Professional Documents
Culture Documents
Self-Stigma and Medication Adherence Among Patients With Mental Illness Treated at Jimma University Medical Center, Southwest Ethiopia
Self-Stigma and Medication Adherence Among Patients With Mental Illness Treated at Jimma University Medical Center, Southwest Ethiopia
Abstract
Background: Self-stigma associated with mental illness has remained a global public health issue affecting social
interactions, health care, productivity and acceptance among others. It is one of important factors contributing to
non-adherence to medication that leads to increased hospitalization and higher healthcare costs. Hence, the study
aimed to assess self-stigma and medication adherence among patients with mental illness treated at the psychiatric
clinic of Jimma University Medical Center (JUMC).
Methods: A cross-sectional, community-level study was conducted at Jimma town. The patient’s data was collected
from records between April and June 2017 and the collected data was analyzed using SPSS version 21. The Internal-
ized Stigma of Mental Illness (ISMI) tool was utilized to measure internalized stigma. Linear regression analysis was
performed to get the final model. Statistical significance association was considered at p-values less than 0.05 and
95% confidence interval was used.
Results: Males comprised more than half (61%) of the total sample of 300 respondents and with a mean age of 34.99
(SD ± 11.51) years. About one-third (32%) of patients had a working diagnosis of schizophrenia followed by major
depressive disorder (24.3%). More than half of them, 182 (60.7%) were adherent to their psychotropic medication.
The overall mean value of self-stigma was 2.16 (SD = 0.867) and 84 (28%) of the respondents had moderate to high
self-stigma. Using ISMI the mean score of alienation was 2.26 (SD = 0.95), stereotype endorsement 2.14 (SD = 0.784),
perceived discrimination 2.18 (SD = 0.90), social withdrawal 2.10 (SD = 0.857) and stigma resistance 2.11 (SD = 0.844).
Increasing age of the patients (std. β = − 0.091, p = 0.009) and living with kids and spouse (std. β = − 0.099, p = 0.038)
were negatively associated with self-stigma whereas increased world health organization disability assessment
schedule (WHODAS) score (β = 0.501, p < 0.001), number of relapses (std. β = 0.183, p < 0.01) and medication non-
adherence (std. β = 0.084, p = 0.021) were positively associated with self-stigma.
Conclusion: The study revealed that there was high self-stigma among patients with mental illness and a significant
association between overall ISMI score and level of medication adherence. These require mental health professionals
*Correspondence: [email protected]
2
Clinical Pharmacy Department, School of Pharmacy, Institute of Health
Sciences, Wollega University, P.O Box 395, Nekemte, Ethiopia
Full list of author information is available at the end of the article
© The Author(s) 2020. This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing,
adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and
the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material
in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material
is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the
permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativeco
mmons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/
zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Abdisa et al. Int J Ment Health Syst (2020) 14:56 Page 2 of 13
and policy-makers should give attention to ways to overcome self-stigma and increase medication adherence among
patients with mental illness.
Keywords: Self-stigma, Mental illness, Medication adherence, Psychiatry, Ethiopia
tremendous burden on people with mental illness [24]. p = 0.2512%, the previous finding reported that the self-
Hence, the study was aimed to assess self-stigma and stigma prevalence was 25.12% [36].
medication adherence among patients with mental ill-
ness treated at the psychiatric clinic of Jimma University (Z α2 )2 p(1 − p)
n=
Medical Center (JUMC). d2
where P = extent of self-stigma, Z = critical value at 95%
Methods
CI of certainty (1.96). d = the margin of error, 5%. n = the
Study setting, design and eligibility criteria
required sample size.
The study was conducted at the community level in
Therefore, the sample size was
Jimma town patients with mental illness. Jimma town
is the capital city of Jimma zone which was founded in (1.96)2 (0.25)(1 − 0.75)
the late 1830s and it occurs at south west part of Ethio- n= , n = 289
(0.05)2
pia at a distance of 352 km southwest from Addis Ababa.
The town has 3 districts (Woreda) and 17 sub-districts Therefore, the sample size by the above calculation was
(kebele). The town has two hospitals; Jimma general 289. By adding 10% non-response rate, the final sample
hospital and Jimma University Medical Center (JUMC). size was 318.
Psychiatric outpatient and inpatient services are avail- For obtaining study population, a systematic sampling
able only at the JUMC. There were more than 1200 indi- method was used because the registration of the patients
viduals who have ever follow-up treatment at psychiatric with mental illness who were only from Jimma town was
clinics of JUMC with different diagnosis and duration used. Patients’ data registration revealed that the total
of treatment. A cross-sectional study was conducted by adult patients with mental illness who had treatment at
reviewing patients’ charts from April 20 to June 20, 2017. the psychiatry clinic of JUMC were 1200 and from these
Adult patients (≥ 18 years) with mental illness who were 698 adult patients were from Jimma town. From these,
listed under the DSM-IV-TR manual and had a history of 318 study participants were selected systematically by
at least one-time psychiatric treatment at the psychiatric using a computer-generated method (k = 2). By review-
clinic of JUMC were included. Patients who were newly ing chart study participants were identified. Finally, the
diagnosed with mental illness within 6 months dura- information of adult psychiatric patients was distributed
tion before data collection, who were not available dur- to health extension workers of the town and psychia-
ing data collection and those who had hearing problems try department staff according to the patients’ address.
were excluded from the study. Information of the study sample was provided to the
health extension workers who contacted the patients
Study variables using tracing methods and with the help of the contact
The outcome or dependent variable was self-stigma and number provided on the registration chart. But 14 of
the primary independent exposure variable was the level them died during data collection. Therefore, they were
of medication adherence. Other were sociodemographic excluded from the study and data collection and analysis
characteristics (age, sex, place of residence, education, was carried out among 304 participants.
