|Year : 2019 | Volume
| Issue : 9 | Page : 1281-1285
Psychometric evaluation of medication adherence rating scale (MARS) among Nigerian patients with schizophrenia
OA Sowunmi, PO Onifade
Department of Clinical Services, Neuropsychiatric Hospital Aro, Abeokuta, Ogun, Nigeria
|Date of Acceptance||02-Mar-2019|
|Date of Web Publication||6-Sep-2019|
Dr. O A Sowunmi
Consultant Emergency Psychiatrist, Neuropsychiatric Hospital, Aro, PMB 2002, Abeokuta, Ogun
Source of Support: None, Conflict of Interest: None
| Abstract|| |
Background: The Medication Adherence Rating Scale (MARS) is a 10-item self-report measure of medication adherence in psychosis which is a vital predictor of illness course and outcome in patient with schizophrenia. The initial and subsequent studies have shown that MARS has good reliability and validity scores after correction for the small sample size in the index study. Aim: This study aimed to determine the psychometric properties of MARS among outpatients with schizophrenia at the outpatient clinic of the Neuropsychiatric Hospital Aro Abeokuta Ogun State Nigeria. Methods: Intra-class correlation coefficient (ICC) was used to determine the internal consistency, item-total correlations, and reliability of the instrument. Factor analysis was done using principal component analysis with varimax rotation. Results: The intra-class correlation coefficient (ICC) for these 10 items (at time T1) was 0.6 with a P value of <0.001 while for the test--retest analysis was 0.7 with a P value of 0.04. A principal components factor analysis with varimax rotation produced a four-factor solution and factor 4 was found to be the most internally consistent, with Cronbach's alpha of 0.63. Conclusion: This study supports the internal consistency, test--retest reliability, and constructs validity of the MARS.
Keywords: Adherence, MARS, Nigeria, schizophrenia
|How to cite this article:|
Sowunmi O A, Onifade P O. Psychometric evaluation of medication adherence rating scale (MARS) among Nigerian patients with schizophrenia. Niger J Clin Pract 2019;22:1281-5
|How to cite this URL:|
Sowunmi O A, Onifade P O. Psychometric evaluation of medication adherence rating scale (MARS) among Nigerian patients with schizophrenia. Niger J Clin Pract [serial online] 2019 [cited 2020 Aug 14];22:1281-5. Available from: http://www.njcponline.com/text.asp?2019/22/9/1281/266162
| Introduction|| |
The use of psychotropic medications are primarily indicated for the treatment of mental disorders including schizophrenia and other psychotic and non-psychotic mental disorders. These medications are effective in treating symptoms associated with several mental disorders especially if adherence to medication is optimal. Medication adherence may be defined as the degree to which a patient's medication use complies or aligns with the recommendations and advise of the prescribing physician while ''nonadherence'' includes both underuse and overuse of medication prescribed.
In previous studies,,, ''medication nonadherence'' is considered to be an ''all or none'' phenomenon where the patient is ''nonadherent'' if they miss 20% of the medication, which may lead to increased risk of relapse or (re-) hospitalization.
However, adherence and nonadherence with these medications is a dynamically changing behaviour that contributes significantly to resolution of symptoms and reintegration into the society. When compared with no medication, psychotropic medications significantly reduce relapse rates in patients with psychiatric disorders who are stabilized on medications over 1 year.
Unfortunately, nonadherence is not uncommon and medication nonadherence in patients with psychiatric morbidities are reported to range from 20% to 89%., More than 50% of the patients with mental disorders, especially schizophrenia, become partially adherent or nonadherent in 1 year, and about 75% in 2 years.,
From the foregoing, it is paramount that medication adherence be monitored effectively. There are three main methods of doing this, namely, patient and clinical self-report, pill counts, and biological measures. Self-report methods are generally the most cost-effective and time-efficient way of obtaining an indication of adherence, although such methods have been reported to overestimate adherence by 30%. It may also be dependent on the phrasing of the question., In comparison, pill counts have the disadvantage of undermining the therapeutic relationship and are also unreliable since there is no guarantee that the tablets removed from the container were actually consumed. Biological measures are less frequently used due to their cost and limited availability.,, The accuracy of biological measurement is compromised by individual differences in metabolism, and is dependent on the period of time between ingestion of the last tablet and testing. None of these methods of adherence measurement is completely accurate, and the concordance rate across these different measures has been reported to be low. Self-report measures are perhaps the easiest to administer and the least costly, whereas biological methods are more objective and expensive but no more accurate.,
One of the common use of self-report questionnaire for monitoring of medication adherence is the Medication Adherence Rating Scale (MARS) which was developed by Thompson et al. in 2000 to be a quick, nonintrusive, and a simple self-reporting measure of medication adherence especially in psychiatric.,,, Despite the wide use of the instrument in Nigeria, it has few published psychometric properties in Nigeria., Thus, the aim of this study was to determine the validity and reliability of MARS among outpatients with schizophrenia at the outpatient clinic of the Neuropsychiatric Hospital Aro Abeokuta Ogun State.
| Methodology|| |
The study centre was Neuropsychiatric Hospital, Aro Abeokuta Ogun State, Nigeria. The hospital has a total bed capacity of 546 for inpatient care. The hospital plays a strategic role in mental health care for patients from all parts of Nigeria and neighbouring countries. All new patients are seen at the Assessment/Emergency unit of the hospital from where they are transferred to either the outpatient clinic or the wards. Those discharged from the wards are also subsequently followed-up at the outpatient clinic. No new patient is seen at the outpatient clinic of the hospital. The clinic runs on Mondays, Tuesdays, Thursdays, and Fridays. About 130 patients were seen daily at the outpatient clinic. Of these, 54 (41.9%) are patients with a diagnoses of schizophrenia.
