Development and validation of the scale to assess satisfaction with medications for addiction treatment-methadone for heroin addiction (SASMAT-METHER). Drug Alcohol Depend. (2014), http://dx.doi.org/10.1016/j.drugalcdep.2014.05.024
G Model DAD-5176; No. of Pages 7
ARTICLE IN PRESS J.P.d.l. Cobos et al. / Drug and Alcohol Dependence xxx (2014) xxx–xxx
The “SASMAT–METHER” is the version of the SASMAT specifically developed to evaluate patient satisfaction with ‘methadone’ to treat ‘heroin’ addiction (METHER is an abbreviation of methadone [MET] plus heroin [HER]). The SASMAT–METHER was obtained by simply inserting the terms ‘methadone’ and ‘heroin’ in the appropriate places on the SASMAT (see the Supplementary materials). 2.2. Participants Participants included methadone-maintained, heroindependent patients (DSM-IV: 304.02; American Psychiatric Association, 1994) who had received methadone for at least the previous 3 months. Exclusion criteria included mental disorders that could hinder patient assessment (e.g., neurocognitive disorders or substance intoxication) and difficulty answering the survey due to limited literacy or poor Spanish language proficiency. All subjects signed an informed consent form and were recruited from the Addiction Unit of Sant Pau Hospital (Barcelona) using accidental sampling. Some participants were receiving indefinite follow-up care at Sant Pau methadone clinic while other participants were referred from other methadone clinics in Barcelona and were at the Sant Pau Hospital Addiction Unit for clinical assessment and/or short-term treatments at the time of the study. No compensation was provided to subjects for participation in this survey. The study protocol was approved by the Institutional Review Board of Sant Pau Hospital. The survey began in February, 2007 and ended in March, 2012. 2.3. Assessments to examine concurrent validity of SASMAT–METHER
3
good to excellent internal consistency reliabilities (Cronbach’s ˛: 0.96, 0.84, 0.95, and 0.98, respectively). 2.4. Statistical analyses Exploratory factor analysis of SASMAT–METHER was conducted using principal component analysis. The Scree test (Cattell, 1961), Horn’s parallel analysis (Horn, 1965), and clinical interpretability of factor structure were used to determine the number of components to retain. A varimax rotation was performed on the resulting component pattern. SASMAT–METHER items with a factor loading 0.32 (i.e., accounting for more than 10% of common variance of two or more factors) were removed. The internal consistency of the components resulting from the previous analysis was estimated using Cronbach’s alpha coefficient (Cronbach, 1951). Test–retest reliability of the SASMAT–METHER was estimated by calculating the intraclass correlation coefficient (ICC). Cronbach’s alpha and ICC results were interpreted according to the descriptive adjectives for ranges of reliability proposed by George and Mallery (2003) and Ciccheti (1994), respectively. Categorical variables were compared using Chi-square tests and continuous variables were compared with independent samples t-test. Pearson correlations were used to express the relationships between SAMAT–METHER scores and other quantitative variables. Tests of significance were two-tailed and considered significant if P < 0.05. All analyses were performed using SPSS version 19.0 (SPSS Inc., Chicago, IL, USA). 3. Results 3.1. Characteristics of participants
The TSQM 1.4 and the Verona Service Satisfaction Scale for Methadone Treatment (VSSS-MT) were used for concurrent validation. The TSQM 1.4 is a 14-item instrument designed to generically assess patient satisfaction with any medication (Atkinson et al., 2004). The four subscales of the TSQM 1.4 are effectiveness, side effects, convenience, and global satisfaction. Responses are obtained on a 5- or 7-point Likert scale (ranging from 1 ‘extremely dissatisfied’ to 7 ‘extremely satisfied’) for all but one item (yes–no response option). The subscale scores are transformed into scores ranging from 0 to 100. The side effects subscale score equals 100 when there is a self-reported absence of side effects. In the present sample, the internal consistency of the TSQM 1.4 demonstrated satisfactory reliability. All but one of the subscales showed at least an acceptable level of internal consistency. Effectiveness, side effects, and global satisfaction subscales showed acceptable to good internal consistency reliabilities (Cronbach’s ˛: 0.76, 0.82, and 0.88, respectively). The only exception was the convenience subscale, whose ˛ coefficient was 0.54. The VSSS-MT assesses satisfaction with services delivered by methadone treatment centres (Pérez de los Cobos et al., 2002). The subscales of VSSS-MT are basic interventions, specific interventions, social worker skills, and psychologist skills. The basic interventions subscale primarily assesses doctors’ and nurses’ skills, and the help received in improving social relationships and self-care. Specific interventions assess only psychosocial interventions. Each VSSS-MT item has a five-point Likert scale response option (1 = terrible to 5 = excellent). VSSS-MT scores are obtained by averaging applicable items and the overall and subscale scores of the VSSS-MT range from 1 to 5. Only the VSSS-MT completed by patients treated at methadone clinics other than the Sant Pau methadone clinic were used in the present study due to potential response bias given that the psychologist who supervised all clinical assessments (N.S.) is a staff member of Addiction Unit of Sant Pau Hospital. In the present study, the four VSSS-MT factors showed
The characteristics of the 241 participants are shown in Table 1. One hundred and ten patients were recruited from the Sant Pau methadone clinic and the rest from other methadone clinics. The SASMAT–METHER surveys provided by 61 participants were excluded from the factor analysis because the respondents either answered “not applicable” on at least one item (n = 55) or left one (n = 5) or two (n = 1) items missing. Table 1 shows the characteristics of all subjects, including the 180 patients whose SASMAT–METHER survey responses were considered suitable for factor analysis as well as the aforementioned 61 patients whose responses were not suitable. No significant differences between the two groups were observed in age, gender, substance use, or pharmacological treatment. The 61 SASMAT–METHER surveys excluded from factor analysis were included in the concurrent validity analyses. In these cases, we averaged the scores of all applicable responses to obtain mean scores from those 61 surveys. The six subjects who had only one or two missing items were not excluded from concurrent validity analyses because their scales were considered valid given that they responded to nearly all of the items (95% or more). 3.2. Psychometric properties of SASMAT–METHER Factor structure. The correlation matrix was considered suitable for factor analytical modelling because the Meyer–Olkin measure yielded a value of 0.862 and Bartlett’s test of sphericity was statistically significant (P < 0.001). According to the Scree test, Horn’s parallel analysis, and clinical interpretability, three components were extracted (Fig. 1). These 3 components consisted of a total of 17 items (7, 5, and 5, respectively). The remaining items (i.e., 24) were removed because their factor loadings in each of the three components were ≤0.55. The 3-factor solution of SASMAT–METHER accounted for 40.4% of the total variance (Table 2). The variance accounted for by factors 1, 2 and 3 were 15.2%, 14.1%, and 11.0%, respectively.
Please cite this article in press as: Cobos, J.P.d.l., et al., Development and validation of the scale to assess satisfaction with medications for addiction treatment-methadone for heroin addiction (SASMAT-METHER). Drug Alcohol Depend. (2014), http://dx.doi.org/10.1016/j.drugalcdep.2014.05.024
G Model
ARTICLE IN PRESS
DAD-5176; No. of Pages 7
J.P.d.l. Cobos et al. / Drug and Alcohol Dependence xxx (2014) xxx–xxx
4
Table 1 Features and pharmacological treatment of participants. Full sample (n = 241) Age (years) Male gender (%) Heroin use Age of onset (years) Time of use (months) Main route of administration (%) Intravenous Intrapulmonary Intranasal History of non-opioid substance use (%) Alcohol Benzodiazepines Cannabis Cocaine Nicotine Methadone treatment Dose (mg/d) Split dosagea (%) Patients with take-home doses (%) Methadone form (%) Tablets Liquid Absence of side effectsb (%) Lifetime duration (months) Patients taking other medications (%)
Patients with suitable SASMAT–METHERs for factor analysis (n = 180)
Patients without suitable SASMAT–METHERs for factor analysis (n = 61)
2 (1) or t-test; p-value
40.2 ± 6.8 75.5
40.0 ± 6.5 76.1
41.3 ± 7.5 73.8
−1.47; 0.144 0.14; 0.713
21.0 ± 6.3 123.0 ± 75.9
21.2 ± 6.6 122.9 ± 77.4
20.4 ± 5.4 123.2 ± 71.9
0.72; 0.470 −0.02; 0.982
66.4 13.7 19.9
67.2 12.8 20.0
63.9 16.4 19.7
0.51; 0.774
66.4 60.6 88.8 92.9 98.3
66.7 58.9 88.3 92.8 98.3
65.6 65.6 90.2 93.4 98.4
0.02; 0.876 0.85; 0.356 0.70; 0.695 0.03; 0.861 0.00; 0.988
65.1 ± 73.0 17.6 90.5
61.5 ± 70.9 16.6 89.4
75.5 ± 78.5 21.2 93.4
−1.29; 0.197 0.58; 0.448 0.84; 0.358
38.2 61.8 35.7 93.6 ± 56.6 70.1
37.2 62.8 34.4 91.4 ± 53.6 72.2
41.0 59.0 39.3 100.3 ± 64.6 63.9
0.27; 0.601 0.48; 0.490 −1.07; 0.288 1.49; 0.222
Values are mean ± S.D. unless otherwise indicated. a Missing information for this variable in 20 participants. b According to item 4 of the TSQM 1.4.