occupation, monthly household expenditure, marital sta-
tus, living condition, religion), various psychiatric illness’ Data collection tools and process
(bipolar disorder, major depressive disorder, schizophre- The data was collected from patients’ charts and using
nia, brief psychotic disorder, schizophreniform disor- interview-based structured questionnaires among the
der, schizoaffective disorder, delusional disorder, panic participants who were addressed by tracing methods in
disorder, generalized anxiety disorder, post-traumatic each kebele of the town. Two Bachelor of Science (B.Sc)
stress disorder, acute stress disorder, social phobia etc.), psychiatric nurses and four mental health master’s stu-
duration of treatments, premorbid occupational status, dents were recruited for data collection.
the status of functional outcome, duration of the illness, The data collection tool had five components: socio-
age at onset of the illness, number of relapses, number demographic variables, clinical details, self-stigma of
of admissions, level of medication adherence and social mental illness (ISMI), WHODAS version 2 (12-items),
support. Oslo scale for social support and Morisky Medication
Adherence Scales (MMAS-4). Clinical inventory part
Sample size and sampling technique includes patients’ working diagnosis, age at onset, dura-
Single population proportion formula was used to tion of illness, number of episodes, number of hospitali-
calculate sample size using; the level of significance zations, duration of treatment, current clinical status,
taken as 95%, (Zα = 1.96), the error of margin, 5% and medication side effects, and previous suicide attempt.
Abdisa et al. Int J Ment Health Syst (2020) 14:56 Page 4 of 13
The Internalized Stigma of Mental Illness (ISMI-29 were scored for all questions of medication adherence
items) tool [13, 38–40] was utilized to measure internal- with a score of ≤ 2 classified as ‘non-adherent’ and > 2 as
ized stigma (internal consistency reliability coefficient of adherent [39, 41].
alpha = 0.90) [41]. The tool was validated and has been
used in several studies in Ethiopia [5, 36, 42]. Each ISMI Data quality management and analysis
item contains a declarative statement about a poten- To ensure the quality of data, the project staff were
tial stigma issue and participants respond to each state- trained for 2 days on the purpose of the study, details
ment by indicating their level of agreement: 1 = strongly of the questionnaires, the technique of interview, the
disagree; 2 = disagree; 3 = agree and 4 = strongly agree. importance of privacy and confidentiality of the respond-
It contains five subscales; alienation (6 items), stereo- ents. The data collection instrument tools were devel-
type endorsement (7 items), discrimination experience oped in English version, then translated to local language
(5 items), social withdrawal (6 items), and stigma resist- (Amharic and Afan Oromo) and later translated back
ance (5 items). Alienation is “the subjective experience of to English version by experts for consistency. Twenty
being less than a full member of society”. The stereotype patients were pretested at the psychiatric clinic, JUMC,
endorsement is “the degree to which patients agreed with those who came for appointments from different parts
common stereotypes about people with a mental illness”. of Jimma Zone except Jimma Town. During data collec-
The discrimination experience measures “respondents’ tion process, all the collected data was reviewed daily and
perceptions of the way they tend to be treated by oth- checked for completeness.
ers”. The social withdrawal measures the self-exclusion The collected data were entered into Epi-data version
from social events (situations) due to mental illness. The 3.1, and then exported to SPSS window version 21 for
stigma resistance subscale is “a person’s ability to resist summarization and further analysis. Before analysis lin-
stigma” [16, 36]. Except for the stigma resistance domain, ear regression assumptions were checked. Simple linear
a higher score of the remaining four subscales indicates regression analysis was carried out to assess the associa-
higher internalized stigma. For that matter, stigma resist- tion between different variables and to identify candi-
ance items were reverse coded. The overall score was dates for multivariate linear regression analysis. Variable
obtained by summing all the answered scores and divided having p-value of less than 0.25 was subjected to mul-
by a total number of items. tivariate linear regression analysis. Then, multivariate
The other tool which was developed by the World linear regression analysis was performed to get the final
Health Organization, The 12-item Self-Report World model. Statistical significance association was considered
Health Organization Disability Assessment Schedule 2.0 at p-values less than 0.05 and 95% confidence interval
(WHODAS-II) [43, 44] was used to establish the level of was used. Self-stigma was identified against medication
impairment associated with mental illness and it assesses adherence status. The mean score for each of the subscale
the level of disability and the number of days lost from and overall self-stigma was done but the final subscale of
work in the previous 30 days. It has 12 items and Likert ISMI was reversely recoded before the identified mean
scale (1–5). Oslo scale for social support has three items score. Finally, the results were presented in tables, graphs
and is used to assess the level of social support of the par- and statements.
ticipants [45, 46]. Oslo-3 Social Support Scale (OSS-3);
the interpretation was made as, a person scored 12–14 Operational definition
(good), 9–11 (moderate), and 3–8 (poor social support) Level of stigma was based on the mean score of ISMI, and
[47]. using similar score categories to the European and Ethio-
The Morisky Medication-taking Adherence Scale pian studies [39, 42]:
(MMAS-4), a four-item self-report measure of medi-
cation-taking behavior, which has high reliability and • < 2 of total score indicated that minimal internalized
validity, was used for assessing patient adherence [48]. stigma.
MMAS-4 was reliable in the study (Cronbach’s α = 0.73) • 2–2.5 of total score indicated that low internalized
done in Jimma university specialized [41]. It measures stigma.
both intentional and unintentional adherence based on • 2.5–3 of total score indicated that moderate internal-
forgetfulness, carelessness, stopping medication when ized stigma.
feeling better, and stopping medication when feel- • 3+ of total score indicated that high internalized
ing worse. Each of the items has dichotomous types of stigma.
