J. Biomedical Science and Engineering, 2010, 3, 602-611
doi:10.4236/jbise.2010.36082 Published Online June 2010 (http://www.SciRP.org/journal/jbise/
JBiSE
).
Published Online June 2010 in SciRes. http://www.scirp.org/journal/jbise
Medication compliance among mentally Ill patients in public
clinics in Kingston and St. Andrew, Jamaica
Andrea E. Pusey-Murray1, Paul A. Bourne1, Stan Warren2, Janet LaGrenade1, Christopher A. D. Charles3
1Department of Community Health and Psychiatry, Faculty of Medical Sciences University of the West Indies, Kingston, Jamaica;
2Department of Sociology, Psychology and Social Work, The University of the West Indies, Mona, Jamaica;
3King Graduate School, Monroe College, Bronx, New York and Center for Victim Support, Harlem Hospital Center, New York,
USA.
Received 26 February 2010; received 5 April 2010; accepted 15 April 2010.
ABSTRACT
The Bellevue and the Hagley Park mental health
outpatient clinics in Jamaica serve the majority of
psychiatric patients in the country, but there is a
dearth of research on medication compliance, which
is a very important mental health issue. Medication
compliance affects intervention outcomes. Therefore,
this study seeks to examine medication compliance
among psychiatric patients in Jamaica. A 33-item
questionnaire which included items on demographics,
health conditions, medication compliance and in-
sightfulness was administered to a sample of 370 par-
ticipants with a response rate of 93%. The majority
of the participants have schizophrenia, followed by
depression, bipolar disorder and drug-induced psy-
chosis. The majority of the participants (65.3%) did
not comply with their prescribed medication regimen.
Medication compliance was significantly related to:
gender (P < 0.05) where males were more likely to
take the prescribed medication, family support (P <
0.05) where the participants who received family
support (the majority being males) were more likely
to take the prescribed medication, and insightfulness
(P < 0.05) where the majority of participants with
insightfulness were females. Locus on control was not
statistically tested but a majority of the non- compli-
ant participants reported that factors external to
themselves had greater control over their disorder.
Conclusion: There are three significant factors that
explain the medication compliance of psychiatric pa-
tients in Jamaica. An important non-tested factor is
locus of control so there needs to be more research to
understand the range of factors that can inform and
improve patient education about medication compli-
ance.
Keywords: Medication Compliance; Mentally Ill; Public
Clinics; Kingston; St. Andrew
1. INTRODUCTION
Mental illness such as schizophrenia, bipolar disorders
and uni-polar depression presents a serious health care
problem in Caribbean countries such as Jamaica and
worldwide. The economic, clinical, and personal bur-
dens associated with schizophrenia make it a leading
public health problem. The earliest epidemiological re-
port on schizophrenia in the Caribbean was from Ja-
maica [1], and indicated that the annual population inci-
dence rate was 150 per 100,000. Burke [2] confirmed
this finding, and reported an annual incidence rate of 136
per 100,000. Both studies were based on mental hospital
admission rates, and it is likely that these studies failed
to access the total number of admissions for the island.
A study of psychiatric admissions to private and public
hospitals across Jamaica [3] reported that the incidence
rate of schizophrenia had fallen from 69 per 100,000 in
1960 to 35 per 100,000 in 1990.
Antipsychotic drugs, the most effective treatment for
acute episodes or exacerbations of schizophrenic illness
[4], allow many patients to leave institutions and live in
the community [5]. Rates of relapse among patients with
schizophrenia who receive medication are two to three
times lower than those among patients receiving placebo
[6], and non-compliance increases the frequency of acute
psychotic episodes and psychiatric hospitalization [7].
Although antipsychotic drugs can have serious adverse
effects [8], many clinicians prescribe them at moderate
doses for as long as possible to prevent relapse. In addi-
tion to antipsychotic agents, patients with schizophrenia
may receive lithium, antidepressants, or benzodiazepines
for concomitant psychiatric disorders [4].
Mental health research and pharmaceutical innovation
have developed a class of drugs referred to as “atypical

A. E. Pusey-Murray et al. / J. Biomedical Science and Engineering 3 (2010) 602-611 603
antipsychotics”, which are used for the treatment of se-
vere schizophrenics who are considered treatment- re-
sistant to traditional or conventional antipsychotic med-
ications, or who experience side-effects severe enough to
require that the patient discontinue use of conventional
antipsychotics [9]. These atypical antipsychotics are
tremendously effective in combating the symptoms of
schizophrenia while avoiding the severe side-effects
often experienced through treatment with conventional
antipsychotics [10]. In Jamaica medications for treating
mental health problems are limited and their availability
varies district by district. Older antipsychotic drugs are
usually available, but there is only limited availability of
the newer, more expensive anti-psychotics. In the North
East, some patients are treated with clozapine by special
arrangement by the district psychiatrist [11].
Adherence to drug regimen is a very important factor
for improvement. Adherence may be defined as the ex-
tent to which a person’s behaviour confirms to medical
or health advice [12]. Patient who do not follow the
treatment schedule and drug regimens prescribed to
them by physician can be described as noncompliant or
not adherent [13]. Medication compliance among bipolar
disorder patients is related to the constructs of the health
belief model (HBM) such as benefits and barriers, sus-
ceptibility and perceived seriousness [14]. However, the
compliance/non-compliance of patients is only moder-
ately predicted by the HBM. The HBM does not cover
some determinants of medication compliance such as
social influence and treatment alliance [15]. Patients
with bipolar disorder who are medication compliant per-
ceive the quality of their life to be higher, have greater
resources to cope with stress and have a stronger belief
that their behaviour controls their health status, unlike
non-compliant patients [16]. The patients’ perception of
the medication is important because cognitive disso-
nance suggests that the perceptual properties of the
medication have particular meanings for the patients.