Factor 1 included all four items from the personal functioning theoretical factor, which evaluates satisfaction with the compatibility of the effects of methadone on various activities. These four items had the highest component loadings in factor 1 (Table 2). Moreover, factor 1 included one item from each of the following theoretical factors: mental state (i.e., impact of taking methadone on ability to enjoy the pleasant things of life), physical state (i.e.,
impact on overall physical health), and pharmacotherapy (i.e., tolerability of methadone, considering the side effects experienced because of this medicine). Based on the contents of the seven items in factor 1, this factor was denominated ‘Personal Functioning and Well-Being’. Factor 2 was named ‘Anti-Addictive Effect on Heroin’ because this factor is comprised of five of the seven items of the theoretical
14
Real Dataset Eigenvalue
12
Random Data Eigenvalue
Eigenvalue
10
8
6
4
2
0 1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44
Principal component Fig. 1. SASMAT–METHER parallel analysis results showing eigenvalues for the actual data and for the 95th percentile random data.
Please cite this article in press as: Cobos, J.P.d.l., et al., Development and validation of the scale to assess satisfaction with medications for addiction treatment-methadone for heroin addiction (SASMAT-METHER). Drug Alcohol Depend. (2014), http://dx.doi.org/10.1016/j.drugalcdep.2014.05.024
G Model DAD-5176; No. of Pages 7
ARTICLE IN PRESS J.P.d.l. Cobos et al. / Drug and Alcohol Dependence xxx (2014) xxx–xxx
5
Table 2 Factor analysis of the SASMAT–METHER (n = 180). Factorsb
SASMAT–METHER itemsa
Compatibility with work/student activity Compatibility with leisure activities Compatibility with everyday activities Compatibility with social activities Impact on overall physical health Impact on enjoying pleasant things of life Tolerability regarding side effects Efficacy. . . . . .in reducing heroin use . . .in reducing heroin craving . . .in preventing heroin withdrawal . . .to stop thinking about heroin . . .to avoid paying attention to everything around me related to heroin . . .in reducing other substance use . . .in reducing other substance craving . . .to avoid paying attention to everything around me related to other substances . . .in preventing other substance withdrawal . . .to stop thinking about other substances a b
1
2
3
0.761 0.751 0.742 0.741 0.663 0.631 0.601
0.195 0.196 0.303 0.149 0.243 0.309 0.275
0.092 0.041 0.038 0.151 0.169 0.165 0.119
0.047 −0.039 0.222 0.031 0.099 0.106 0.126 0.125 0.029 0.213
0.761 0.753 0.670 0.650 0.578 0.089 0.217 0.231 0.059 0.175
0.178 0.240 −0.039 0.310 0.199 0.861 0.853 0.831 0.820 0.797
Component loadings >0.55 are highlighted in boldface in order to facilitate the interpretation of results. The names of factors are ‘Personal Functioning and Well-Being’ (1), ‘Anti-Addictive Effect on Heroin’ (2), ‘Anti-Addictive Effect on Other Substances’ (3).
factor with the same name. Similarly, factor 3 was called ‘AntiAddictive Effect on Other Substances’ because it consists of all five of the items from the theoretical factor of the same name. The only two items from the theoretical domain ‘Anti-Addictive Effect on Heroin’ not included in factor 2 (i.e., the efficacy of methadone to normalize the physical or psychological changes produced by heroin) were those that had no counterparts in factor 3 (anti-addictive effect on other substances). As a consequence, factor 2 and factor 3 wordings only differ in that factor 2 refers to heroin and factor 3 to other substances of abuse (Table 2). Internal consistency. Cronbach’s alpha coefficient values for the three factors were good to excellent: personal functioning and wellbeing: 0.89 (good); anti-addictive effect on heroin: 0.83 (good); anti-addictive effect on other substances: 0.92 (excellent). Deleting an item did not improve the internal consistency of the subscale. Thus, the 3 subscales were maintained in their current composition in the ensuing statistical analyses. The 17 items of SASMAT–METHER taken as a single composite scale yielded a Cronbach’s alpha of 0.90 (good). Test–retest reliability. A retest was performed in a subgroup of 23 patients. The average interval between the test and the retest was 3.8 (S.D. = 1.5) days. ICC point estimates for the three factors were good to excellent (95% confidence intervals given in brackets): personal functioning and well-being: 0.84 (0.63–0.93); anti-addictive effect on heroin: 0.89 (0.73–0.95); anti-addictive effect on other substances: 0.66 (0.20–0.86). The 17 items on the SASMAT–METHER taken as a single composite scale yielded an excellent ICC value of 0.83 (0.61–0.93). Concurrent validity. All correlations between SASMAT–METHER and overall and subscale TSQM 1.4 scores provided by the 241 participants were statistically significant (Table 3). SASMAT–METHER overall scores accounted for 48% of the total variance in TSQM 1.4 overall scores (Pearson r = 0.69). Values of r < 0.3 were obtained only in the four correlations between anti-addictive effect on heroin or anti-addictive effect on other substances factors, and side effects or convenience TSQM factors. Consistent with these findings, these TSQM 1.4 factors correlated more strongly with personal functioning and well-being, which assesses compatibility between the medication and personal activities and could be negatively affected by a perceived lack of convenience in taking the medication and by the side-effects of methadone. Nonetheless, personal functioning and well-being also had a stronger correlation with the TSQM effectiveness factor versus the other two SASMAT–METHER factors
that refer to the effectiveness of methadone to treat addiction to heroin or other substances. Pearson correlations between SASMAT–METHER and VSSS-MT overall and subscale scores were performed only for the 131 participants untreated at the Sant Pau methadone clinic (Table 4). Overall SASMAT–METHER scores accounted for 7% of total variance in VSSSMT overall scores (Pearson r = 0.26). Moreover, values of r > 0.3 were not obtained in any correlation between SASMAT–METHER and VSSS-MT factors. Therefore, SASMAT–METHER was more strongly correlated with TSQM than with VSSS-MT (Tables 3 and 4). SASMAT–METHER factors correlated more strongly with the basic interventions factor than with the other VSSS-MT factors. This is an expected finding given that basic interventions is the only VSSS-MT factor that assesses the skills of professionals directly involved in methadone prescription and dispensation: doctors and nurses. 4. Discussion As the results of the present study show, SASMAT–METHER is able to measure satisfaction with methadone in heroin-dependent patients quickly, specifically, and multidimensionally. It is notable that all three of the factors that make up the SASMAT–METHER were easy to interpret clinically. This interpretability is due to the fact that the contents of these factors had a marked similarity to three of the factors constructed from theoretical assumptions. Another finding that supports the validity of the factors obtained in this analysis is that there is a coherent association between these factors and the TSQM 1.4 and VSSS-MT. In addition, the SASMAT–METHER factors had a stronger correlation – as expected – with the TSQM 1.4 factors than with those of the VSSS-MT. In terms of reliability, all SAMAT–METHER factors showed good-toexcellent internal consistency and test–retest reliability. The three factors of the SASMAT–METHER (personal functioning and well-being, anti-addictive effect on heroin, and anti-addictive effect on other substances) presented a similar degree of relevance: each factor explained, respectively, 15%, 14% and 11% of the total variance. Thus, the perceived disability and discomfort associated with methadone use accounts for a proportion of variance in patient satisfaction that is similar to, or even greater than the perceived effect of methadone on heroin addiction. Moreover, the effectiveness of methadone in treating addictions to substances other than heroin is also a very important component of patient satisfaction with this medication.
Please cite this article in press as: Cobos, J.P.d.l., et al., Development and validation of the scale to assess satisfaction with medications for addiction treatment-methadone for heroin addiction (SASMAT-METHER). Drug Alcohol Depend. (2014), http://dx.doi.org/10.1016/j.drugalcdep.2014.05.024
G Model DAD-5176; No. of Pages 7
ARTICLE IN PRESS J.P.d.l. Cobos et al. / Drug and Alcohol Dependence xxx (2014) xxx–xxx
6
Table 3 Pearson correlations between SASMAT–METHER and TSQM 1.4 scores (n = 241). TSQM 1.4
SASMAT–METHER
Personal functioning and well-being Anti-addictive effect on heroin Anti-addictive effect on other substances Overall * **
Effectiveness
Side effects
Convenience
Global satisfaction
Overall
0.61** 0.51** 0.39** 0.62**
0.50** 0.17* 0.24** 0.38**
0.41** 0.26** 0.21* 0.36**
0.67** 0.63** 0.48** 0.74**
0.73** 0.50** 0.44** 0.69**
P < 0.01, two-tailed. P < 0.001, two-tailed.
Table 4 Pearson correlations between SASMAT–METHER and VSSS-MT scores (n = 131)a . VSSS-MTb
SASMAT–METHER
Personal Functioning and Well-Being Anti-Addictive Effect on Heroin Anti-Addictive Effect on Other Substances Overall a b † *
Basic interventions (n = 131)
Specific interventions (n = 121)
Social worker skills (n = 92)
Psychologist skills (n = 87)
Overall (n = 131)
0.22† 0.17 0.23* 0.27*
0.20† 0.03 0.08 0.13
0.14 −0.004 0.06 0.08
0.17 0.18 0.11 0.20
0.23* 0.16 0.21† 0.26*
Patients treated at Sant Pau methadone clinic centre were not included in this analysis in order to avoid potential response bias on VSSS-MT questions. The number of patients included in each VSSS-MT subscale was different due to that the answer ‘not applicable’ was used with different frequency on the subscales. P < 0.05, two-tailed. P < 0.01, two-tailed.