responses (yes, no). In all, 0 point was allocated to a yes
response and 1 point to a no response. The total score Adherence to medication The level of drug adherence
ranged from 0 (non-adherent) to 4 (adherent). The results was measured based on MMAS-4 score. Accordingly,
Abdisa et al. Int J Ment Health Syst (2020) 14:56 Page 5 of 13
Sex
Results
Male 185 61.7
Socio‑demographic characteristics
Female 115 38.3
Of the total 318 participants contacted by tracers, 14
Marital status
people died during the study period. Only 304 partici-
Single 180 60.7
pants were included in the study with response rate of
Married 80 26.7
300 (98.67%). Of the total respondents, males comprised
Divorced 22 7.3
185 (61.7%) and the mean age of the respondents was
34.99 (SD = 11.51) years. One hundred and eighty-five Widowed 11 3.7
(61.0%) were single and the majority of them were Mus- Separated 5 1.7
lims (43.0%). One hundred forty-five (48.3%) of them Religion
were Oromo, and 192 (64.2%) of them were living with Muslim 129 43.0
their family. One hundred forty (46.7%) of the respond- Orthodox 121 40.3
ents were jobless and about one-third (34.3%) had pri- Protestant 44 14.7
mary educational level (Table 1). Catholic 4 1.3
Othersa 2 0.7
Ethnicity
Clinical characteristics
Oromo 145 48.3
About one-third (32%) of patients had a working diag-
Amhara 75 25.0
nosis of schizophrenia followed by major depressive
Dawuro 28 9.3
disorder (24.3%). The average duration of mental illness
Gurage 21 7.0
among the respondents was 98.25 (SD = 67.608) months
Kafa 19 6.3
ranged from 3 to 12 months. The mean value of dura-
tion before seeking treatment was 14.28 (SD = 24.31) Othersb 31 4.0
months. The average duration that respondents had Living condition
been getting treatment was 69.11 (SD = 64.98) months, With family (father or mother) 194 64.7
ranged zero to 348 months. After the first episode of With wife/husband and kids 62 20.7
their illness, respondents experienced an average value Living alone 17 5.7
of 2.17 (SD = 2.511) times relapse of their illness with a With relatives/friends 12 4.0
maximum of 17 times. The average number of admis- Living on street 5 1.7
sions to the hospital was 0.87 (SD = 1.289) with the Othersc 8 2.7
range of no admission to seven times. The mean age of Occupation
the participants at the onset of mental illness was 26.90 Jobless 139 46.3
(SD = 11.484) years. Employed 54 18.0
Around 74.7% of the respondents were on the work Student 29 9.7
(had jobs) and around two-thirds (65.7%) of them were Merchant 26 8.7
single when they had the onset of mental illness. Less House wife 23 7.7
than two-thirds of the respondents (63.3%) had a his- Daily laborer 16 5.3
tory of discontinuation of their medication without Retired 7 2.3
medical advice. Fifty-five (18.3%) respondents had poor Farmer 6 2.0
social supports whereas 39 (13.0%) of them had high Educational status
social supports. About 30.7% of respondents had a his- Illiterates 19 6.3
tory of attempting suicide to kill themselves with and Able to read/write, but no formal 6 2.0
education
without strong intentions. About 13% of the respond-
Primary 103 34.3
ents attempted suicide due to their illness alone whereas
Secondary 89 29.7
6.8% of them attempted due to their illness comorbid
College and above 83 27.7
with social problems like lack of support. More than two-
a
thirds (69.3%) of the respondents had current follow up Catholic, Wakeffata
b
from psychiatric clinics, but only 63.3% of the respond- Yem, Kefa, Sidamo, Wolayita
c
In prison, dormitory
ents were on treatment during the data collection period
(Table 2).
Abdisa et al. Int J Ment Health Syst (2020) 14:56 Page 6 of 13
Table 2 Clinical diagnosis of the sample (n = 300) Medication non-adherence, having diagnosed with
Diagnosis Frequency (n) Percent (%)
schizophrenia, living with others other than parents,
increased WHODAS score (decreased functional out-
Schizophrenia 96 32.0 come), stopped their treatment during data collection,
Major depressive disorder 73 24.3 having poor social support, increased number of admis-
Bipolar disorder 58 19.3 sion, having frequent relapse of the illness and longer
Brief psychotic disorder 20 6.7 duration of the illness before starting treatment were
Schizophreniform disorder 14 4.7 associated with significantly higher self-stigma whereas
Schizoaffective disorder 12 4.0 the remaining variables listed in Table 4 were associated
Post-traumatic disorder 7 2.3 with lower level of self-stigma upon binary linear regres-
Generalized anxiety disorder 6 2.0 sion (Table 4).
Delusional disorder 5 1.7 Multivariable regression revealed the age of the
Othersa 9 3.0 patient (std. β = − 0.091, p = 0.009) and living with kids
a
Dementia, social phobia, somatoform disorder, sleep disorder, substance and spouse (std. β = − 0.099, p = 0.038) were negatively
disorder associated with self-stigma whereas increased WHO-
DAS score (β = 0.501, p < 0.01), number of relapses (std.
β = 0.183, p < 0.01) and medication non-adherence (std.
Level of medication adherence and self‑stigma β = 0.084, p = 0.021) were positively associated with self-
Medication adherence was assessed using the Morisky stigma. Increasing the age of the patients and living with
medication adherence scale (MMAS) which contains four kids and spouses was significantly associated with lower
components. With this scale, 182 (60.7%) were adherent level of internalized stigma whereas the rest variables
to their medication and 39.3% were not-adherent. From were associated with higher level of internalized stigma
39.3% of the respondents who were non-adherents, 26% (Table 5).
of them were males. Those individuals forgetting taking
of their medication at right time and right dose were 127
(42.3%), careless in taking their medication 106 (35.3%), Association of self‑stigma and level of medication
discontinued their medication when felt better 88 (29.3%) adherence
and those who stopped their medication when their ill- By dichotomizing the level of medication adherence into
ness becomes worse were only 52 (17.3%). adherent and non-adherent, it was associated with self-
The self-stigma was evaluated using the ISMI scale stigma. In the bivariate (β = 0.034, p < 0.001) and multi-
that contains 29-items with four-point Likert scale and variate linear regression model (std. β = 0.084, p < 0.021),
has five subscales. The mean score of each sub-scale was being non-adherent to the medication was associated
done with different tools. Accordingly, mean alienation with higher self-stigma.
score 2.26 (SD = 0.95), mean stereotype endorsement Among respondents who were adherent to medi-
score 2.14 (SD = 0.784), mean perceived discrimination cations, minimal stigma was detected among 63.2%
score 2.18 (SD = 0.90), mean social withdrawal score whereas moderate to high stigma was detected in 17.5%.
