Personality traits and other factors associated with psychotropic medication non-adherence at two hospitals in Uganda. A cross-sectional study

2.1. Aim, design, setting and study population

Our aim was to determine the prevalence of psychotropic medication non-adherence and associated personality traits among patients with mental illness. Before reaching that objective, we first assessed others possible factors (confounders) that could influence medication non-adherence in our study participants. A descriptive cross-sectional study employing quantitative method was conducted from 1^st July to 1^st September 2023 among people with mental disorders attending psychiatric outpatient clinics at Kampala International University-Teaching Hospital and Jinja Regional Referral Hospital. Patients diagnosed with mental disorder and taking psychotropic medications for at least 6 months, aged 18 years old and above were included in the study. Patient with active psychiatric symptoms were excluded from the study.

2.2 Sample size estimation, recruitment, ethical considerations and data Collection Process

Sample size was calculated using the single proportion formula for the present study (Daniel formula, 1995): n = Z² P (1-P)/d² which gave 360 participants. By adding 10% we found a total of 396 patients with mental illness who were recruited from outpatient clinics of Kampala International University-Teaching Hospital and Jinja Regional Referral Hospital. Patients were chosen using simple random sampling method.

Patients who came for psychiatric review at outpatient clinics and fulfilling the inclusion criteria were approached and explained the purpose of the study. Approval to carry out the study was sought and obtained from Research Ethics committee of Bishop Stuart University (the research number is: BSU-REC-2023-99) and it has been conducted according to the principles expressed in the declaration of Helsinki. Before data collection, written informed consent from participants was obtained after fully explaining the details of the study to them in English and local Languages. Before asking consent from the patients, we first assessed their mental capacity using the Functional test of capacity. Participant were not forced to enroll themselves if they don’t want to. Participants were free to withdraw from the study any time he/she wishes without coercion or compromise of care they are entitled to. As patients with mental illness are among the vulnerable population, a written informed consent was also obtained from the next of kin and/ legally authorized representative for study participation.

Those who consented to participate in the study were given a questionnaire with sociodemographic informations, clinical characteristics, Medication Adherence Rating Scale (MARS) and the Ten Item Personality Inventory (TIPI) scale. Identification of participants was by means of codes.

The sociodemographic informations included age, gender, residence, level of education, marital status, employment status and family support. The questionnaire included also the diagnosis of the patient (confirmed by clinical interview using DSM-V-TR diagnostic criteria), associated comorbidities, duration of illness and duration of continuous therapy, number of tablets taken by the patient per day, the frequency as well as side effects of medicines.

Patients’ medication adherence was assessed using Medication Adherence Rating Scale (MARS). The MARS comprised of 10 items with response of yes or no and a total score of 10. A total score then revealed the level of adherence, scoring less than 6 was considered as non-adherence while 6 and above was considered as adherence (17).

To assess the personality traits, we used the Ten-Item Personality Inventory, which is a short version of Big Five Inventory. It consists of ten pairs of adjectives generated using a 7-point likert scale, with 7 (strongly agree) being the highest and 1 being the lowest (strongly disagree) (18). TIPI scoring (‘‘R’’ denotes reverse-scored items): Extraversion (1, 6R); Agreeableness (2R, 7); Conscientiousness (3, 8R); Neuroticism (4, 9R); Openness to Experiences (5, 10R). The score of the personality trait is the average of the 2 items. If the score of that personality trait is above the mean, the person was scored high on that trait. The questionnaire and the tools were pre-tested before being used in the study.

2.3 Data Processing and analysis

After being prepared and encoded, data were entered in Microsoft Excel version 2016 before being exported to Stata 15 for analysis. Proportions, percentages, and frequencies were calculated and presented in a table. Both bivariate and multivariate logistic regression analysis were used to assess personality traits and others factors associated with psychotropic medication adherence. For the multivariate analysis, variables that had a p-value below 0.2 at bivariate analysis were taken into consideration. The odds ratio (ORs) for the association were also presented, along with the appropriate 95% CI and p-values. Factors with a p-value was less than 0.05 were significantly associated with non-adherence. The personality trait for which p-value was less than 0.05 after controlling for the other possible factors was considered as significantly associated with non-adherence to psychiatric drugs.

3.1. Sociodemographic and clinical characteristics of study participants

Males made up 234 (59.1%) of the study participants, with the majority falling into the 18–30 (42.7%) and 31–45 (41.2%) age categories. Majority were from pastoral areas 294(74.2%) and had the mental illness for 1-5 years (71.2%). The other baseline characteristics are shown in table 1(A) and 1(B) below

3.2. Prevalence of psychotropic medication adherence among people with mental illness at KIU-TH and JRRH

In this study, 213 patients were adherent while 183 were non adherent to psychotropic medication translating to 53.79% (95%CI 49.2-58.6) for adherence and 46.21(95%CI 41.4-50.8) for non-adherence respectively.