Although personal functioning and well-being was the only empirical factor that included items from more than one theoretical factor, it is relatively easy to interpret this factor clinically. Several reasons explain this relatively easy interpretability of this factor. First, this factor included the entirety (four items) of the theoretical factor “Personal Functioning”, which was designed to evaluate satisfaction with the compatibility of methadone with work/study, everyday, leisure, and social activities. Second, the other three items from factor 1 assess aspects related to patient well-being, which is assumed to affect personal functioning. Thus, one of these items evaluates the repercussion of methadone on hedonic response, which surely influences the subject’s ability to perform activities, especially leisure ones. The other two items from this factor measure satisfaction with methadone in terms of tolerability of its side effects and the impact of methadone on overall physical health. The items included in anti-addictive effect on heroin (factor 2), and anti-addictive effect on other substances (factor 3) show that patient satisfaction depends, in large part, on the effectiveness of methadone to decrease the use of heroin or other substances, or to reduce thoughts about, cravings for, or attention paid to those substances, or to reduce withdrawal symptoms. Although methadone has not been proven efficacious in the treatment of addictions to non-opioid substances, our results indicate that patients nevertheless expect methadone to be efficacious in treating the abuse of such substances. Staff members could induce these expectations by regulating methadone administration in a contingent manner, not only to heroin use, but also to the use of other substances such as cocaine. In addition, patients whose previous experience with methadone has shown that methadone reduces symptoms (e.g., craving) of addiction to cocaine (Foltin and Fischman, 1998) might have expectations in this regard. In contrast, patients who report an increase in the addictive effects of cocaine in the context of methadone maintenance treatment (Preston et al., 1996) would be unlikely to harbour any expectations with regard to the anti-addictive effects of methadone on secondary substances and this lack of such expectations could promote dissatisfaction with methadone. Apart from global satisfaction, the only theoretical factors that did not contribute any items to the 17-item definitive version of the SASMAT–METHER were initiation and acceptability. Initiation may
have little relevance because methadone treatment had already been successfully implemented in all participants as evidenced by the fact that the study participants had been taking it for at least the last 3 months. Similarly, it is possible that our hypothesis regarding the association between satisfaction and acceptability is not consistent in patients who currently accept methadone as a medication, as was the case with all participants in the present study. Nevertheless, initiation or acceptability may be closely related to satisfaction in patients who have experience with methadone treatment but who refuse to continue – or to reinitiate – this treatment (Reisinger et al., 2009; Peterson et al., 2010). The present study has several limitations. First, the SASMAT–METHER might not be the most patient-centred tool to measure satisfaction with methadone given that the patients did not participate in developing this scale, nor did they evaluate its pertinence a posteriori (Trujols et al., 2013, 2014). Second, the representativeness of the sample is limited because the participants were recruited from a convenience sample. Moreover, the sample did not include patients who had taken methadone in the past but were not taking methadone at the time of the study. Although assessment of these patients could help to better understand the relationship between satisfaction with methadone and acceptability of this medication, SASMAT–METHER was originally designed to assess and improve the treatment of patients who are currently on methadone treatment. Third, the size (n = 180) of the subsample used to analyze the main components of the initial version of the 44-item SASMAT–METHER was modest according to the rule-of-thumb that recommends a ratio of at least 5 participants per item evaluated (Gorsuch, 1983). However, the minimum sample size is not invariant across studies because the stability of the factor structure is critically related to the size of factor loadings and the number of variables per factor (MacCallum et al., 1999). According to a Monte Carlo study performed by Guadagnoli and Velicer (1988), stable factor structures are obtained when, as in the present study, each factor has a minimum of 4 variables with factor loadings >0.60 and the sample includes more than 150 participants. Later simulation studies (e.g., de Winter et al., 2009) have shown that even smaller sample sizes are sufficient for factor structures that, like ours, have few factors (2 or 3) and are
Please cite this article in press as: Cobos, J.P.d.l., et al., Development and validation of the scale to assess satisfaction with medications for addiction treatment-methadone for heroin addiction (SASMAT-METHER). Drug Alcohol Depend. (2014), http://dx.doi.org/10.1016/j.drugalcdep.2014.05.024
G Model DAD-5176; No. of Pages 7
ARTICLE IN PRESS J.P.d.l. Cobos et al. / Drug and Alcohol Dependence xxx (2014) xxx–xxx
clearly represented (e.g., mid-range factor loadings: 0.50–0.60) by a sufficient number of variables (e.g., 5). The good psychometric properties of the SASMAT–METHER encourage its use as a research tool and in clinical practice. This new tool could be useful to clinicians who wish to incorporate the patient perspective into methadone treatment management. Moreover, SASMAT–METHER could be of value to verify if patient satisfaction with methadone is associated with adherence and/or clinical outcomes in heroin addiction treatment. In addition to carrying out further studies to establish the predictive validity of SASMAT–METHER, other studies will be required to determine the discriminant validity of the tool, and to replicate its reliability and factor structure in other samples of heroin-dependent patients. Role of funding source Funding for this study was provided by grant PI06/0531 from Fondo de Investigación Sanitaria, Instituto de Salud Carlos III (Spanish Ministries of Economy and Competitiveness, and Health, Social Services and Equality). The Instituto de Salud Carlos III had no further role in study design; in the collection, analysis and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication. Contributors José Pérez de los Cobos and Joan Trujols developed the SASMAT, wrote the protocol, and managed the literature searches. ˜ performed data collection. Joan Trujols analyzed the Núria Sinol ˜ psychometric properties of SASMAT–METHER and Nuria Sinol undertook the rest of the statistical analysis. José Pérez de los Cobos and Francesca Batlle wrote the first draft of the manuscript. All authors substantially contributed to and have approved the final manuscript. Conflict of interest José Pérez de los Cobos declares having received grant support for research and educational activities from Reckitt-Benckiser. All other authors report no financial or other relationship relevant to the subject of this study. Acknowledgements We are grateful to the patients who kindly participated in the study, and to Vanessa García, Saiko Allende and Isabel Blásquiz for their secretarial support. We would also like to thank Quintiles, Inc. (Durham, NC) for giving us permission to use the Spanish version of the TSQM 1.4 in this study. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.drugalcdep. 2014.05.024. References American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Mental Disorders: DSM-IV, fourth ed. American Psychiatric Association, Washington, DC. Atkinson, M.J., Sinha, A., Hass, S.L., Colman, S.S., Kumar, R.N., Brod, M., Rowland, C.R., 2004. Validation of a general measure of treatment satisfaction, the Treatment Satisfaction Questionnaire for Medication (TSQM), using a national panel study of chronic disease. Health Qual. Life Outcomes 2, 12. Cattell, R.B., 1961. The Scree test for the number of factors. Multivar. Behav. Res. 1, 245–276.
7