2.10 (SD = 0.857) and mean stigma resistance score 2.11 Of the respondents who were non-adherent to their
(SD = 0.844). The stigma resistance component was medication, 44.1% had moderate to high internalized
reversely recoded because the maximum value indicates stigma. Among the respondents who were forgetting to
that resistance is low in opposite of other components take their medication, 26.0% of them had high levels of
of IMSI. The overall mean value of self-stigma was 2.16 stigma and nearly half (47.2%) of respondents who were
(SD = 0.867). careless in taking of their medication had moderate to
About 157 (52.3%), 59 (19.7%), 42 (14%), and 42 (14%), high self-stigma. About 37.5% of respondents with a his-
of the respondents had a minimal, low, moderate and tory of medication discontinuation when they felt well
high level of self-stigma based on total stigma score., from their illness, had moderate to high self-stigma.
respectively. Overall, around one-third (28%) of the About 51.9% of the respondents who had a history of
respondents had moderate to high self-stigma (Table 3). medication discontinuation when felt their illness worsen
without medical advice had moderate to high self-stigma
score (Table 6).
Factors associated with self‑stigma Overall, more than half (53.4%) of the respondents who
For one by one regression association of each dependent were non-adherent scored above the total mean score of
factor with self-stigma around 17 variables were associ- self-stigma and discrimination (Fig. 1).
ated and candidate for multivariate linear regression. When comparing the respondents with a history of
never forgetting to take their medication, respondents
Abdisa et al. Int J Ment Health Syst (2020) 14:56 Page 7 of 13
Alienation (yes)
Feeling of worthless 100 33.3
Miserable life due to having mental illness (MI) 127 42.3
Others people don’t understand me 101 33.7
Feeling of shame due to having MI 110 36.7
Feeling bad or annoyed due to having MI 133 44.3
Stereotype (yes)
Low self esteem 98 32.7
Public misconceptions work on me 102 34.0
People knows that I am mentally ill by just looking at me 90 30.0
PWMI are danger to other 45 15.9
Decisions about you shall be made by others 47 15.7
PWMI will not live better life 59 19.7
PWMI should not have marital life 49 16.3
Perceived discrimination (yes)
Because of my MI, I can’t do good to the public 49 16.3
Because of my MI other people discriminates me 92 30.7
Others believe that I will not be successful because of my MI 108 36.0
Other people will not give me value/concern because of my MI 95 31.6
Other people treat me as a child or inferior because of my MI 89 29.6
Other people prefer not to approach me or don’t want close relation with me 80 26.7
Social withdrawal (yes)
Not to bother others about me, am not talk to them about my self 130 43.4
I have no good social relationship due to the fear of behaving bad because of my MI 100 33.4
Misconception from the public made me isolate my self 88 29.4
I don’t feel good in a public place 89 29.7
Avoid healthy social contact with people to protect discrimination 80 16.7
For the sake of my family not to feel ashamed because of me, I isolate myself from social relation 80 26.6
Stigma resistance (no)
Feel comfortable when other see my symptom of MI 55 18.3
I am living a kind of life I want to live 207 46.6
I can live a better life despite my MI 174 58.0
PWMI can contribute well to the public 216 72.0
My MI makes me be stronger in my life 101 53.7
MI mental illness, PWMI patient with mental illness
Table 5 Variables associated with self-stigma in the final model of adjusted multivariate linear regression analysis
of the sample
Coefficientsa
treated with holy water and exorcism. Most Ethiopians adherence to spiritual explanations, particularly for seri-
have faith in traditional healers and procedures [50]. The ous psychotic symptoms, and their openness to both
possible factors for the high risk of stigma are ignorance, modern and traditional treatments speak to the impor-
perceived fear of injury, actual (perceived) absence of tance of developing a holistic and inclusive approach to
treatment, and cultural misconceptions about the nature psycho-education and treatment for this population [52].
of the mental illness, associating the illness with the A clearer understanding of the cross-cultural nature of
supernatural explanation [5–7, 51]. The rural population’s stigma and discrimination experienced by people living
Abdisa et al. Int J Ment Health Syst (2020) 14:56 Page 9 of 13
Status of adherence
Adherent 115 (63.2%) 35 (19.2%) 23 (12.6%) 9 (4.9%)
Non adherent 42 (35.6%) 24 (20.3%) 19 (16.1%) 33 (28.0%)
Total 157 (52.3%) 59 (19.7%) 42 (14.0%) 42 (14.0%)
MMAS-4
Forgetting their medication 43 (33.9%) 29 (22.8%) 22 (17.3%) 33 (26.0%)
Careless in taking their medication 34 (32.1%) 22 (20.8%) 18 (17.0%) 32 (30.2%)
Discontinue their medication when felt better 38 (43.2%) 17 (19.3%) 14 (15.9%) 19 (21.6%)
Discontinue their medication when felt their illness 16 (30.8%) 9 (17.3%) 9 (17.3%) 18 (34.6%)
becomes worsened
MMAS Morisky medication adherence scale
150
Nigeria [13, 53, 54]. The study done in Iran took place
127
among patients with bipolar disorders while the study in
100 Nigeria was only among patients with severe mental dis-
63
50
55 55
orders. Additionally, cross-sectional facility-based study
conducted in Addis Ababa, Ethiopia reported that nearly
0 three-quarters of respondents (71%) expressed strong
adherent non adherent
agreement to at least one internalized stigma item [42]. A
s gma less than 2.16 score s gma greater than 2.16
study by Gebeyehu et al. reported that (25.7%) of patients
Fig. 1 Stigma score around the overall mean score with the level of with severe mental disorders attending Bahir Dar Fel-
medication adherence among the sample ege Hiwote Referral hospital had perceived stigma [29].
Additionally, study by Adewuya et al. in Lagos, Nigeria
reported that high self-stigma was found in 21.6% of the
outpatients in the study [13]. The correlates of high self-
with mental disorders will also be an important avenue
stigma study in Nigeria included level of social support,
for future investigation.
duration of illness, level of insight and working status.