3.3. Sociodemographic and clinical factors associated with psychotropic medication adherence among people with mental illness at KIU-TH and JRRH

At bivariate level of analysis, the variables that had a p value less than 0.2, and therefore qualified for multivariate analysis were: sex, education level, employment status, marital status, family support, beliefs about mental illness, pill frequency, side effects, chronic disease and substance abuse.

At multivariate level of analysis, the variables that were independently associated with psychotropic medication non-adherence were poor family support, belief that mental illness is caused by witchcraft/sorcery, having side effects and substance use. The rest of the details of multivariate analysis are shown in table 2 below.

3.4. Personality traits associated with psychotropic medication adherence among people with mental illness

At bivariate analysis, only neuroticism and agreeableness were the two personality with p value less than 0.2 and qualified for multivariate analysis (Table 3).

At multivariate analysis after controlling for potential confounders which included absence of family support, belief that mental illness is caused by witchcraft/sorcery, having side effects and substance use, agreeableness and neuroticism personality traits remained significantly associated with non-adherence. A participant with personality trait neuroticism was 7.424 times more likely to be non-adherent (aOR=7.424, CI=3.890-14.168, P<0.001). A participant with personality trait agreeableness was 0.062 times less likely to be non-adherent (aOR=0.062, CI=0.024-0.160, P<0.001). This means that a participant with personality trait agreeableness was 16.12 (1/0.062) times more likely to be adherent (Table 4).

Personality traits associated with psychotropic medication non-adherence

In our study, the two personality traits that were significantly associated with drug non-adherence after controlling for the significant sociodemographic and clinical factors were Neuroticism and Agreeableness.

In our study participant with neuroticism traits were more likely to be non-adherent to psychotropic medications. Persons with predominant neuroticism personality traits are likely to have instability in emotions, self-doubt, irritability, anxiety, and depression. They experience generally negative emotions which can lead to poor adherence to medications. Our results were in line with other studies done in different parts of the world (21). However, a study done in Italy using the short version of Big Five Inventory (BFI-10) to assess personality traits has found that conscientiousness was also associated with non-adherence to medication (21). The difference can be explained by the small simple size (13 participants) and the longitudinal nature of the study.

Agreeableness was also significantly associated with psychotropic medication non-adherence. A participant with predominant agreeableness traits was less likely to be non-adherent to medications. The patients with predominant agreeableness personality traits tend to agree and follow other’s opinions which can help to adhere to instructions given by health workers and as result the increase in the level of adherence to medications.

This is similar to the results of a study done in Sweden who found that agreeableness influences positively the adherence behavior (21) and a systematic review done in India, studies showed that Agreeableness was among the personality traits negatively associated with non-adherence for patients with chronic conditions (21).

However, in USA, the most personality traits associated with non-adherence among participants with bipolar affective disorder were extraversion, openness and conscientiousness (21). This difference was due to the study population (our study was done in all patients with different mental disorders not only in patients with bipolar affective disorder) and also, he used the revised Five Factor Inventory (Neo-FFI) with 60 items. In U.K only conscientiousness was the personality trait influencing non-adherence behavior (21).

This difference may be explained by the different versions of the Big Five Inventory used to assess personality traits, the methodology used in the study (longitudinal and multicenter) but also the sample size (1062 patients).

Other factors associated with psychotropic medication non-adherence were significantly associated with medication non-adherence were poor family support, belief that mental illness is caused by witchcraft/sorcery, presence of side effects and substance use.

Poor family support had negative influence on medication adherence. In Uganda where there is lack of social protection agencies, families are the principal sources of support and many families in low-income countries have social and financial difficulties (22). Family support improve medication adherence by taking care of the patient, reminding him to take medicine, providing basics needs and money for transport to hospital or for medicine. The false beliefs about causes of mental illness influenced psychotropic medication non-adherence. The beliefs about the causes of mental illness influence the choice to adhere to modern treatment (23) which lead to poor compliance to biological treatment. The false beliefs about the causes of mental illness lead to poor insight about illness and poor insight leads to psychotropic medication non-adherence (24). The presence of side effects of medication was significantly associated with medication non-adherence in our study. Individuals who experienced side effects were more likely to be non-adherent compared to those without side effects. Patients who experience severe side effects can intentionally stop medication to relieve the side effects. Substance use was also significantly associated with psychotropic medication non-adherence in our study. Patients who used psychoactive substances were more likely to be non-adherent compared to others. Psychoactive substance impair cognition and rationale thinking and are also associated with increased adverse effects and interactions with psychotropic medications (25) which can increase non-adherence to psychotropic medications.

Strengths of the study

• The study was done in public and private referral hospitals; the findings may apply to both public and private institutions.

• This is the first study exploring the association between personality traits and psychotropic medication non-adherence in Uganda and in the East African region.

Limitations of the study

• The study was cross-sectional which measures data at a single point in time

• The study was done in only two hospitals chosen by convenience, so the results of the study cannot be generalized for the country.

• The tools used were self-reported which can decrease the accuracy of the responses.

Areas of further research

Longitudinal studies on several sites of the country are needed to generalize the results for the country.