Psychosocial Determinants of Medication Adherence Among Psychiatric Outpatients

Medication adherence refers to compliance by patients in taking their prescribed medications as instructed by their doctors. Lack of treatment adherence has shown to be associated with negative attitude to pharmacotherapy. This study was set out to determine the influence of psychosocial factors on medication adherence among outpatient of Federal Neuro-psychiatric Hospital Yaba, Lagos. The study adopted survey design with a randomized sample of 277 participants selected purposively. The participant responded to Big Five Personality Inventory (BFI), Religiosity Oriented Test (ROT) and Morisky Medication Adherence Scale (MMAS-8). Data was analyzed using descriptive and inferential statistics. The result revealed a 44.8% and a 55.2% prevalence of high and low medication adherence respectively. Observed significant independent determinants of medication adherence among the outpatients include personality traits (R2 =.034 p = .000) and religiosity (R2 =.037 p = .001). Personality traits and religiosity jointly predicted medication adherence (R2=.024, p= .000). Respondents’ nature of illness (F (4,273) = 8.81, p=. 000) as well as level of education (F (5,271) = 5.964, p = .000) significantly influenced medication adherence. Findings concluded that personality traits, religiosity, nature of illness and educational level are psycho-social determinants of medication adherence among outpatients. Authors recommend psychoeducation to enlightenment of patients on the importance of adherence to medication.

and extraversion. According to Tsang and McCullough (2003) some forms of religiosity correlate significantly with physical and mental health, tolerance, pro-social behaviour and positive interpersonal relationships. However, McCullough, et al., (2003) reported a positive significant correlation between religiousness and openness to experience.
Medication non-adherence is an important barrier to the successful treatment of the mentally ill and is a frequent cause of exacerbations in psychopathology, psychotic relapse and re hospitalization (Ascher-Syanum, et al., 2006). The association between socio-demographic factors, symptoms severity and type of medication has remained controversial, (Lacro et al. 2002). In an effort to ultimately increase adherence, researchers have examined both the static (i.e., fixed) and dynamic (i.e., variable) factors that most strongly affect overall compliance in individuals with mental illness. Studies have explored numerous static variables such as gender (Atwood & Beck, 1985;Buchanan, 1992), ethnic minority status (Tunnicliffe, Harrison, & Standen, 1992), age of illness onset (Leclerc, et al., 2013), type of psychotic symptoms (Duncan & Rogers, 1998), and level of formal education (Barkof, et al., 2011;Leclerc, et al., 2013). Cooper, Carpenter, Katona, Schroll, Wagner, Fialova, and Livingston (2005) reported that married individuals had higher treatment adherence levels which were attributed to perceived support from spouses. Wang et. al. (2002) found higher medication adherence levels in females than males.
The Theory of Reasoned Action (Fishbein, 1967), suggests a patient's knowledge and ability to obtain knowledge influences the development of their beliefs. When faced with self-management tasks, patients weigh many factors, including their beliefs about the necessity or concerns about medicines which ultimately leads to the decision to adhere or not to adhere to the prescribed treatments (Horne & Weinman, 1999). In an examination of the influence of belief about medicines on adherence, Emilson and colleagues (2011) found a positive relationship between necessity of medication for controlling illness and adherence. Beliefs about medicine has shown to be an important factor that most affects medication adherence in individuals with diabetes (Horne, 1997;Horne & Weinman, 1999, & Horne, 2006. A study examining belief about medicines and adherence in patients with type II diabetes carried out by Sweileh, Zyoud, Abu Nab', Deleq and Al-Jabi, (2014) demonstrated that the participants who believed their medicines to be of concern and anxiety were shown to be non-adherent to treatments.
Medication non-compliance is a major risk factor for relapse in mental illness (Marder, 1998;Geddes et al 2003). It is estimated that fewer than 10% of people with unipolar major depression take the appropriate therapeutic dose of medication, and more than 20% fail to even fail their first prescription (Keller & Boland 1998). Furthermore, people often discontinue their medication long before it is advised by their medical practitioner. Adherence to taking antipsychotic medication is particularly problematic, and a wide range of factors (Marland & Cash 2001;Sweileh, et al, 2014).
In Nigeria, the subject of mental health is often culturally evaded as many people are not usually inclined to discuss it openly (Bakare, 2014). Unlike in western world there is paucity of studies on medication adherence carried out in the developing world. This is not surprising taking into cognizance the ignorance, myth solely based on tradition and superstition, negatively shape the attitude of most people towards the etiology and solutions of most illnesses, especially mental disorders in this part of the world (Chukwujekwu, & Stanley, 2008). There is a dearth of information on medication adherence among psychiatric patients especially in the area of the psychological and social factors in Nigeria. Hence, this present study explores some extraneous psychosocial factors that influence medication adherence among outpatients of Neuro-psychiatric Hospital Lagos Nigeria. Our focus includes ascertaining if personality traits and religiosity significantly predicts medication adherence, establishing the influence of nature of illness as well as educational background on medication adherence among outpatients of Federal Neuro-psychiatric Hospital (NPH) Yaba, Lagos Nigeria.