The majority of respondents had working a diagnosis of
This study found that social withdrawal (exclusion from
schizophrenia followed by major depressive disorder. The
social events) was less experienced than other compo-
mean value of time duration for which the respondents
nents of ISMI whereas alienation (being inferior) was the
had followed their treatment was 69.11 months and more
highest score in agreement with the study done in Jimma,
than half of the respondents had history of discontinua-
Dilla and Europe that identified high alienation score
tion of their medication without medical advice. About
[5, 36, 54]. But a study in Nigeria revealed that ISMI
30.7% of the respondents had feeling of socially discrimi-
scoring was highest on ‘dis-crimination experiences’
nated against due to their illness.
(mean = 2.69, SD = 1.88), followed by ‘social withdrawal’
In this finding, the overall self-stigma score was 2.16
(mean 2.40, SD = 1.51) [13].
and 28% of the respondents had moderate to high self-
The increasing age of the patients was significantly
stigma. The prevalence of the self-stigma was lower com-
associated with a lower level of internalized stigma. But
pared with the study done in Jimma [36] and Dilla [5],
a study by Kamaradova et al. reported the levels of self-
whereas it was comparable with the study done in Addis
stigma expressed by the total ISMI scores did not signifi-
Ababa [42]. The first study in Jimma and Dilla were done
cantly correlate with either age or age of disorder onset
among patients with mental illness only those on fol-
[1]. This leads to more adaptation and creates some rela-
low up and facility-based, while the current study was
tion with the community which may decrease the magni-
done both at community-based and tracing methods.
tude of self-stigma.
Additionally, the current study also took place among all
In our study finding, 55 (18.3%) had poor social sup-
patients who had a history of treatment at JUMC without
ports whereas 39 (13.0%) of them had high social sup-
limiting the duration and clinical diagnosis.
ports. A closely related finding was reported by Gebeyehu
The prevalence of self-stigma was higher in this study
et al. where 45.9% had good social support [29]. Living
compared with the studies done in Iran, Europe and
Abdisa et al. Int J Ment Health Syst (2020) 14:56 Page 10 of 13
with kids and spouses was negatively associated with [5, 64]. One study done in Czech Republic, by placing
self-stigma up on multivariable linear regression. This self-stigma as independent (predictor factors) for medi-
finding correlates with the previous study by Asrat et al. cation adherence, identified that there was negative asso-
that showed a lack of family support was independently ciation between self-stigma and medication adherence
associated with internalized stigma. Accordingly, patients among schizophrenic patients [1]. Contrary to this, one
getting good family support had less risk for internal- study done in Sweden identified that there was no signifi-
ized stigma than their counterparts [5]. These studies cant association between the two factors in the adjusted
strongly supported that patients getting better social model [38].
and family support had a better prognosis and favorable Even though self-stigma and discrimination could
outcome against internalized stigma. Furthermore, they occur among all patients, a high mean score of self-
are actively involved in anti-stigma campaigns and other stigma was common among non-adherent whereas low
activities. mean score was common among adherents than non-
The current study found that 39.7% of respondents adherents. About 39.3% scored greater than the overall
were non-adherent with their medication. This find- mean value of ISMI score. Visiting psychiatric clinics
ing was in line with global non-adherence rates among and taking medications are directly associated with the
patients with mental illness between 30 and 65% [55, disclosure of their health status. In general, the present
56]. The finding was also comparable with those from study showed that higher self-stigma level was linked to
Jimma, Ethiopia 41.2% [41], and India 42% [57]. Contrary non-adherence. This underlines the importance of inter-
to this, the prevalence was higher than in India 26% [58] ventions aimed at lowering the levels of self-stigma in
and Addis Ababa, Ethiopia (26.5%) [42]. The finding was patients.
also lower than a study conducted in Nigeria (48.0%) [59], Although the treatment gap is a useful construct to
Curitiba, Parana (49%) [60], Nigeria 54.2% [61], Bahir measure access and equitability of care, it fails to com-
Dar, Ethiopia (55.2%) [29], South Africa 63.2% [62] and municate the real-life consequences of the treatment gap
Pakistan’s study (64.75%) [63]. This variation could be and the urgent need to address care disparities in Ethio-
due to the differences in socio-demography of the study pia [65]. Lack of capacity is deeply rooted in the lack of
population, selection criteria for the study participants, psychiatric training available in Ethiopia. Resources for
study design, sample size and tools used for the assess- learning about mental illness fuel the fire of stigma and
ment of adherence. It might also relate to the differences ignorance as well. Medical journals exist in Ethiopia, but
in the health care setting, community culture or level of rarely ever broach the topic of mental illness [66]. Stigma
knowledge on drug adherence of health providers. has long been viewed as a major barrier to mental health
For example, the lower sample size was taken in Paki- reform and community integration for people with men-
stan, using a convenient, non-probability sample of 135 tal disorders. Whereas, knowledge of the social dimen-
follow-up patients were assessed. The study done in sions of stigma and discrimination on mental patients
South Africa took only among schizophrenic patients has begun to accumulate, much remains to be known
and the study done in Nigeria was carried out at facility- about the extent and models of self-stigma and how to
based and among patients struggling with illness a sign measure it.
and symptoms by using self-administered questionnaires.