Sample size determination
The sample size (n) calculation and oversampling was calculated as proposed by Cochrane. The calculation gave a total sample size (n) of 220.
The MARS is a 10-item self-report instrument in which a yes/no response is given to questions asked. It was developed from two previous scales, the 30-item Drug Attitudes Inventory (DAI) by Hogan et al. and the four-item Morisky Medication Adherence Questionnaire (MAQ). The aim was to develop a more reliable and valid tool for assessing medication adherence behaviour in patients with psychosis. The total scores range from 0 (low likelihood of medication adherence) to 10 (high likelihood). This shows an understanding that adherence is a continuous variable and that an individual can reach a decision anywhere between complete adherence and complete nonadherence. Adherence is usually equivalent to scores greater than 5. The 10-item MARS can be found on [Table 1].
The development sample for the MARS consisted of 66 participants with psychosis and the scale showed good internal consistency (alpha = 0.75). Three factors were identified. These were considered to represent “medication adherence behaviour” (items 1-4), “attitude toward taking medication” (items 5–8), and “negative side-effects and attitudes to psychotropic medication” (items 9, 10). A positive correlation of 0.6 between the MARS adherence score and blood lithium levels suggested construct validity.
This is a study of diagnostic accuracy. Participants were selected by systematic random sampling of every 4th patient registered to be seen at the outpatient clinic of the hospital. A total of 220 respondents who had diagnosis of schizophrenia were older than 18 years and were able to read and write in English participated in the first stage. However, for the retest stage of the study, 114 respondents could be reached, the rest were lost to follow-up. Ethical approval was obtained from the research and ethics committee of the hospital and permission was obtained from the managing consultants. Consent was obtained from all participants and attention of the managing consultants was drawn to their corresponding patients who had problems with medication adherence.
| Data Analysis|| |
Data was analyzed using the Statistical Package for Social Science (SPSS version 21) Computer Software. The level of significance was set at P = <0.05. Intraclass correlation coefficient (ICC) was used to determine the internal consistency of the instrument using responses from the first test. ICC was also used to determine the test--retest reliability of the instrument using total score at test and retest. Factor analysis using principal component analysis with varimax rotation was used to confirm the robustness of the original construct. Item--total correlations was also done with intraclass correlation coefficient using two-way mixed method with absolute agreement to determine the concurrent validity of the questionnaire.
| Result|| |
A total of 240 participants were recruited but 20 were excluded because of incomplete questionnaire. The mean MARS score for the 220 participants was 7.09 (SD = 2.08), with a range of 1-10 and the median score of 7. The ICC for these 10 items (at time T1) was 0.6 with a P value of < 0.001. The ICC for the test--retest analysis was 0.7 with a P value of 0.04. Other details of the analysis can be found on [Table 2].
|Table 2: Internal consistence and test-retest analysis of 10-item medication adherence rating scale (MARS)|
Click here to view
A principal components factor analysis with varimax rotation, retaining factors with an Eigenvalue greater than 1, produced a four-factor solution. See [Table 3] and [Table 4]. After rotation, factor 1 (items 1 and 2) accounted for 24.65% of the variance, factor 2 (items 4, 6, 9, and 10) for 12.99%, factor 3 (3 and 5) 12.88%, and factor 4 (7 and 8) for 11.20%. Thus, in total the rotated factor solution accounted for 61.72% of the total variance. The internal consistency of the four factors was examined and factor 4 (which only consisted of two items) was found to be the most internally consistent, with Cronbach's alpha of 0.63. Factor 1 had of 0.57, and factor 2 had a value of 0.55 and factor 3 had a of 0.54.
|Table 3: Principal components factor analysis of 10-item medication adherence rating scale (MARS)|
Click here to view
|Table 4: Rotated component matrix of 10-item medication adherence rating scale (MARS)|
Click here to view
| Discussion|| |
The following range of MARS total scores (1-10) was present in this sample of individuals who attended the outpatient clinic of the Neuropsychiatric Hospital Aro Abeokuta, indicating widely varying attitudes towards taking prescribed medications. Our results show that the MARS is reliable even over a long interval of 12 months, although we had an attrition rate of 48.1%.