Research Questions
In the light of the aforementioned problem gaps, the following research questions will this study: 1. Does personality traits significantly determine medication adherence among outpatients of NPH Yaba Lagos Nigeria? 2. To what degree will religiosity significantly determine medication adherence among the participants? 3. Does nature of illness significantly influence medication adherence among the psychiatric outpatients? 4. Will educational level significantly influence medication adherence among outpatients of Federal Neuropsychiatric Hospital Yaba Lagos Nigeria?

Participants
The target population was outpatient in Neuro-psychiatric Hospital (NPH), Yaba Lagos, Nigeria. Only outpatients who were above 18 years, who could read and understand English language were purposively selected using a cross sectional approach to participate in this study. Outpatients who were fondly psychotic patients were excluded from the study. Using the sample size determination formula by Araoye (2004) a total of 277 outpatients were selected.

Measures
A battery of three instruments was adapted and used as tools for data collection. Big-Five Inventory (BFI) by John, Donahue, & Kentle, (1991). The version used is a forty four (44) item scale graded on a five-degree scale from completely disagree=1 to completely agree=5. Cronbach's Alpha coefficients for the five factors are neuroticism r= 0.78, extraversion r = 0.61, conscientiousness r = 0.68, agreeableness r = 0.74 and openness r = 0.75. The BFI has been used on Nigerian populations and was found to have similar psychometric properties when compared to western population (Omoluabi 2002, Okhakhume 2015, Ugalahi 2017). In a validation study Umeh (2004) highlighted a comparative report of American and Nigerian population; this is summarized in table 1. Comparison table of BFI mean and standard deviation of American and Nigerian population (Umeh, 2004) as cited by Oladimeji (2018).
Religious Orientation Test (ROT) a six-item scale that identifies two types of religious orientations: superficial religious orientation and deep religious orientation. The scale has been found to demonstrate good internal consistency. Idehen (2001) used the ROT on a sample of 160 and found alpha values between 0.76 and 0.80 and an internal consistency Cronbach alpha of 0.80, with a test retest reliability correlation of 0.75.
Morisky Medication Adherence Scale (MMAS-8) is an 8-item instrument developed by Morisky et.al (2008) was used to assess medication adherence among patients with different clinical conditions. MMAS-8 has been validated for use in Nigeria (Abiola, Udofia, & Zakari, 2013). It has shown good psychometric properties, with Cronbach's alpha of 0.77 and internal consistency reliability of α = 0.75

Data Analyses
The data was analyzed using SPSS package 23. Descriptive statistics (Simple percentages) and inferential statistics (One way ANOVA, and linear regression) were used for this study.

Patterns of Medication Adherence
As presented in the Table 2, out of two hundred seventy-seven 277 respondents in study 125 (44.8%) reported high medication adherence while 152 (55.2%) reported a low medication adherence. 60 independent variables on medication adherence. The result summarized on Table 3 indicated that there was significant joint influence of personality traits on medication adherence, [F (5,277) = 27.87, R 2 = .034; p =.000] with the variables accounting for 3.4% of the variance in medication adherence. Further result show that extraversion (β=.13; t= 2.01), agreeableness (β= -.27; t = -4.46) and conscientiousness (β= -.17; t = -7.09), had significant independent beta contributions in predicting medication adherence among the outpatients. A simple linear regression analysis was carried out to determine the degree to which religiosity independently and significantly predicted medication adherence among the participants. Result indicated that religiosity independently significantly predicted the medication adherence among the sample, (R² = .037, p = .001). The analysis summarized in Table 4 suggests that 3.7% variance of medication adherence is explained by religiosity among the patients. .001 Hypothesis Three: Nature of illness will significantly influence medication adherence among outpatients of Federal Neuro-psychiatric hospital Yaba.
A one-way between subjects ANOVA was conducted to compare the medication adherence of the outpatients of Federal Neuro-psychiatric hospital Lagos Nigeria based on nature of illness categories. Their nature of illness were classified in five groups namely schizophrenia, bipolar, major depression disorder, substance use, and others. The independent variable represented the five different categories, while the dependent variable was medication adherence scores. Table 5 is a summary of means and standard deviations for each of the five identified nature of illness categories.  Table 6 show the analysis of variance for the medication adherence of the outpatients based on the nature of their illness categories. There was a significant difference in medication adherence of the respondents based on the nature of their illness at the p<.05 level for the five levels [F (4,273) = 8.81, p = .000]. Post hoc comparisons using the Tukey HSD test indicated that the mean score for schizophrenia category (M = 15.00, SD = 1.47) was significantly different than Bipolar category (M = 14.00, SD = 1.48), substance use category (M = 13.92, SD = 1.32), and others category (M = 13.92, SD = 1.37). However, major depression disorder category (M = 15.04, SD = 1.66) did not significantly differ from schizophrenia category. Taken together, these results suggest that nature of illness really do have significant influence on medication adherence. Specifically, our results suggest that among the outpatients using the Yaba Psychiatric Hospital Lagos, schizophrenia and major depression disorder patients adhere less to medication than do the bipolar, substance use disorder and other categories. Hypothesis Four: Educational level will significantly influence medication adherence among outpatients of Federal Neuro-psychiatric hospital Lagos, Nigeria.
A one-way between subjects ANOVA was conducted to compare the medication adherence of the outpatients of Federal Neuro-psychiatric hospital Lagos Nigeria based on educational level categories. Their highest educational background were classified in six groups namely Primary education, Junior Secondary School Certificate (JSSC), Senior Secondary Certificate Education (SSSCE), Ordinary National Diploma (OND), Bachelors /Higher National Diploma (BSc/ HND), Master's degree / PhD. The independent variable represented the six different categories, while the dependent variable was medication adherence scores. Table 7 is a summary of means and standard deviations for each of the six identified educational qualification categories.  Table 8 show the analysis of variance for the medication adherence of the outpatients based on their educational level categories.  . Taken together, these results suggest that educational level really do have significant influence on medication adherence. Specifically, our results suggest that among the outpatients using the Yaba Psychiatric Hospital Lagos, primary education level, Junior secondary school education and Ordinary National Diploma educational level holders adhere less to medication than do the SSSCE, HND/BSc and Master's/PhD categories.