Adherence at any given moment does not guarantee sub- Limitations of the study
sequent adherence since the definition attributed by the One of the study’s limitations is the fact that most data
people to the use of the drugs and the motivation to do were obtained using self-reported questionnaires. The
so are flexible, they are influenced by their perception, self-report method used to measure self-stigma and
attitude, external factors, and past experiences. Thus, medication adherence might substantially overestimate
even adherent patients need great focus and monitoring adherence, as it relies on patient response. To get the
regarding the use of medication. According to the avail- study participants, we used a general survey method on
able evidence, non-adherence to psychiatric medica- patients with mental illness who had a history of treat-
tions is high and is one of the top public health problems ment at the psychiatric clinic of JUMC which might lead
requiring due attention and intervention [41]. to recall bias and great variability in the information of
In this study, significant association was identified the current data (can significantly affect the reliability
between self-stigma and levels of medication adherence. and validity of the result). The cross-sectional design of
Study participants who didn’t adhere to their medication the study fails to assess patients’ adherence behaviors
as ordered by health care providers were experiencing over time, and although this approach is helpful to inves-
more self-stigma than those patients who were adher- tigate associations between variables, it cannot attribute
ent. It was in agreement with other research outcomes cause and effect. Another limitation of this study pertains
Abdisa et al. Int J Ment Health Syst (2020) 14:56 Page 11 of 13
along with other similar data from Ethiopia and other Author details
low-income countries, supports the need to incorporate 1
Department of Psychiatry, School of Nursing and Midwifery, Institute
culturally appropriate methods of addressing internalized of Health Sciences, Wollega University, Nekemte, Ethiopia. 2 Clinical Pharmacy
Department, School of Pharmacy, Institute of Health Sciences, Wollega
stigma into rehabilitation packages for this group. University, P.O Box 395, Nekemte, Ethiopia. 3 Department of Psychiatry, College
Strategies focusing on early detection and treatment of of Health Science, Jimma University, Jimma, Ethiopia. 4 Department of Pediat-
mental illness will play an important role in stigma pre- rics and Child Health, Wollega University Referral Hospital, Nekemte, Ethiopia.
5
Department of Obstetrics and Gynecology, School of Medicine, Institute
vention. Awareness creation, community mobilization, of Health Sciences, Wollega University, Nekemte, Ethiopia. 6 Department
and service expansion strategies are cost-effective and of Surgery, School of Medicine, Institute of Health Science, Wollega University,
efficient ways to act against stigma. Further, strengthen- Nekemte, Ethiopia. 7 Medical Microbiology Unit, Department of Medical
Laboratory Science, Institute of Health Sciences, Wollega University, Nekemte,
ing the social network (support) of mentally ill people Ethiopia. 8 Department of Nursing, Institute of Health Sciences, Wollega
with themselves and with the community is found to be University, Nekemte, Ethiopia.
necessary. Working with the media to raise awareness
Received: 20 April 2020 Accepted: 23 July 2020
of mental health issues and establish best practices for
reporting and for depicting mental illness is an effective
approach with the potential to have a positive impact on
public perceptions.
Abdisa et al. Int J Ment Health Syst (2020) 14:56 Page 12 of 13
References 24. Boyd JE, Katz EP, Link BG, Phelan JC. The relationship of multiple aspects
1. Kamaradova D, Latalova K, Prasko J, Kubinek R, Vrbova K, Mainerova of stigma and personal contact with someone hospitalized for mental
B, Cinculova A, Ociskova M, Holubova M, Smoldasova J. Connection illness, in a nationally representative sample. Soc Psychiatry Psychiatr
between self-stigma, adherence to treatment, and discontinuation of Epidemiol. 2010;45(11):1063–70.
medication. Patient Preference Adherence. 2016;10:1289. 25. Livingston JD, Boyd JE. Correlates and consequences of internalized
2. Link BG, Phelan JC. Conceptualizing stigma. Ann Rev Sociol. stigma for people living with mental illness: a systematic review and
2001;27(1):363–85. meta-analysis. Soc Sci Med. 2010;71(12):2150–61.
3. Rector NA, Stolar N, Grant P. Schizophrenia: cognitive theory, research, 26. Sayers J. The world health report 2001-Mental health: new understand-
and therapy. New York: Guilford Press; 2011. ing, new hope. Bull World Health Organ. 2001;79:1085.
4. Corrigan PW, Rowan D, Green A, Lundin R, River P, Uphoff-Wasowski K, 27. Kessler RC, Berglund PA, Bruce ML, Koch JR, Laska EM, Leaf PJ, Mander-
White K, Kubiak MA. Challenging two mental illness stigmas: personal scheid RW, Rosenheck RA, Walters EE, Wang PS. The prevalence and
responsibility and dangerousness. Schizophr Bull. 2002;28(2):293–309. correlates of untreated serious mental illness. Health Serv Res. 2001;36(6
5. Asrat B, Ayenalem AE, Yimer T. Internalized stigma among patients Pt 1):987.
with mental illness attending psychiatric follow-up at Dilla University 28. Corrigan P. How stigma interferes with mental health care. Am Psychol.
Referral Hospital, Southern Ethiopia. Psychiatry J. 2018. https://doi. 2004;59(7):614.
org/10.1155/2018/1987581. 29. Gebeyehu DA, Mulat H, Bekana L, Asemamaw NT, Birarra MK, Takele WW,
6. World Health Organization. Reducing stigma and discrimination against Angaw DA. Psychotropic medication non-adherence among patients
older people with mental disorders: a technical consensus statement. with severe mental disorder attending at Bahir Dar Felege Hiwote Refer-
Geneva: World Health Organization; 2002. ral hospital, north west Ethiopia, 2017. BMC Res Notes. 2019;12(1):102.
7. Graham N, Lindesay J, Katona C, Bertolote JM, Camus V, Copeland JR, 30. Teferra S, Hanlon C, Beyero T, Jacobsson L, Shibre T. Perspectives on
de Mendonça Lima CA, Gaillard M, Gély Nargeot MC, Gray J. Reducing reasons for non-adherence to medication in persons with schizophrenia
stigma and discrimination against older people with mental disorders: a in Ethiopia: a qualitative study of patients, caregivers and health workers.
technical consensus statement. Int J Geriatr Psychiatry. 2003;18(8):670–8. BMC Psychiatry. 2013;13(1):168.
8. Girma E, Tesfaye M. Patterns of treatment seeking behavior for mental 31. Lacro JP, Dunn LB, Dolder CR, Leckband SG, Jeste DV. Prevalence of and
illnesses in Southwest Ethiopia: a hospital based study. BMC Psychiatry. risk factors for medication nonadherence in patients with schizo-
2011;11(1):138. phrenia: a comprehensive review of recent literature. J Clin Psychiatry.