The internal consistency of the MARS was moderate (alpha = 0.69), but lower than the value produced by Thompson (alpha = 0.75) and in another study done in Nigeria (alpha = 0.76)., However, it was higher than the value reported by Fialko et al. (alpha = 0.60). Fialko opined that this may not represent a weakness of the scale because instrument like MARS with binary response choice, small number of items, and scale multidimensionality are expected to have lower alpha values. In addition, Fialko et al. reported that it is likely that the internal consistency of the MARS could be improved either by adding more response options or by adding more items, it is debatable whether this would constitute an improvement to the measure, or whether it would compromise its quick, simple format.
This study produced a four-factor component for MARS and the current study also examined the relationships of the individual factor scores. These factors correlation was almost of equal strength to the whole scale correlation. At the development of MARS, three factors were identified. These were considered to represent “medication adherence behaviour” (items 1-4), “attitude toward taking medication” (items 5–8), and “negative side-effects and attitudes to psychotropic medication” (items 9, 10). In our study, factor 1 contained items 1 and 2 which corresponded to the medication adherence behaviour factor of the initial study. Factor 4 contained 7 and 8 which corresponded to the attitude towards medication and factor 2 contained items 4, 6, 9, and 10 which corresponded to the negative side-effects and attitudes to psychotropic medication.
In the initial study, item 6 loaded on the attitude towards medication factor which is also included in the negative side-effects and attitudes to psychotropic medication factor and thus may not be out of place in this study. Furthermore, item 4 loaded on the medication adherence behaviour factor of the initial study but on factor 2 which corresponded to the negative side-effects and attitudes to psychotropic medication of the initial study. It appears that item 4 was viewed more as an attitude rather than a behaviour by the participants of this study and may have influenced the loading observed in this study. It has been earlier suggested  that the finding that attitude may not translate into adherence behaviour may not be surprising when the multiple factors involved in determining adherence behaviour are considered. It is possible that people with a positive attitude towards medication may forget to take it while others with a negative attitude may take medication to please someone else especially those that might be regarded as significant others. Factor 3 (items 3 and 5) appeared to have correlated with insight into illness and insight into the effects of medication.
It therefore appears that where the concern is simply whether or not someone is taking their medication, factor 1 (items 1 and 2) and 4 (items 7 and 8) may be a better indicator than the whole MARS scale. This four item subscale (items 1, 2, 7, and 8) (which corresponds to Morisky's MAQ), from which the MARS was developed is quicker to administer, and appears valid for this purpose.
The current study has a number of limitations. The interval between the initial test and the second test varied across participants which were influenced by various social and clinical factors of the patients determined by the attending psychiatrists. Some patients had as much as 6 month interval between clinic visits. However, this study is not without its strength. The sample was not mixed (containing only outpatients), and was an improvement suggested by an earlier study.
In summary, the MARS is a quick, nonintrusive measure of medication adherence. Items in the MARS about attitude to medication may be informative to clinicians identifying barriers to adherence in individual cases, but do not appear to be valuable in predicting adherence behaviour over a large sample. Factor 1 and 4, corresponding to the MAQ, may be superior for this purpose.
In conclusion, the study supports the internal consistency, test--retest reliability, and construct validity of the instrument.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Lally J, MacCabe JH. Antipsychotic medication in schizophrenia: A review. Br Med Bull 2015;114:169-79.
Staring ABP, Van der Gaag M, Mulder CL. Schizophrenia and antipsychotic medication-Better adherence, better outcomes? Schizophr Res 2013;151:296-7.
Phan SV. Medication adherence in patients with schizophrenia. Int J Psychiatry Med 2016;51:211-9.
Sendt KV, Tracy DK, Bhattacharyya S. A systematic review of factors influencing adherence to antipsychotic medication in schizophrenia-spectrum disorders. Psychiatry Res 2015;225:14-30.
Arute JE, Eniojukan JF, Eboigbe N. Adherence to psychotropic medications by outpatients in psychiatric hospital, Uselu Benin City, Nigeria. Value Heal 2014;7:460-1.
Hogan TP, Awad AG, Eastwood R. A self-report scale predictive of drug compliance in schizophrenics: Reliability and discriminative validity. Psychol Med 1983;13:177-83.
Thompson K, Kulkarni J, Sergejew AA. Reliability and validity of a new medication adherence rating scale (MARS) for the psychoses. Schizophr Res 2000;42:241-7.
Fialko L, Garety PA, Kuipers E, Dunn G, Bebbington PE, Fowler D, et al
. A large-scale validation study of the medication adherence rating scale (MARS). Schizophr Res 2008;100:53-9.
Stirratt MJ, Dunbar-Jacob J, Crane HM, Simoni JM, Czajkowski S, Hilliard ME, et al
. Self-report measures of medication adherence behavior: Recommendations on optimal use. Transl Behav Med 2015;5:470-82.
Owie GO, Olotu SO, James BO. Reliability and validity of the medication adherence rating scale in a cohort of patients with schizophrenia from Nigeria. Trends Psychiatry Psychother 2018;40:85-92.
Cochran WG. The χ2
Test of Goodness of Fit. The Annals of Mathematical Statistics 2013;23:315-45.
[Table 1], [Table 2], [Table 3], [Table 4]