Christensen, J.K., Møller, I.W., Rønsted, P., Angelo, H.R., Johansson, B., 1991. Dose–effect relationship of disulfiram in human volunteers I: clinical studies. Pharmacol. Toxicol. 68, 163–165. Ciccheti, D.V., 1994. Guidelines, criteria, and rules of thumb for evaluating normed and standardized assessment instruments in psychology. Psychol. Assess. 6, 284–290. Cronbach, L.J., 1951. Coefficient alpha and the internal structure of tests. Psychometrika 16, 297–334. de Winter, J.C.F., Dodou, D., Wieringa, P.A., 2009. Exploratory factor analysis with small sample sizes. Multivar. Behav. Res. 44, 147–181. Fischer, B., Chin, A.T., Kuo, I., Kirst, M., Vlahov, D., 2002. Canadian illicit opiate users’ views on methadone and other opiate prescription treatment: an exploratory qualitative study. Subst. Use Misuse 37, 495–522. Foltin, R.W., Fischman, M.W., 1998. Effects of “binge” use of intravenous cocaine in methadone-maintained individuals. Addiction 93, 825–836. Gasquet, I., Tcherny-Lessenot, S., Lepine, J.P., Falissard, B., 2006. Patient satisfaction with psychotropic drugs: sensitivity to change and relationship to clinical status, quality-of-life, compliance and effectiveness of treatment. Results from a nationwide 6-month prospective study. Eur. Psychiatry 21, 531–538. George, D., Mallery, P., 2003. SPSS for Windows Step by Step: A Simple Guide and Reference 11. 0 Update, fourth ed. Allyn and Bacon, Boston, MA. Gharabawi, G.M., Greenspan, A., Rupnow, M.F., Kosik-Gonzalez, C., Bossie, C.A., Zhu, Y., Kalali, A.H., Awad, A.G., 2006. Reduction in psychotic symptoms as a predictor of patient satisfaction with antipsychotic medication in schizophrenia: data from a randomized double-blind trial. BMC Psychiatry 6, 45. Goldstein, R.Z., Craig, A.D., Bechara, A., Garavan, H., Childress, A.R., Paulus, M.P., Volkow, N.D., 2009. The neurocircuitry of impaired insight in drug addiction. Trends Cogn. Sci. 13, 372–380. Gorsuch, R.L., 1983. Factor Analysis, second ed. Lawrence Erlbaum, Hillsdale, NJ. Guadagnoli, E., Velicer, W.F., 1988. Relation of sample size to the stability of component patterns. Psychol. Bull. 103, 265–275. Horn, J.L., 1965. A rationale and test for the number of factors in factor analysis. Psychometrika 30, 179–185. Johnson, R.E., Strain, E.C., Amass, L., 2003. Buprenorphine: how to use it right. Drug Alcohol Depend. 70 (2 Suppl.), S59–S77. Kayman, D.J., Goldstein, M.F., Deren, S., Rosenblum, A., 2006. Predicting treatment retention with a brief “Opinions About Methadone” scale. J. Psychoactive Drugs 38, 93–100. MacCallum, R.C., Widaman, K.F., Zhang, S., Hong, S., 1999. Sample size in factor analysis. Psychol. Methods 4, 84–99. Neira-León, M., Barrio, G., Brugal, M.T., de la Fuente, L., Ballesta, R., Bravo, M.J., Silva, T.C., Rodríguez-Martos, A., Project Itinere Group, 2006. Do young heroin users in Madrid, Barcelona and Seville have sufficient knowledge of the risk factors for unintentional opioid overdose? J. Urban Health 83, 477–496. ˜ ˜ Pérez de los Cobos, J., Sinol, N., Trujols, J., Banuls, E., Batlle, F., Tejero, A., 2011. Drug-dependent inpatients reporting continuous absence of spontaneous drug craving for the main substance throughout detoxification treatment. Drug Alcohol Rev. 30, 403–410. Pérez de los Cobos, J., Trujols, J., Valderrama, J.C., Valero, S., Puig, T., 2005. Patient perspectives on methadone maintenance treatment in the Valencia region: dose adjustment, participation in dosage regulation, and satisfaction with treatment. Drug Alcohol Depend. 79, 405–412. Pérez de los Cobos, J., Valero, S., Haro, G., Fidel, G., Escuder, G., Trujols, J., Valderrama, J.C., 2002. Development and psychometric properties of the Verona Service Satisfaction Scale for methadone-treated opioid-dependent patients (VSSS-MT). Drug Alcohol Depend. 68, 209–214. Peterson, J.A., Schwartz, R.P., Mitchell, S.G., Reisinger, H.S., Kelly, S.M., O’Grady, K.E., Brown, B.S., Agar, M.H., 2010. Why don’t out-of-treatment individuals enter methadone treatment programmes? Int. J. Drug Policy 21, 36–42. Preston, K.L., Sullivan, J.T., Strain, E.C., Bigelow, G.E., 1996. Enhancement of cocaine’s abuse liability in methadone maintenance patients. Psychopharmacology (Berl.) 123, 15–25. Reisinger, H.S., Schwartz, R.P., Mitchell, S.G., Peterson, J.A., Kelly, S.M., O’Grady, K.E., Marrari, E.A., Brown, B.S., Agar, M.H., 2009. Premature discharge from methadone treatment: patient perspectives. J. Psychoactive Drugs 41, 285–296. Schwartz, R.P., Kelly, S.M., O’Grady, K.E., Mitchell, S.G., Brown, B.S., 2011. Antecedents and correlates of methadone treatment entry: a comparison of out-of-treatment and in-treatment cohorts. Drug Alcohol Depend. 115, 23–29. Shikiar, R., Rentz, A.M., 2004. Satisfaction with medication: an overview of conceptual, methodologic, and regulatory issues. Value Health 7, 204–215. Sitzia, J., Wood, N., 1997. Patient satisfaction: a review of issues and concepts. Soc. Sci. Med. 45, 1829–1843. Sweileh, W.M., Ihbesheh, M.S., Jarar, I.S., Sawalha, A.F., Abu Taha, A.S., Zyoud, S.H., Morisky, D.E., 2012. Antipsychotic medication adherence and satisfaction among Palestinian people with schizophrenia. Curr. Clin. Pharmacol. 7, 49–55. ˜ Trujols, J., Iraurgi, I., Sinol, N., Portella, M.J., Pérez, V., Pérez de los Cobos, J., 2012. Satisfaction with methadone as a medication: psychometric properties of the Spanish version of the treatment satisfaction questionnaire for medication. J. Clin. Psychopharmacol. 32, 69–74. Trujols, J., Iraurgi, I., Oviedo-Joekes, E., Guàrdia-Olmos, J.A., 2014. A critical analysis of user satisfaction surveys in addiction services: opioid maintenance treatment as a representative case study. Patient Prefer. Adherence 8, 107–117. ˜ Trujols, J., Portella, M.J., Iraurgi, I., Campins, M.J., Sinol, N., Pérez de los Cobos, J., 2013. Patient-reported outcome measures: are they patient-generated, patientcentred or patient-valued? J. Ment. Health 22, 555–562.
Please cite this article in press as: Cobos, J.P.d.l., et al., Development and validation of the scale to assess satisfaction with medications for addiction treatment-methadone for heroin addiction (SASMAT-METHER). Drug Alcohol Depend. (2014), http://dx.doi.org/10.1016/j.drugalcdep.2014.05.024
ARTICLE IN PRESS Drug and Alcohol Dependence xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Drug and Alcohol Dependence journal homepage: www.elsevier.com/locate/drugalcdep
Development and validation of the scale to assess satisfaction with medications for addiction treatment-methadone for heroin addiction (SASMAT-METHER)夽 ˜ a , Francesca Batlle a José Pérez de los Cobos a,∗ , Joan Trujols a,b , Núria Sinol a Addictive Behaviours Unit, Department of Psychiatry, Hospital de la Santa Creu i Sant Pau, Sant Pau Biomedical Research Institute (IIB Sant Pau), Autonomous University of Barcelona School of Medicine, Sant Antoni Maria Claret 167, 08025 Barcelona, Spain b Centro de Investigación Biomédica en Red de Salud Mental (CIBERSAM), Madrid, Spain
a r t i c l e
i n f o
Article history: Received 30 October 2013 Received in revised form 24 April 2014 Accepted 31 May 2014 Available online xxx Keywords: Patient satisfaction Methadone Heroin addiction Factor structure Validation study
a b s t r a c t Objective: To develop and examine the psychometric properties of a scale to specifically assess satisfaction with methadone in heroin-dependent patients. Methods: The 44-item preliminary version of the scale to assess satisfaction with medications for addiction treatment–methadone for heroin addiction (SASMAT–METHER) was obtained from a pool of items designed to assess satisfaction with any medication-addiction combination. Theoretical domains of the initial SASMAT–METHER were overall satisfaction, pharmacotherapy, initiation, anti-addictive effect on heroin, mental state, physical state, personal functioning, acceptability, and anti-addictive effect on secondary substances. The Treatment Satisfaction Questionnaire for Medication 1.4 version (TSQM 1.4) and the Verona Service Satisfaction Scale for Methadone Treatment (VSSS-MT) were used for concurrent validation. Participants included heroin-dependent patients receiving methadone treatment for at least the last 3 months. Results: The preliminary version of the SASMAT–METHER scale was completed by 241 patients, with 180 surveys considered suitable for factor analysis. Principal component analysis of these SASMAT–METHER surveys revealed a 3-factor structure that accounted for 40.4% of total variance. Based on similarities between empirically-obtained factors and theoretical domains, factors 1 through 3 were named ‘Personal Functioning and Well-Being’ (7 items), ‘Anti-Addictive Effect on Heroin’ (5 items), and ‘AntiAddictive Effect on Other Substances’ (5 items). All factors showed good to excellent internal consistency (Cronbach’s ˛: 0.83–0.92) and test–retest reliability (intraclass correlation coefficients: 0.66–0.89). Correlations between overall SASMAT–METHER and TSQM 1.4 scores were stronger (Pearson r = 0.69) than correlations between overall SASMAT–METHER and VSSS-MT scores (Pearson r = 0.26). Conclusion: These results present evidence for the validity and reliability of SASMAT–METHER. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction In accordance with the fundamental role of expectations in treatment satisfaction (Sitzia and Wood, 1997), patient satisfaction with a medication can be defined as a subjective condition that is determined by perceived differences between expectations and the
夽 Supplementary material can be found by accessing the online version of this paper at http://dx.doi.org and by entering doi:... ∗ Corresponding author. Tel.: +34 93 5537665; fax: +34 93 5537666. E-mail address: [email protected] (J.P.d.l. Cobos).