9. Yen C-F, Chen C-C, Lee Y, Tang T-C, Yen J-Y, Ko C-H. Self-stigma and its 2002;63:892–909.
correlates among outpatients with depressive disorders. Psychiatr Serv. 32. Feldhaus T, Falke S, von Gruchalla L, Maisch B, Uhlmann C, Bock E, Lencer
2005;56(5):599–601. R. The impact of self-stigmatization on medication attitude in schizophre-
10. Werner P, Aviv A, Barak Y. Self-stigma, self-esteem and age in persons with nia patients. Psychiatry Res. 2018;261:391–9.
schizophrenia. Int Psychogeriatr. 2008;20(1):174–87. 33. Demoz Z, Legesse B, Teklay G, Demeke B, Eyob T, Shewamene Z, Abera
11. Thornicroft G, Brohan E, Kassam A, Lewis-Holmes E. Reducing stigma M. Medication adherence and its determinants among psychiatric
and discrimination: candidate interventions. Int J Ment Health Syst. patients in an Ethiopian referral hospital. Patient Preference Adherence.
2008;2(1):3. 2014;8:1329.
12. Loganathan S, Murthy SR. Experiences of stigma and discrimination 34. Jenkins JH, Carpenter-Song EA. Awareness of stigma among persons with
endured by people suffering from schizophrenia. Indian J Psychiatry. schizophrenia: marking the contexts of lived experience. J Nerv Ment Dis.
2008;50(1):39. 2009;197(7):520–9.
13. Adewuya AO, Owoeye AO, Erinfolami A, Ola BA. Correlates of self-stigma 35. Freudenreich O, Cather C, Evins AE, Henderson DC, Goff DC. Attitudes of
among outpatients with mental illness in Lagos, Nigeria. Int J Soc Psy- schizophrenia outpatients toward psychiatric medications: relationship to
chiatry. 2011;57(4):418–27. clinical variables and insight. J Clin Psychiatry. 2004;65:1372–6.
14. Gaebel W, Baumann AE. Interventions to reduce the stigma associated 36. Girma E, Tesfaye M, Froeschl G, Möller-Leimkühler AM, Dehning S, Müller
with severe mental illness: experiences from the open the doors program N. Facility based cross-sectional study of self stigma among people with
in Germany. Can J Psychiatry. 2003;48(10):657–62. mental illness: towards patient empowerment approach. Int J Ment
15. Stuart H. Fighting the stigma caused by mental disorders: past per- Health Syst. 2013;7(1):21.
spectives, present activities, and future directions. World Psychiatry. 37. Khairy N, Hamdi E, Sidrak A, Sabry N, Nasreldin M, Khoweiled A, Loza N.
2008;7(3):185–8. Impact of the first national campaign against the stigma of mental illness.
16. Ghanean H, Nojomi M, Jacobsson L. Internalized stigma of mental illness Egypt J Psychiatry. 2012;33(1):35.
in Tehran, Iran. Stigma Res Action. 2011;1(1):11–7. 38. Brain C, Sameby B, Allerby K, Quinlan P, Joas E, Lindström E, Burns T, Waern
17. Hansson L, Stjernswärd S, Svensson B. Perceived and anticipated M. Stigma, discrimination and medication adherence in schizophrenia:
discrimination in people with mental illness—an interview study. Nord J results from the Swedish COAST study. Psychiatry Res. 2014;220(3):811–7.
Psychiatry. 2014;68(2):100–6. 39. Ritsher J, Otilingam PG, Grajales M. Internalized stigma of mental
18. Corrigan P, Thompson V, Lambert D, Sangster Y, Noel JG, Campbell J. illness: psychometric properties of a new measure. Psychiatry Res.
Perceptions of discrimination among persons with serious mental illness. 2003;121(1):257–65.
Psychiatr Serv. 2003;54(8):1105–10. 40. Boyd JE, Adler EP, Otilingam PG, Peters T. Internalized stigma of
19. Dickerson FB, Sommerville J, Origoni AE, Ringel NB, Parente F. Experi- mental illness (ISMI) scale: a multinational review. Compr Psychiatry.
ences of stigma among outpatients with schizophrenia. Schizophr Bull. 2014;55(1):221–31.
2002;28(1):143–55. 41. Tesfay K, Girma E, Negash A, Tesfaye M, Dehning S. Medication non-
20. Mulvale G, Chodos H, Bartram M, MacKinnon MP, Abud M. Engaging civil adherence among adult psychiatric out-patients in Jimma Univer-
society through deliberative dialogue to create the first Mental Health sity specialized hospital, Southwest Ethiopia. Ethiop J Health Sci.
Strategy for Canada: changing directions, changing lives. Soc Sci Med. 2013;23(3):227–36.
2014;123:262–8. 42. Assefa D, Shibre T, Asher L, Fekadu A. Internalized stigma among patients
21. Goldbloom D, Bradley L. The Mental Health Commission of Canada: the with schizophrenia in Ethiopia: a cross-sectional facility-based study. BMC
first five years. Ment Health Rev J. 2012;17(4):221–8. Psychiatry. 2012;12(1):239.
22. Sewilam AM, Watson AM, Kassem AM, Clifton S, McDonald MC, Lipski R, 43. Axelsson E, Lindsäter E, Ljótsson B, Andersson E, Hedman-Lagerlöf E. The
Deshpande S, Mansour H, Nimgaonkar VL. Suggested avenues to reduce 12-item Self-Report World Health Organization Disability Assessment
the stigma of mental illness in the Middle East. Int J Soc Psychiatry. Schedule (WHODAS) 2.0 administered via the internet to individuals with
2015;61(2):111–20. anxiety and stress disorders: a psychometric investigation based on data
23. Rüsch N, Angermeyer MC, Corrigan PW. Mental illness stigma: con- from two clinical trials. JMIR Ment Health. 2017;4(4):e58.
cepts, consequences, and initiatives to reduce stigma. Eur Psychiatry. 44. Basavarajappa C, Kumar KS, Suresh VC, Kumar CN, Ravishankar V, Vana-
2005;20(8):529–39. moorthy U, Mehta U, Waghmare A, Thanapal S. What score in WHODAS
2.0 12-item interviewer version corresponds to 40% psychiatric disability?