Discussions
The research finding of our first hypothesis implied that each of the dimensions of personality traits interact together in determining level of outpatient medication adherence. This finding supports Booth-Kewley and Vickers, (1994) who reported that the personality traits of agreeableness and conscientiousness have positive effects on health behavior. In addition, Christensen and Smith, (1995) found conscientiousness to be positively associated with adherence to medication regimens.
The result of our second hypothesis revealed that religiosity is a significant independent predictor of medication adherence among outpatients of Federal Neuro-psychiatric Hospital Yaba Lagos. This is in support of Koenig et al. (2012), who affirmed that religious beliefs may predict medication compliance in mental illness to a large extent; he further affirmed that more religious persons seem to be the more likely of such to involve in disease-screening practices and comply better with prescribed medical treatments. Added to that, spiritual needs are common phenomenon among people for instance; religious beliefs influence medical decision making, drugs adherence (Brink, 1993).
Result of our third hypothesis revealed that nature of illness had significant influence on medication adherence. Schizophrenia and major depression disorder outpatients adhere less to medication than do the bipolar, substance use disorder and other categories. Studies show that between a third and half of medicines that are prescribed for long term conditions are not used as recommended (Haynes, McDonald, Garg, & Montague, 2002). Researchers have established significant relationship between psychiatric symptom severity and treatment adherence (Duncan & Rogers, 1998;Tsang et al., 2010), such that as the severity of mental illness symptoms increases, medication adherence decreases (Kampman, et al., 2002;Tsang et al., 2010). Among patients with Schizophrenia, those with more severe positive and negative symptoms may be less likely to comply with recommended treatment (Macpherson, Jerrom & Hughes 1997).
Educational level was found to significantly influence medication adherence among the psychiatric outpatients. Participants with primary education level, Junior secondary school education and Ordinary National Diploma educational level holders were observed to adhere less to medication than do the SSSCE, HND/BSc and Master's degree / PhD categories. Patient's level of education had been reported as a risk factor to medication adherence since poor understanding of the medical instructions, recommendations, or steps to follow adequate treatment does not allow successful healthcare completion with the serious and negative impact that this could have in the patient's wellbeing (Villafuerte-Galvez, J., et al, 2015).

Conclusions and Recommendations
From the analysis and interpretations of our findings the following conclusions can be deduced. First, personality traits are significant predictors of medication adherence among psychiatric outpatients. Secondly, religiosity is a significantly independent predictor of medication adherence among the participants. Finally, significant difference in nature of illness as well as difference in educational level on medication adherence exists among outpatients of Federal Neuro-psychiatric Hospital Yaba Lagos Nigeria. Psycho-education on the importance of adherence to medication should be given to outpatients on each visit to psychiatric hospital.

Consent and Ethical Approval
The research intention and proposed procedures for carrying the research was subjected to scrutiny by the Internal Research Ethic Committee (IREC) of Redeemer's University, Ede, Osun State Nigeria and approval granted before the study was embarked upon. A letter of approval was equally obtained from the, Research Ethics Committee of the Federal Neuro-psychiatric Hospital Yaba Lagos Nigeria. The approval was taken to the head of the outpatient ward by the researchers. Respondents who were available and willing to be part of the study on each clinic day were used for this study. The respondents were made to sign an informed consent form which was detached from the questionnaire immediately in order to keep their anonymity. The questionnaires were distributed to them and were given a day to respond to it. Participants were further informed that they could withdraw at any time from the study without any penalty. For confidentiality and other ethical considerations, each respondent was made to sign a consent form. The respondents were given adequate information about the research and were by no means coerced to participate. They all choose to participate on their own volition and were also given the opportunity to ask questions after brief education on what the study is all about in other to diffuse any hidden misconception.