actual experience of taking the medication. Research in other areas of psychiatry has shown that patient satisfaction with medications is positively associated with adherence and/or clinical outcomes (Gasquet et al., 2006; Gharabawi et al., 2006; Sweileh et al., 2012). Given this positive association, it may be possible to increase the effectiveness of medications currently used to treat addictions by assessing patient satisfaction with those medications. It appears highly probable that patient satisfaction with methadone is a clinically relevant factor in the treatment of heroin addiction. One report found that nearly 30% of patients were either dissatisfied or had mixed feelings about methadone as a medication (Pérez de los Cobos et al., 2005). Consistent with this finding, some
http://dx.doi.org/10.1016/j.drugalcdep.2014.05.024 0376-8716/© 2014 Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: Cobos, J.P.d.l., et al., Development and validation of the scale to assess satisfaction with medications for addiction treatment-methadone for heroin addiction (SASMAT-METHER). Drug Alcohol Depend. (2014), http://dx.doi.org/10.1016/j.drugalcdep.2014.05.024
G Model DAD-5176; No. of Pages 7 2
ARTICLE IN PRESS J.P.d.l. Cobos et al. / Drug and Alcohol Dependence xxx (2014) xxx–xxx
heroin-dependent patients have negative views about the effects of methadone on their physical or psychological state, an attitude that has been associated with worse engagement with this treatment (Fischer et al., 2002; Kayman et al., 2006; Schwartz et al., 2011). Dissatisfied patients are highly unlikely to adhere to methadone treatment for the extended period of time needed to successfully treat heroin dependence, especially given the chronic nature and high relapse rate associated with this disorder. To our knowledge, no instruments have been developed to specifically evaluate satisfaction with medications with a proven efficacy in addiction treatment. Instead, researchers must rely on generic instruments designed to assess satisfaction with all types of medications, an example of which is the Treatment Satisfaction Questionnaire for Medication 1.4 version (TSQM 1.4; Atkinson et al., 2004). The TSQM 1.4 contains various domains and this feature is important given that a medication could produce desirable (e.g., symptom relief) or undesirable results (e.g., adverse effects). Although the TSQM 1.4 is a reliable and valid measure of satisfaction with methadone to treat heroin addiction (Trujols et al., 2012), it is a generic instrument that does not explicitly evaluate certain patient experiences that can strongly influence satisfaction with the medication. The TSQM 1.4 includes questions about side effects, but it does not mention interactions between the medication and substances of abuse. Importantly, some of these interactions are associated with very disturbing experiences for patients such as unintentional overdoses due to synergic effects of medications and substances of abuse (Neira-León et al., 2006), withdrawal from the main substance precipitated by medications (Johnson et al., 2003), and alcohol–disulfiram reaction (Christensen et al., 1991). Likewise, the TSQM 1.4 does not assess patient experiences in terms of the compatibility of medications with social, occupational, and other key activities involved in addiction rehabilitation. Moreover, it seems likely that the generic questions regarding symptom relief has important limitations in the case of substance-dependent patients given that such patients frequently present impaired insight into addiction symptoms (Goldstein et al., 2009). This deficit could account for the absence of spontaneous substance craving reported by more than 20% of patients with severe addictions during a 14-day detoxification treatment program (Pérez de los Cobos et al., 2011). We assume that the specific questioning strategy is more efficacious than the generic one to reduce impaired insight on relief of addiction symptoms and, in general, for eliciting patient opinions about certain experiences that may be associated with dissatisfaction with medications. The objective of the present study was to develop an instrument to specifically and multidimensionally assess satisfaction with methadone in heroin-dependent patients. This new instrument was developed from a pool of items applicable to any medication–addiction combination. We denominated this pool of items the ‘Scale to Assess Satisfaction with Medications for Addiction Treatment’ (SASMAT) (see the Supplementary material). In this study we examine the factor structure, reliability, and concurrent validity of the SASMAT for methadone treatment of heroin addiction (SASMAT–METHER) (see the Supplementary material). The TSQM 1.4 and the Verona Service Satisfaction Scale for Methadone Treatment (VSSS-MT; Pérez de los Cobos et al., 2002) are used for concurrent validation. We expect the SASMAT–METHER to have a stronger positive relationship to the TSQM 1.4 than to the VSSS-MT given that the TSQM 1.4 assesses satisfaction with methadone whereas the VSSS-MT is focused on satisfaction with services delivered by treatment centres. Some association between SASMAT–METHER and VSSS-MT is expected because we have found a correlation between the answer to a single question about satisfaction with methadone as a medicine and overall VSSS-MT scores (Pérez de los Cobos et al., 2005).
2. Methods 2.1. Development of SASMAT and the preliminary version of SASMAT–METHER The SASMAT was developed by a team consisting of a psychiatrist (J.P.) and a psychologist (J.T.). Development of the SASMAT was based on the researchers’ 20+ years of professional experience in addiction treatment combined with a review of the relevant literature. The authors agree with Shikiar and Rentz (2004), who concluded that the degree of patient satisfaction with a medication is a function of both the process of taking the medication and the results thereof. Moreover, we agree with the aforementioned authors that the main domains of patient satisfaction with a medication are effectiveness of symptom relief, the pharmacologicrelated features of the medication, and the impact of the medication on health-related quality of life. In addition, we believe that the acceptability of the medication is a key component in patient satisfaction with the pharmacological treatment of addictions. The SASMAT is an overinclusive pool of 44 items, all of which may be of value to assess patient satisfaction with any medication–addiction combination. The SASMAT was divided into nine theoretical domains, with each domain composed of a number of items directly proportional to their conceptual importance. The domains are as follows (the number of items in each domain is given in brackets): overall satisfaction (3), pharmacotherapy (8), initiation (3), anti-addictive effect on main substance (7), mental state (5), physical state (5), personal functioning (4), acceptability (4), and anti-addictive effect on secondary substances (5). The English version of SASMAT and the rationale for the contents of each domain are included in the Supplementary materials. The SASMAT instructions clearly indicate that the questions are specifically designed only to assess satisfaction with the medication assessed but not with psychosocial interventions or other factors (e.g., the attitude of physicians) that accompany (or may accompany) pharmacological treatment. The minimum assessment period for the SASMAT is 3 months, a period of time that should allow patients to have acquired sufficient experience with the medication. However, it should be noted that SASMAT has been designed to assess the overall experience (current or previous) with the medication. The response options on all the SASMAT items, except those included in the acceptability domain, use a five-point Likert scale (1 = terrible, 2 = generally unsatisfactory, 3 = mixed, 4 = generally satisfactory, 5 = excellent). Questions that evaluate satisfaction with the efficacy of the medication to interfere with the effects of the main addiction substance or which assess compatibility between the medication under evaluation with other medications or with work/study activities also include an additional response option: ‘not applicable’. This extra option was included in these cases to account for situations in which the patient did not take the main substance of abuse, took other medications, or performed work/study activities while receiving treatment with the medication. Likewise, the five items that assess the anti-addictive effect of methadone on secondary substances of abuse (e.g., cocaine, nicotine, etc.) may not be applicable if these secondary substances were not consumed at least twenty times during the treatment period. On the acceptability domain, two items ask patients if they would take the medication again or recommend it to a friend; the response options for these questions range from 1 (‘no, definitely not’) to 5 (‘yes, definitely yes’). Similarly, on the same acceptability domain, the two questions on thoughts or behaviours of abandoning the medication range from ‘often’ (1) to ‘never’ (5). All response options of the SASMAT are presented with alternate directionality. SASMAT scores are obtained by averaging the applicable items. Thus, the scores for each domain and overall for the SASMAT can range from 1 to 5.
Please cite this article in press as: Cobos, J.P.d.l., et al., Development and validation of the scale to assess satisfaction with medications for addiction treatment-methadone for heroin addiction (SASMAT-METHER). Drug Alcohol Depend. (2014), http://dx.doi.org/10.1016/j.drugalcdep.2014.05.024
G Model DAD-5176; No. of Pages 7
ARTICLE IN PRESS J.P.d.l. Cobos et al. / Drug and Alcohol Dependence xxx (2014) xxx–xxx
The “SASMAT–METHER” is the version of the SASMAT specifically developed to evaluate patient satisfaction with ‘methadone’ to treat ‘heroin’ addiction (METHER is an abbreviation of methadone [MET] plus heroin [HER]). The SASMAT–METHER was obtained by simply inserting the terms ‘methadone’ and ‘heroin’ in the appropriate places on the SASMAT (see the Supplementary materials). 2.2. Participants Participants included methadone-maintained, heroindependent patients (DSM-IV: 304.02; American Psychiatric Association, 1994) who had received methadone for at least the previous 3 months. Exclusion criteria included mental disorders that could hinder patient assessment (e.g., neurocognitive disorders or substance intoxication) and difficulty answering the survey due to limited literacy or poor Spanish language proficiency. All subjects signed an informed consent form and were recruited from the Addiction Unit of Sant Pau Hospital (Barcelona) using accidental sampling. Some participants were receiving indefinite follow-up care at Sant Pau methadone clinic while other participants were referred from other methadone clinics in Barcelona and were at the Sant Pau Hospital Addiction Unit for clinical assessment and/or short-term treatments at the time of the study. No compensation was provided to subjects for participation in this survey. The study protocol was approved by the Institutional Review Board of Sant Pau Hospital. The survey began in February, 2007 and ended in March, 2012. 2.3. Assessments to examine concurrent validity of SASMAT–METHER
3
good to excellent internal consistency reliabilities (Cronbach’s ˛: 0.96, 0.84, 0.95, and 0.98, respectively). 2.4. Statistical analyses Exploratory factor analysis of SASMAT–METHER was conducted using principal component analysis. The Scree test (Cattell, 1961), Horn’s parallel analysis (Horn, 1965), and clinical interpretability of factor structure were used to determine the number of components to retain. A varimax rotation was performed on the resulting component pattern. SASMAT–METHER items with a factor loading 0.32 (i.e., accounting for more than 10% of common variance of two or more factors) were removed. The internal consistency of the components resulting from the previous analysis was estimated using Cronbach’s alpha coefficient (Cronbach, 1951). Test–retest reliability of the SASMAT–METHER was estimated by calculating the intraclass correlation coefficient (ICC). Cronbach’s alpha and ICC results were interpreted according to the descriptive adjectives for ranges of reliability proposed by George and Mallery (2003) and Ciccheti (1994), respectively. Categorical variables were compared using Chi-square tests and continuous variables were compared with independent samples t-test. Pearson correlations were used to express the relationships between SAMAT–METHER scores and other quantitative variables. Tests of significance were two-tailed and considered significant if P < 0.05. All analyses were performed using SPSS version 19.0 (SPSS Inc., Chicago, IL, USA). 3. Results 3.1. Characteristics of participants
The TSQM 1.4 and the Verona Service Satisfaction Scale for Methadone Treatment (VSSS-MT) were used for concurrent validation. The TSQM 1.4 is a 14-item instrument designed to generically assess patient satisfaction with any medication (Atkinson et al., 2004). The four subscales of the TSQM 1.4 are effectiveness, side effects, convenience, and global satisfaction. Responses are obtained on a 5- or 7-point Likert scale (ranging from 1 ‘extremely dissatisfied’ to 7 ‘extremely satisfied’) for all but one item (yes–no response option). The subscale scores are transformed into scores ranging from 0 to 100. The side effects subscale score equals 100 when there is a self-reported absence of side effects. In the present sample, the internal consistency of the TSQM 1.4 demonstrated satisfactory reliability. All but one of the subscales showed at least an acceptable level of internal consistency. Effectiveness, side effects, and global satisfaction subscales showed acceptable to good internal consistency reliabilities (Cronbach’s ˛: 0.76, 0.82, and 0.88, respectively). The only exception was the convenience subscale, whose ˛ coefficient was 0.54. The VSSS-MT assesses satisfaction with services delivered by methadone treatment centres (Pérez de los Cobos et al., 2002). The subscales of VSSS-MT are basic interventions, specific interventions, social worker skills, and psychologist skills. The basic interventions subscale primarily assesses doctors’ and nurses’ skills, and the help received in improving social relationships and self-care. Specific interventions assess only psychosocial interventions. Each VSSS-MT item has a five-point Likert scale response option (1 = terrible to 5 = excellent). VSSS-MT scores are obtained by averaging applicable items and the overall and subscale scores of the VSSS-MT range from 1 to 5. Only the VSSS-MT completed by patients treated at methadone clinics other than the Sant Pau methadone clinic were used in the present study due to potential response bias given that the psychologist who supervised all clinical assessments (N.S.) is a staff member of Addiction Unit of Sant Pau Hospital. In the present study, the four VSSS-MT factors showed
The characteristics of the 241 participants are shown in Table 1. One hundred and ten patients were recruited from the Sant Pau methadone clinic and the rest from other methadone clinics. The SASMAT–METHER surveys provided by 61 participants were excluded from the factor analysis because the respondents either answered “not applicable” on at least one item (n = 55) or left one (n = 5) or two (n = 1) items missing. Table 1 shows the characteristics of all subjects, including the 180 patients whose SASMAT–METHER survey responses were considered suitable for factor analysis as well as the aforementioned 61 patients whose responses were not suitable. No significant differences between the two groups were observed in age, gender, substance use, or pharmacological treatment. The 61 SASMAT–METHER surveys excluded from factor analysis were included in the concurrent validity analyses. In these cases, we averaged the scores of all applicable responses to obtain mean scores from those 61 surveys. The six subjects who had only one or two missing items were not excluded from concurrent validity analyses because their scales were considered valid given that they responded to nearly all of the items (95% or more). 3.2. Psychometric properties of SASMAT–METHER Factor structure. The correlation matrix was considered suitable for factor analytical modelling because the Meyer–Olkin measure yielded a value of 0.862 and Bartlett’s test of sphericity was statistically significant (P < 0.001). According to the Scree test, Horn’s parallel analysis, and clinical interpretability, three components were extracted (Fig. 1). These 3 components consisted of a total of 17 items (7, 5, and 5, respectively). The remaining items (i.e., 24) were removed because their factor loadings in each of the three components were ≤0.55. The 3-factor solution of SASMAT–METHER accounted for 40.4% of the total variance (Table 2). The variance accounted for by factors 1, 2 and 3 were 15.2%, 14.1%, and 11.0%, respectively.