Abdisa et al. Int J Ment Health Syst (2020) 14:56 Page 13 of 13
A comparative study against IDEAS. J Psychosoc Rehabil Ment Health. 57. Nirojini PS, Bollu M, Nadendla R. Prevalence of medication non-adher-
2016;3(1):21–6. ence among the psychiatric patients-results from a survey conducted in
45. Abiola T, Udofia O, Zakari M. Psychometric properties of the 3-item oslo a tertiary care hospital. Int J Pharm Pharm Sci. 2014;6(4):461–3.
social support scale among clinical students of Bayero University Kano, 58. Shoib S, Mushtaq R, Dar MM, Mir JA, Shah T, Singh R, Ahmad J, Kyser S.
Nigeria. Malays J Psychiatry. 2013;22(2):32–41. Psychosocial and medical factors affecting treatment compliance in
46. Kocalevent R-D, Berg L, Beutel ME, Hinz A, Zenger M, Härter M, Nater U, patients attending psychiatric hospital: a study from Kashmir. 2014.
Brähler E. Social support in the general population: standardization of the 59. Adewuya AO, Owoeye OA, Erinfolami AR, Coker AO, Ogun OC, Okewole
Oslo social support scale (OSSS-3). BMC Psychol. 2018;6(1):31. AO, Dada MU, Eze CN, Bello-Mojeed MA, Akindipe TO. Prevalence and cor-
47. Dalgard O. Social support-Consequences for individual and society. relates of poor medication adherence amongst psychiatric outpatients in
EUPHIX, EUphact Bilthoven: RIVM. http://www.euphix.org. EUphact southwestern Nigeria. Gen Hosp Psychiatry. 2009;31(2):167–74.
Determinants of health Environment Social support 16 March 2009. 60. Borba LdO, Maftum MA, Vayego SA, Mantovani MdF, Felix J, Kalinke
http://www.euphix.org/object_document/o5522n27411.html. Accessed LP. Adherence of mental therapy for mental disorder patients to drug
12 Feb 2017. health treatment [corrected]. Revista da Escola de Enfermagem da USP.
48. Morisky DE, Green LW, Levine DM. Concurrent and predictive valid- 2018;52:e03341.
ity of a self-reported measure of medication adherence. Med Care. 61. Ibrahim A, Pindar S, Yerima M, Rabbebe I, Shehu S, Garkuwa H, Bashir I,
1986;24:67–74. Wakil M, Yahya S. Medication-related factors of non adherence among
49. Workneh T, Emirie G, Kaba M, Mekonnen Y, Kloos H. Perceptions of health patients with schizophrenia and bipolar disorder: outcome of a cross-
and illness among the Konso people of southwestern Ethiopia: persis- sectional survey in Maiduguri, North-eastern Nigeria. J Neurosci Behav
tence and change. J Ethnobiol Ethnomed. 2018;14(1):18. Health. 2015;7(5):31–9.
50. Hodes R. Cross-cultural medicine and diverse health beliefs, Ethiopians 62. Mahaye S, Mayime T, Nkosi S, Mahomed FN, Ntuli L, Pramlal J, Setlhabana
abroad. West J Med. 1997;166(1):29. O, Oosthuizen F. Medication adherence of psychiatric patients in an
51. Girma E, Möller-Leimkühler AM, Dehning S, Mueller N, Tesfaye M, Froeschl outpatient setting. Afr J Pharm Pharmacol. 2012;6(9):608–12.
G. Self-stigma among caregivers of people with mental illness: toward 63. Taj F, Tanwir M, Aly Z, Khowajah AA, Tariq A, Syed FK, Waqar F, Shahzada
caregivers’ empowerment. J Multidiscip Healthc. 2014;7:37. K. Factors associated with non-adherence among psychiatric patients
52. Monteiro NM, Balogun SK. Urban and rural perceptions of mental illness at a tertiary care hospital, Karachi, Pakistan: a questionnaire based cross-
in Ethiopia. IOSR J Humanit Soc Sci. 2013;8:43–51. sectional study. J Pak Med Assoc. 2008;58(8):432.
53. Sadighi G, Khodaei MR, Fadaie F, Mirabzadeh A, Sadighi A. Self 64. Ayenalem E, Tiruye TY, Muhammed MS. Impact of self stigma on quality
stigma among people with bipolar-I disorder in Iran. Iran Rehabil J. of life of people with mental illness at Dilla University referral hospital,
2015;13(1):28–32. South Ethiopia. Am J Health Res. 2017;5(5):125–30.
54. Brohan E, Elgie R, Sartorius N, Thornicroft G, Group G-ES. Self-stigma, 65. Fekadu A, Thornicroft G. Global mental health: perspectives from Ethio-
empowerment and perceived discrimination among people with schizo- pia. Glob health Action. 2014;7:25447.
phrenia in 14 European countries: The GAMIAN-Europe study. Schizophr 66. Tobin Bridget. Mental Health Care in Ethiopia, 2014. https://www.borge
Res. 2010;122(1–3):232–8. nmagazine.com/mental-health-care-ethiopia/. Accessed 14 July 2017.
55. Yang J, Ko Y-H, Paik J-W, Lee M-S, Han C, Joe S-H, Jung I-K, Jung H-G,
Kim S-H. Symptom severity and attitudes toward medication: impacts
on adherence in outpatients with schizophrenia. Schizophr Res. Publisher’s Note
2012;134(2–3):226–31. Springer Nature remains neutral with regard to jurisdictional claims in pub-
56. Kassis IT, Ghuloum S, Mousa H, Bener A. Treatment non-compliance of lished maps and institutional affiliations.
psychiatric patients and associated factors: are patients satisfied from
their psychiatrist? J Adv Med Med Res. 2014;4:785–96.