Prevalence and predictors of sub-optimal medication adherence among patients with severe mental illnesses in a tertiary psychiatric facility in Maiduguri, North-eastern Nigeria

Introduction Sub-optimal adherence constitutes a significant impediment to the management of severe mental illnesses (SMIs) as it negatively impacts on the course of the illness and the treatment outcome. In this study, the levels of adherence, prevalence and the predictors of sub-optimal adherence were assessed in a sub-Saharan African setting. Methods Three hundred and seventy (370) respondents with diagnoses of schizophrenia, bipolar disorder or severe depression were randomly enrolled and interviewed at the out-patient department of the Federal Neuropsychiatric Hospital, Maiduguri in northeastern Nigeria. An anonymous sociodemographic questionnaire and a clinical proforma designed by the authors, Oslo social support scale and the 8-item Morisky Medication Adherence Scale (MMAS-8) were used for data collection. Results The prevalence of sub-optimal adherence was 55.7%. The independent predictors of sub-optimal adherence were; seeking for traditional/ spiritual treatment (Odds Ratio (O.R.) = 6.523, 95% C.I. = 3.773 - 11.279, P = < 0.001), male gender (O.R. = 3.307, 95% C.I. = 1.907 - 5.737, P = < 0.001), low levels of insight (O.R. = 1.753, 95 C.I. = 1.220 - 2.519, P = 0.002), and low social support levels (O.R. = 1.528, 95% C.I. = 1.097 - 2.129, P = 0.012). Conclusion Based on the outcome of the study, we recommend the development of psycho-educational programmes on adherence and the active involvement of the relations and significant others in the management of patients with SMIs in sub-Saharan Africa.


Introduction
The severe mental illnesses (SMIs) are characterized by a chronic, and sometimes, fluctuating course as well as their association with significant functional impairments/disabilities [1][2][3][4]. The main aims of managing such disorders are to achieve and maintain remission of symptoms and, to reduce the accompanying impairments [5].
Psychotropic medications together with other non-pharmacological modalities constitute the cornerstones in the management of these disorders as espoused by the biopsychosocial model [6]. Since the SMIs are chronically disabling, their management specifically with medications are also longstanding and in some cases lifelong. A key determinant of the success of pharmacotherapy in patients with SMIs is their adherence to the medication regimen, which is defined as the extent to which medication intake behaviour corresponds with the recommendations of the health care provider [7][8][9][10].
Hence, non-adherence to medications (treatment) is the degree to which the patient does not carry out the clinical recommendations of a treating physician. Non-adherence to medications is a complex and multidimensional healthcare problem as it constitutes a major obstacle to translating treatment efficacy in research settings into effectiveness in clinical practice [11][12][13]. Research evidences abound that have shown its clinical significance on the course of the illness and treatment outcomes [14][15][16]. In addition, there can be a profound impact on the cost of care, as well as significant impediments to the patients´ long term adaptations, including the social, vocational and academic functioning [17,18].
In unipolar depression, the estimated non-adherence rates ranged from 13% to 52.7% [28][29][30][31]. The reported rates for non-adherence were 41.2% in Ethiopia and 74% in Egypt on the African continent [32,33]. In Nigeria, the recorded rates of non-adherence among patients with SMIs ranged from 48% to 55.5% in Southern Nigeria, and 49.4% in Kaduna and 34.2% in Jos, North-western and Northcentral Nigeria respectively [34][35][36][37]. Sub-optimal adherence in psychiatric populations may be attributed to multifactorial influences Such as; age, gender, poor insight, negative attitude towards medications, shorter duration of illness, poor therapeutic alliance and poor social support [19,21,27,34]. Africans, have their psychological and social peculiarities with regards to mental illnesses and such variables could affect their adherence to pharmacotherapy in the context of severe mental illnesses. Each African setting has its own peculiarities in terms of belief systems and social support as well as their effects on health seeking and adherence. To the best of our knowledge, this is the first study from North-eastern Nigeria that addresses this topical issue among patients with SMIs as a prototype sub-Saharan setting. It sought to ascertain; 1) the levels of adherence to pharmacotherapy among patients with severe mental illnesses and 2) determine the sociodemographic and clinical predictors of sub-optimal adherence among the subjects.

Study design and setting
This was a hospital-based cross-sectional study conducted at the outpatient department of the Federal Neuropsychiatric Hospital, Maiduguri, Borno State, Nigeria. As a matter of policy, all diagnoses made in the institution were according to the tenth edition of the International Classification of Diseases and health-related disorders (ICD -10) criteria. Clinically generated data for each subject enrolled were matched to the ICD -10 criteria by Consultant Psychiatrists for quality assurance purposes.