Please cite this article in press as: Cobos, J.P.d.l., et al., Development and validation of the scale to assess satisfaction with medications for addiction treatment-methadone for heroin addiction (SASMAT-METHER). Drug Alcohol Depend. (2014), http://dx.doi.org/10.1016/j.drugalcdep.2014.05.024
G Model
ARTICLE IN PRESS
DAD-5176; No. of Pages 7
J.P.d.l. Cobos et al. / Drug and Alcohol Dependence xxx (2014) xxx–xxx
4
Table 1 Features and pharmacological treatment of participants. Full sample (n = 241) Age (years) Male gender (%) Heroin use Age of onset (years) Time of use (months) Main route of administration (%) Intravenous Intrapulmonary Intranasal History of non-opioid substance use (%) Alcohol Benzodiazepines Cannabis Cocaine Nicotine Methadone treatment Dose (mg/d) Split dosagea (%) Patients with take-home doses (%) Methadone form (%) Tablets Liquid Absence of side effectsb (%) Lifetime duration (months) Patients taking other medications (%)
Patients with suitable SASMAT–METHERs for factor analysis (n = 180)
Patients without suitable SASMAT–METHERs for factor analysis (n = 61)
2 (1) or t-test; p-value
40.2 ± 6.8 75.5
40.0 ± 6.5 76.1
41.3 ± 7.5 73.8
−1.47; 0.144 0.14; 0.713
21.0 ± 6.3 123.0 ± 75.9
21.2 ± 6.6 122.9 ± 77.4
20.4 ± 5.4 123.2 ± 71.9
0.72; 0.470 −0.02; 0.982
66.4 13.7 19.9
67.2 12.8 20.0
63.9 16.4 19.7
0.51; 0.774
66.4 60.6 88.8 92.9 98.3
66.7 58.9 88.3 92.8 98.3
65.6 65.6 90.2 93.4 98.4
0.02; 0.876 0.85; 0.356 0.70; 0.695 0.03; 0.861 0.00; 0.988
65.1 ± 73.0 17.6 90.5
61.5 ± 70.9 16.6 89.4
75.5 ± 78.5 21.2 93.4
−1.29; 0.197 0.58; 0.448 0.84; 0.358
38.2 61.8 35.7 93.6 ± 56.6 70.1
37.2 62.8 34.4 91.4 ± 53.6 72.2
41.0 59.0 39.3 100.3 ± 64.6 63.9
0.27; 0.601 0.48; 0.490 −1.07; 0.288 1.49; 0.222
Values are mean ± S.D. unless otherwise indicated. a Missing information for this variable in 20 participants. b According to item 4 of the TSQM 1.4.
Factor 1 included all four items from the personal functioning theoretical factor, which evaluates satisfaction with the compatibility of the effects of methadone on various activities. These four items had the highest component loadings in factor 1 (Table 2). Moreover, factor 1 included one item from each of the following theoretical factors: mental state (i.e., impact of taking methadone on ability to enjoy the pleasant things of life), physical state (i.e.,
impact on overall physical health), and pharmacotherapy (i.e., tolerability of methadone, considering the side effects experienced because of this medicine). Based on the contents of the seven items in factor 1, this factor was denominated ‘Personal Functioning and Well-Being’. Factor 2 was named ‘Anti-Addictive Effect on Heroin’ because this factor is comprised of five of the seven items of the theoretical
14
Real Dataset Eigenvalue
12
Random Data Eigenvalue
Eigenvalue
10
8
6
4
2
0 1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44
Principal component Fig. 1. SASMAT–METHER parallel analysis results showing eigenvalues for the actual data and for the 95th percentile random data.
Please cite this article in press as: Cobos, J.P.d.l., et al., Development and validation of the scale to assess satisfaction with medications for addiction treatment-methadone for heroin addiction (SASMAT-METHER). Drug Alcohol Depend. (2014), http://dx.doi.org/10.1016/j.drugalcdep.2014.05.024
G Model DAD-5176; No. of Pages 7
ARTICLE IN PRESS J.P.d.l. Cobos et al. / Drug and Alcohol Dependence xxx (2014) xxx–xxx
5
Table 2 Factor analysis of the SASMAT–METHER (n = 180). Factorsb
SASMAT–METHER itemsa
Compatibility with work/student activity Compatibility with leisure activities Compatibility with everyday activities Compatibility with social activities Impact on overall physical health Impact on enjoying pleasant things of life Tolerability regarding side effects Efficacy. . . . . .in reducing heroin use . . .in reducing heroin craving . . .in preventing heroin withdrawal . . .to stop thinking about heroin . . .to avoid paying attention to everything around me related to heroin . . .in reducing other substance use . . .in reducing other substance craving . . .to avoid paying attention to everything around me related to other substances . . .in preventing other substance withdrawal . . .to stop thinking about other substances a b
1
2
3
0.761 0.751 0.742 0.741 0.663 0.631 0.601
0.195 0.196 0.303 0.149 0.243 0.309 0.275
0.092 0.041 0.038 0.151 0.169 0.165 0.119
0.047 −0.039 0.222 0.031 0.099 0.106 0.126 0.125 0.029 0.213
0.761 0.753 0.670 0.650 0.578 0.089 0.217 0.231 0.059 0.175
0.178 0.240 −0.039 0.310 0.199 0.861 0.853 0.831 0.820 0.797
Component loadings >0.55 are highlighted in boldface in order to facilitate the interpretation of results. The names of factors are ‘Personal Functioning and Well-Being’ (1), ‘Anti-Addictive Effect on Heroin’ (2), ‘Anti-Addictive Effect on Other Substances’ (3).
factor with the same name. Similarly, factor 3 was called ‘AntiAddictive Effect on Other Substances’ because it consists of all five of the items from the theoretical factor of the same name. The only two items from the theoretical domain ‘Anti-Addictive Effect on Heroin’ not included in factor 2 (i.e., the efficacy of methadone to normalize the physical or psychological changes produced by heroin) were those that had no counterparts in factor 3 (anti-addictive effect on other substances). As a consequence, factor 2 and factor 3 wordings only differ in that factor 2 refers to heroin and factor 3 to other substances of abuse (Table 2). Internal consistency. Cronbach’s alpha coefficient values for the three factors were good to excellent: personal functioning and wellbeing: 0.89 (good); anti-addictive effect on heroin: 0.83 (good); anti-addictive effect on other substances: 0.92 (excellent). Deleting an item did not improve the internal consistency of the subscale. Thus, the 3 subscales were maintained in their current composition in the ensuing statistical analyses. The 17 items of SASMAT–METHER taken as a single composite scale yielded a Cronbach’s alpha of 0.90 (good). Test–retest reliability. A retest was performed in a subgroup of 23 patients. The average interval between the test and the retest was 3.8 (S.D. = 1.5) days. ICC point estimates for the three factors were good to excellent (95% confidence intervals given in brackets): personal functioning and well-being: 0.84 (0.63–0.93); anti-addictive effect on heroin: 0.89 (0.73–0.95); anti-addictive effect on other substances: 0.66 (0.20–0.86). The 17 items on the SASMAT–METHER taken as a single composite scale yielded an excellent ICC value of 0.83 (0.61–0.93). Concurrent validity. All correlations between SASMAT–METHER and overall and subscale TSQM 1.4 scores provided by the 241 participants were statistically significant (Table 3). SASMAT–METHER overall scores accounted for 48% of the total variance in TSQM 1.4 overall scores (Pearson r = 0.69). Values of r < 0.3 were obtained only in the four correlations between anti-addictive effect on heroin or anti-addictive effect on other substances factors, and side effects or convenience TSQM factors. Consistent with these findings, these TSQM 1.4 factors correlated more strongly with personal functioning and well-being, which assesses compatibility between the medication and personal activities and could be negatively affected by a perceived lack of convenience in taking the medication and by the side-effects of methadone. Nonetheless, personal functioning and well-being also had a stronger correlation with the TSQM effectiveness factor versus the other two SASMAT–METHER factors
that refer to the effectiveness of methadone to treat addiction to heroin or other substances. Pearson correlations between SASMAT–METHER and VSSS-MT overall and subscale scores were performed only for the 131 participants untreated at the Sant Pau methadone clinic (Table 4). Overall SASMAT–METHER scores accounted for 7% of total variance in VSSSMT overall scores (Pearson r = 0.26). Moreover, values of r > 0.3 were not obtained in any correlation between SASMAT–METHER and VSSS-MT factors. Therefore, SASMAT–METHER was more strongly correlated with TSQM than with VSSS-MT (Tables 3 and 4). SASMAT–METHER factors correlated more strongly with the basic interventions factor than with the other VSSS-MT factors. This is an expected finding given that basic interventions is the only VSSS-MT factor that assesses the skills of professionals directly involved in methadone prescription and dispensation: doctors and nurses. 4. Discussion As the results of the present study show, SASMAT–METHER is able to measure satisfaction with methadone in heroin-dependent patients quickly, specifically, and multidimensionally. It is notable that all three of the factors that make up the SASMAT–METHER were easy to interpret clinically. This interpretability is due to the fact that the contents of these factors had a marked similarity to three of the factors constructed from theoretical assumptions. Another finding that supports the validity of the factors obtained in this analysis is that there is a coherent association between these factors and the TSQM 1.4 and VSSS-MT. In addition, the SASMAT–METHER factors had a stronger correlation – as expected – with the TSQM 1.4 factors than with those of the VSSS-MT. In terms of reliability, all SAMAT–METHER factors showed good-toexcellent internal consistency and test–retest reliability. The three factors of the SASMAT–METHER (personal functioning and well-being, anti-addictive effect on heroin, and anti-addictive effect on other substances) presented a similar degree of relevance: each factor explained, respectively, 15%, 14% and 11% of the total variance. Thus, the perceived disability and discomfort associated with methadone use accounts for a proportion of variance in patient satisfaction that is similar to, or even greater than the perceived effect of methadone on heroin addiction. Moreover, the effectiveness of methadone in treating addictions to substances other than heroin is also a very important component of patient satisfaction with this medication.
Please cite this article in press as: Cobos, J.P.d.l., et al., Development and validation of the scale to assess satisfaction with medications for addiction treatment-methadone for heroin addiction (SASMAT-METHER). Drug Alcohol Depend. (2014), http://dx.doi.org/10.1016/j.drugalcdep.2014.05.024
G Model DAD-5176; No. of Pages 7
ARTICLE IN PRESS J.P.d.l. Cobos et al. / Drug and Alcohol Dependence xxx (2014) xxx–xxx
6
Table 3 Pearson correlations between SASMAT–METHER and TSQM 1.4 scores (n = 241). TSQM 1.4
SASMAT–METHER
Personal functioning and well-being Anti-addictive effect on heroin Anti-addictive effect on other substances Overall * **
Effectiveness
Side effects
Convenience
Global satisfaction
Overall
0.61** 0.51** 0.39** 0.62**
0.50** 0.17* 0.24** 0.38**
0.41** 0.26** 0.21* 0.36**
0.67** 0.63** 0.48** 0.74**
0.73** 0.50** 0.44** 0.69**
P < 0.01, two-tailed. P < 0.001, two-tailed.
Table 4 Pearson correlations between SASMAT–METHER and VSSS-MT scores (n = 131)a . VSSS-MTb
SASMAT–METHER
Personal Functioning and Well-Being Anti-Addictive Effect on Heroin Anti-Addictive Effect on Other Substances Overall a b † *
Basic interventions (n = 131)
Specific interventions (n = 121)
Social worker skills (n = 92)
Psychologist skills (n = 87)
Overall (n = 131)
0.22† 0.17 0.23* 0.27*
0.20† 0.03 0.08 0.13
0.14 −0.004 0.06 0.08
0.17 0.18 0.11 0.20
0.23* 0.16 0.21† 0.26*
Patients treated at Sant Pau methadone clinic centre were not included in this analysis in order to avoid potential response bias on VSSS-MT questions. The number of patients included in each VSSS-MT subscale was different due to that the answer ‘not applicable’ was used with different frequency on the subscales. P < 0.05, two-tailed. P < 0.01, two-tailed.
Although personal functioning and well-being was the only empirical factor that included items from more than one theoretical factor, it is relatively easy to interpret this factor clinically. Several reasons explain this relatively easy interpretability of this factor. First, this factor included the entirety (four items) of the theoretical factor “Personal Functioning”, which was designed to evaluate satisfaction with the compatibility of methadone with work/study, everyday, leisure, and social activities. Second, the other three items from factor 1 assess aspects related to patient well-being, which is assumed to affect personal functioning. Thus, one of these items evaluates the repercussion of methadone on hedonic response, which surely influences the subject’s ability to perform activities, especially leisure ones. The other two items from this factor measure satisfaction with methadone in terms of tolerability of its side effects and the impact of methadone on overall physical health. The items included in anti-addictive effect on heroin (factor 2), and anti-addictive effect on other substances (factor 3) show that patient satisfaction depends, in large part, on the effectiveness of methadone to decrease the use of heroin or other substances, or to reduce thoughts about, cravings for, or attention paid to those substances, or to reduce withdrawal symptoms. Although methadone has not been proven efficacious in the treatment of addictions to non-opioid substances, our results indicate that patients nevertheless expect methadone to be efficacious in treating the abuse of such substances. Staff members could induce these expectations by regulating methadone administration in a contingent manner, not only to heroin use, but also to the use of other substances such as cocaine. In addition, patients whose previous experience with methadone has shown that methadone reduces symptoms (e.g., craving) of addiction to cocaine (Foltin and Fischman, 1998) might have expectations in this regard. In contrast, patients who report an increase in the addictive effects of cocaine in the context of methadone maintenance treatment (Preston et al., 1996) would be unlikely to harbour any expectations with regard to the anti-addictive effects of methadone on secondary substances and this lack of such expectations could promote dissatisfaction with methadone. Apart from global satisfaction, the only theoretical factors that did not contribute any items to the 17-item definitive version of the SASMAT–METHER were initiation and acceptability. Initiation may
have little relevance because methadone treatment had already been successfully implemented in all participants as evidenced by the fact that the study participants had been taking it for at least the last 3 months. Similarly, it is possible that our hypothesis regarding the association between satisfaction and acceptability is not consistent in patients who currently accept methadone as a medication, as was the case with all participants in the present study. Nevertheless, initiation or acceptability may be closely related to satisfaction in patients who have experience with methadone treatment but who refuse to continue – or to reinitiate – this treatment (Reisinger et al., 2009; Peterson et al., 2010). The present study has several limitations. First, the SASMAT–METHER might not be the most patient-centred tool to measure satisfaction with methadone given that the patients did not participate in developing this scale, nor did they evaluate its pertinence a posteriori (Trujols et al., 2013, 2014). Second, the representativeness of the sample is limited because the participants were recruited from a convenience sample. Moreover, the sample did not include patients who had taken methadone in the past but were not taking methadone at the time of the study. Although assessment of these patients could help to better understand the relationship between satisfaction with methadone and acceptability of this medication, SASMAT–METHER was originally designed to assess and improve the treatment of patients who are currently on methadone treatment. Third, the size (n = 180) of the subsample used to analyze the main components of the initial version of the 44-item SASMAT–METHER was modest according to the rule-of-thumb that recommends a ratio of at least 5 participants per item evaluated (Gorsuch, 1983). However, the minimum sample size is not invariant across studies because the stability of the factor structure is critically related to the size of factor loadings and the number of variables per factor (MacCallum et al., 1999). According to a Monte Carlo study performed by Guadagnoli and Velicer (1988), stable factor structures are obtained when, as in the present study, each factor has a minimum of 4 variables with factor loadings >0.60 and the sample includes more than 150 participants. Later simulation studies (e.g., de Winter et al., 2009) have shown that even smaller sample sizes are sufficient for factor structures that, like ours, have few factors (2 or 3) and are
Please cite this article in press as: Cobos, J.P.d.l., et al., Development and validation of the scale to assess satisfaction with medications for addiction treatment-methadone for heroin addiction (SASMAT-METHER). Drug Alcohol Depend. (2014), http://dx.doi.org/10.1016/j.drugalcdep.2014.05.024
G Model DAD-5176; No. of Pages 7
ARTICLE IN PRESS J.P.d.l. Cobos et al. / Drug and Alcohol Dependence xxx (2014) xxx–xxx
clearly represented (e.g., mid-range factor loadings: 0.50–0.60) by a sufficient number of variables (e.g., 5). The good psychometric properties of the SASMAT–METHER encourage its use as a research tool and in clinical practice. This new tool could be useful to clinicians who wish to incorporate the patient perspective into methadone treatment management. Moreover, SASMAT–METHER could be of value to verify if patient satisfaction with methadone is associated with adherence and/or clinical outcomes in heroin addiction treatment. In addition to carrying out further studies to establish the predictive validity of SASMAT–METHER, other studies will be required to determine the discriminant validity of the tool, and to replicate its reliability and factor structure in other samples of heroin-dependent patients. Role of funding source Funding for this study was provided by grant PI06/0531 from Fondo de Investigación Sanitaria, Instituto de Salud Carlos III (Spanish Ministries of Economy and Competitiveness, and Health, Social Services and Equality). The Instituto de Salud Carlos III had no further role in study design; in the collection, analysis and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication. Contributors José Pérez de los Cobos and Joan Trujols developed the SASMAT, wrote the protocol, and managed the literature searches. ˜ performed data collection. Joan Trujols analyzed the Núria Sinol ˜ psychometric properties of SASMAT–METHER and Nuria Sinol undertook the rest of the statistical analysis. José Pérez de los Cobos and Francesca Batlle wrote the first draft of the manuscript. All authors substantially contributed to and have approved the final manuscript. Conflict of interest José Pérez de los Cobos declares having received grant support for research and educational activities from Reckitt-Benckiser. All other authors report no financial or other relationship relevant to the subject of this study. Acknowledgements We are grateful to the patients who kindly participated in the study, and to Vanessa García, Saiko Allende and Isabel Blásquiz for their secretarial support. We would also like to thank Quintiles, Inc. (Durham, NC) for giving us permission to use the Spanish version of the TSQM 1.4 in this study. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.drugalcdep. 2014.05.024. References American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Mental Disorders: DSM-IV, fourth ed. American Psychiatric Association, Washington, DC. Atkinson, M.J., Sinha, A., Hass, S.L., Colman, S.S., Kumar, R.N., Brod, M., Rowland, C.R., 2004. Validation of a general measure of treatment satisfaction, the Treatment Satisfaction Questionnaire for Medication (TSQM), using a national panel study of chronic disease. Health Qual. Life Outcomes 2, 12. Cattell, R.B., 1961. The Scree test for the number of factors. Multivar. Behav. Res. 1, 245–276.