Participants
The sample size was calculated using a prevalence rate of 49.4% for non-adherence among patients with mental illnesses in northern  asked and a score of 1 is assigned for yes and a score of zero for no for any item answered: Do you accept that you have an illness? Do you think that you require treatment? And, do you think you require your medications to stay well? The total score range between 0 and 3. A total score of zero is regarded as "no insight", total score of 1-2 is "partial insight", and a total score of 3 is interpreted as "full insight". which is an 8-item instrument which was developed by Donald Morisky to assess medication adherence among patients with different clinical conditions. In interpreting the outcome, the scores are graded as follows: <6 is low adherence, 6 to <8 is medium adherence and high adherence is equal to 8 [41]. For the purpose of this study, sub-optimal adherence is defined as MMAS-8 score of less than 8 while score of 8 is considered adherent. Similar cut-off score was adopted for non-adherence among hypertensive outpatients with comorbid psychiatric conditions in Ghana [42].

Ethical consideration
Ethical clearance was obtained from the institutional review board of Federal Neuropsychiatric Hospital, Maiduguri. Written informed consents were also obtained from all the participants. In order to ensure confidentiality, codes were used for data entry and analysis.

Data analyses
The data were analyzed using statistical package for social sciences (SPSS) version 20. Descriptive statistics were used to represent the characteristics of the participants. Bivariate analyses were used to explore the associations between the psychosocial variables and sub-optimal medication adherence among the participants. Binary logistic regression was then conducted to determine the independent predictors of sub-optimal adherence among the subjects. Sub-optimal adherence was used as independent variable while the factors found to be significant on bivariate analysis were used as covariates. Significance was computed at p < 0.05, twotailed.

Results
Of the 390 respondents enrolled, the data of 370 respondents were finally analyzed yielding an overall response rate of 94.87%. The data of 20 respondents were not analyzed due to: refusal to grant informed consent (n=8), comorbid psychoactive substance use (n=5), presence of florid psychopathology, particularly, auditory hallucinations and delusions (n=5), and those whose questionnaires could not be analyzed due to missing data (n=2). In terms of the sociodemographic characteristics of the respondents: males constituted 56.5%, their ages ranged from 18 to 63 years with an average of 35.06 years (SD+ 9.63). Over 70% of the respondents Page number not for citation purposes 4 were ≤ 40 years of age. Semi-skilled and unskilled workers as well as the unemployed constituted over 60% of the respondents and about 51% were unmarried. About 43% of the respondents believed that mental illnesses were caused by either demonic possession or retribution for ´sins´ committed by one´s ancestors.
The clinical characteristics of the respondents revealed that about 63% 0f the respondents had their illnesses for less than or equal to 4 years and nearly 56% of them had sought for either traditional African/spiritual forms of treatment apart from the conventional (orthodox) care. About 31% had poor social support while 69% had moderate to strong social support. About 58% had full insight, while, about 18% and 24% had no and partial insights respectively.

Levels of adherence and prevalence of sub-optimal adherence
One hundred and sixty eight (45.4%), thirty eight (10.3%), and one hundred and sixty four (44.3%) had low, medium and high levels of medication adherence respectively. The prevalence rate for suboptimal adherence was 55.7% using an MMAS score of <8 as adapted for this study. The findings are depicted in Table 1.

Socio-demographic variables associated with sub-optimal adherence
Of all the sociodemographic variables analysed for association with sub-optimal medication adherence only gender (χ 2 =50.415, df=1, p=<0.001) and belonging to lower occupational classes (χ 2 =57.93, df=4, p=<0.001) were found to be statistically significant. These are presented in Table 2.

Logistic regression analysis for variables associated with sub-optimal adherence
Logistic regression analyses revealed that only gender (Odds ratio   They may also be better motivated to adhere to treatment protocols in order to retain their jobs. The independent clinical predictors of sub-optimal adherence in this study were; diagnoses, forms of treatment sought, levels of social support and insight. Based on the outcome, patients with a diagnosis of schizophrenia were over 3 times more likely to be sub-optimally adherent than those with either bipolar disorder or unipolar depression. This might be attributed to residual symptomatology and difficulties with interpersonal adjustment that may be commoner in schizophrenia than either bipolar disorder or unipolar depression as reported by In terms of social support as an independent predictor, those with low levels of social support were about 2 times more likely to be sub-optimally adherent. This finding is in tandem with that of

Limitations of the study
The limitations of the study were; (1) this is a questionnaire based study; objective methods of assessing adherence such as pill counts and metabolite bioassay could have been more reliable indicators of sub-optimal adherence (2) the cross-sectional nature of this study cannot permit drawing causal inference.
Page number not for citation purposes 6

Conclusion
Over half of the participants in this study were sub-optimally adherent to their medications we, therefore, recommend: (1) developing psycho-educational programmes that will address the misconceptions about the aetiology and spectra of symptoms associated with severe mental illnesses, (2) the active involvement of relatives and significant others in treatment planning and management of the patients with SMIs.

Competing interests
The authors declare no competing interests.