7
Christensen, J.K., Møller, I.W., Rønsted, P., Angelo, H.R., Johansson, B., 1991. Dose–effect relationship of disulfiram in human volunteers I: clinical studies. Pharmacol. Toxicol. 68, 163–165. Ciccheti, D.V., 1994. Guidelines, criteria, and rules of thumb for evaluating normed and standardized assessment instruments in psychology. Psychol. Assess. 6, 284–290. Cronbach, L.J., 1951. Coefficient alpha and the internal structure of tests. Psychometrika 16, 297–334. de Winter, J.C.F., Dodou, D., Wieringa, P.A., 2009. Exploratory factor analysis with small sample sizes. Multivar. Behav. Res. 44, 147–181. Fischer, B., Chin, A.T., Kuo, I., Kirst, M., Vlahov, D., 2002. Canadian illicit opiate users’ views on methadone and other opiate prescription treatment: an exploratory qualitative study. Subst. Use Misuse 37, 495–522. Foltin, R.W., Fischman, M.W., 1998. Effects of “binge” use of intravenous cocaine in methadone-maintained individuals. Addiction 93, 825–836. Gasquet, I., Tcherny-Lessenot, S., Lepine, J.P., Falissard, B., 2006. Patient satisfaction with psychotropic drugs: sensitivity to change and relationship to clinical status, quality-of-life, compliance and effectiveness of treatment. Results from a nationwide 6-month prospective study. Eur. Psychiatry 21, 531–538. George, D., Mallery, P., 2003. SPSS for Windows Step by Step: A Simple Guide and Reference 11. 0 Update, fourth ed. Allyn and Bacon, Boston, MA. Gharabawi, G.M., Greenspan, A., Rupnow, M.F., Kosik-Gonzalez, C., Bossie, C.A., Zhu, Y., Kalali, A.H., Awad, A.G., 2006. Reduction in psychotic symptoms as a predictor of patient satisfaction with antipsychotic medication in schizophrenia: data from a randomized double-blind trial. BMC Psychiatry 6, 45. Goldstein, R.Z., Craig, A.D., Bechara, A., Garavan, H., Childress, A.R., Paulus, M.P., Volkow, N.D., 2009. The neurocircuitry of impaired insight in drug addiction. Trends Cogn. Sci. 13, 372–380. Gorsuch, R.L., 1983. Factor Analysis, second ed. Lawrence Erlbaum, Hillsdale, NJ. Guadagnoli, E., Velicer, W.F., 1988. Relation of sample size to the stability of component patterns. Psychol. Bull. 103, 265–275. Horn, J.L., 1965. A rationale and test for the number of factors in factor analysis. Psychometrika 30, 179–185. Johnson, R.E., Strain, E.C., Amass, L., 2003. Buprenorphine: how to use it right. Drug Alcohol Depend. 70 (2 Suppl.), S59–S77. Kayman, D.J., Goldstein, M.F., Deren, S., Rosenblum, A., 2006. Predicting treatment retention with a brief “Opinions About Methadone” scale. J. Psychoactive Drugs 38, 93–100. MacCallum, R.C., Widaman, K.F., Zhang, S., Hong, S., 1999. Sample size in factor analysis. Psychol. Methods 4, 84–99. Neira-León, M., Barrio, G., Brugal, M.T., de la Fuente, L., Ballesta, R., Bravo, M.J., Silva, T.C., Rodríguez-Martos, A., Project Itinere Group, 2006. Do young heroin users in Madrid, Barcelona and Seville have sufficient knowledge of the risk factors for unintentional opioid overdose? J. Urban Health 83, 477–496. ˜ ˜ Pérez de los Cobos, J., Sinol, N., Trujols, J., Banuls, E., Batlle, F., Tejero, A., 2011. Drug-dependent inpatients reporting continuous absence of spontaneous drug craving for the main substance throughout detoxification treatment. Drug Alcohol Rev. 30, 403–410. Pérez de los Cobos, J., Trujols, J., Valderrama, J.C., Valero, S., Puig, T., 2005. Patient perspectives on methadone maintenance treatment in the Valencia region: dose adjustment, participation in dosage regulation, and satisfaction with treatment. Drug Alcohol Depend. 79, 405–412. Pérez de los Cobos, J., Valero, S., Haro, G., Fidel, G., Escuder, G., Trujols, J., Valderrama, J.C., 2002. Development and psychometric properties of the Verona Service Satisfaction Scale for methadone-treated opioid-dependent patients (VSSS-MT). Drug Alcohol Depend. 68, 209–214. Peterson, J.A., Schwartz, R.P., Mitchell, S.G., Reisinger, H.S., Kelly, S.M., O’Grady, K.E., Brown, B.S., Agar, M.H., 2010. Why don’t out-of-treatment individuals enter methadone treatment programmes? Int. J. Drug Policy 21, 36–42. Preston, K.L., Sullivan, J.T., Strain, E.C., Bigelow, G.E., 1996. Enhancement of cocaine’s abuse liability in methadone maintenance patients. Psychopharmacology (Berl.) 123, 15–25. Reisinger, H.S., Schwartz, R.P., Mitchell, S.G., Peterson, J.A., Kelly, S.M., O’Grady, K.E., Marrari, E.A., Brown, B.S., Agar, M.H., 2009. Premature discharge from methadone treatment: patient perspectives. J. Psychoactive Drugs 41, 285–296. Schwartz, R.P., Kelly, S.M., O’Grady, K.E., Mitchell, S.G., Brown, B.S., 2011. Antecedents and correlates of methadone treatment entry: a comparison of out-of-treatment and in-treatment cohorts. Drug Alcohol Depend. 115, 23–29. Shikiar, R., Rentz, A.M., 2004. Satisfaction with medication: an overview of conceptual, methodologic, and regulatory issues. Value Health 7, 204–215. Sitzia, J., Wood, N., 1997. Patient satisfaction: a review of issues and concepts. Soc. Sci. Med. 45, 1829–1843. Sweileh, W.M., Ihbesheh, M.S., Jarar, I.S., Sawalha, A.F., Abu Taha, A.S., Zyoud, S.H., Morisky, D.E., 2012. Antipsychotic medication adherence and satisfaction among Palestinian people with schizophrenia. Curr. Clin. Pharmacol. 7, 49–55. ˜ Trujols, J., Iraurgi, I., Sinol, N., Portella, M.J., Pérez, V., Pérez de los Cobos, J., 2012. Satisfaction with methadone as a medication: psychometric properties of the Spanish version of the treatment satisfaction questionnaire for medication. J. Clin. Psychopharmacol. 32, 69–74. Trujols, J., Iraurgi, I., Oviedo-Joekes, E., Guàrdia-Olmos, J.A., 2014. A critical analysis of user satisfaction surveys in addiction services: opioid maintenance treatment as a representative case study. Patient Prefer. Adherence 8, 107–117. ˜ Trujols, J., Portella, M.J., Iraurgi, I., Campins, M.J., Sinol, N., Pérez de los Cobos, J., 2013. Patient-reported outcome measures: are they patient-generated, patientcentred or patient-valued? J. Ment. Health 22, 555–562.
Please cite this article in press as: Cobos, J.P.d.l., et al., Development and validation of the scale to assess satisfaction with medications for addiction treatment-methadone for heroin addiction (SASMAT-METHER). Drug Alcohol Depend. (2014), http://dx.doi.org/10.1016/j.drugalcdep.2014.05.024