These meanings can support or distract the patients from
medication compliance [17]. Patients with a major de-
pressive disorder who have a superior medication com-
pliance index are more likely to show improved scores
on the Hamilton Depression Scale [18]. The medication
events monitoring system (MEMS) used in a study with
schizophrenic disorder patients reveals a 63% mean
compliance rate for the first month and a decline from
56% to 45% over the next five months. The medication
compliance of these patients can be monitored with
electronic monitoring devices, but data recovery and
compliance require methodological improvements [19].
Among patients with psychotic disorders there was a
significant relationship between medication adherence
and involuntary admission, substance abuse, not gradu-
ating from school, and a history of abusive behaviour.
Patients with paranoid or negative symptoms were less
compliant in taking their medication. However, patients
who were changed from a typical to an atypical antipsy-
chotic medication were more compliant than patients
who remained on the typical antipsychotic medication.
The patients who had higher medication compliance
experienced much greater improvement of their psychi-
atric symptoms [20]. Medication self-management am-
ong patients with schizophrenia or affective disorder is
important for medication compliance. However, patients
who are self-managing their medication, guided by the
principles of motivational interviewing, have better atti-
tudes towards medications and insights about their ill-
ness compared to controls, but the group difference is
not significant [21].
Patients with schizophrenia, bipolar disorder and uni-
polar depression, who were educated about the nature of
their disorder and its pharmacological management,
were more compliant in outpatient follow-ups, and dis-
played less fear of being addicted to the medication and
dealing with the side effects of the drug [22]. Severely
mentally ill patients who were a part of the Medication
Usage Skills for Effectiveness (MUSE) Program that
taught cueing, life skill techniques, visual feedback and
about the data displayed on the medication cap showed
an overall mean compliance rate of 76%, compared to a
57% compliance rate for the control group [23].
The problem of poor adherence to therapeutic regimen
has been a matter of concern to the professionals for
years. The paucity of data on cost-related medication
adherence problems has important implications not only
for estimating their clinical significance, but also for
understanding the extent to which adherence problems
vary across socioeconomic groups. To our knowledge
there has not been any study that has examined drug
compliance among mentally ill patients in Jamaica who
are treated with oral antipsychotics. The aim of this
study was to examine the medication compliance of
psychiatric patients in Jamaica.
2. MATERIALS AND METHODS
2.1. Sample
In Jamaica the single mental hospital is Bellevue Mental
Hospital in Kingston. Its three acute care wards of 50
beds each serve a designated geographic catchment area
of the parishes of Kingston, St. Andrew, and St. Cath-
erine, a population that is 70 percent urban [24]. Of the
hospital’s patients, 40% are older than 65 years, 60% of
them are regarded as chronically ill, and 300 of them
have lived in the hospital for most of their adult lives.
The University Hospital West Indies (UHWI), also lo-
cated in Kingston, has a 20-bed psychiatric unit. It is an
acute unit with an average length of stay of 15 days. Ja-

604 A. E. Pusey-Murray et al. / J. Biomedical Science and Engineering 3 (2010) 602-611
maica’s rural and western regions are served partly by
Cornwall Regional Hospital. This 60-bed unit offers a
full range of services, but only 30 beds are used because
of staff shortages [11].
The Bellevue Hospital in Kingston and the Hagley
Park Health Centre in Saint Andrew investigated in this
study are two of the main psychiatric clinics in Jamaica.
They attend to in excess of 90% of the psychiatric
out-patient cases in the country. The current study used
cross-sectional survey to gather the data from partici-
pants. Using a sampling error of ± 3.0% and a 95% con-
fidence interval, the calculated sample size was 370 par-
ticipants. Based on the proportionality of cases seen over
the last 2 months, this was used to proportion the sample
size from each institution: 200 participants from the
Bellevue Hospital Clinic and 170 from the Hagley Park
Health Centre. The response rate was 93.0% with 186
participants from the Bellevue Outpatient Clinic and 158
from the Hagley Park Health Centre.
Data was collected over a three-month period, from
March to May, 2008. A team of trained data collectors
were retrained in keeping with the peculiarities of the
task. The researcher was a part of the process, and regu-
lar checks were done to ensure consistency among inter-
viewers. The inclusion/exclusion criterion was based on
mental illness and medication use. Mentally ill patients
who were visiting the clinics for the first time were ex-
cluded from the study because they would not have de-
veloped a pattern of medication compliance (or non-
compliance).
2.2. The Instrument
A 33-item questionnaire was used to collect the data.
The questions included: 1) demographic information, 2)
medication compliance, 3) reasons for taking (or not
taking) medication, 4) accessibility and affordability of
medication, 5) use and prevalence of home remedies, 6)
family support and 7) perception of the management of
mental illness (Appendix 1).
2.2.1. Test-and-Pretesting the Instrument
The instrument was developed in consultation with aca-
demics at the University of the West Indies, Mona, Ja-
maica, caregivers/guardians at the Harbour View Health
Centre in St. Andrew, health practitioners, and non-
prospective mentally ill patients for the sample. The in-
strument was modified in keeping with the comments of
the various stakeholders, with specific emphasis on
mentally ill patients’ perspectives, suggestions and rec-
ommendations. A pilot of 50 questionnaires was tested
at the Harbour View Health Centre on ill patients, and
they provided useful information regarding ambiguities,
which was then used to modify the final instrument.
2.3. Measure
Non-compliance in medication denotes the failure or
refusal of an individual to take the prescribed medica-
tion(s) as recommended by the medical practitioner.
Compliance is adhering to the prescription of oral or
other forms of medication as stipulated by the medical
practitioner. It was measured using the item “I take my
medication as instructed by the doctor.” The options
ranged from always, most times, occasionally and rarely,
to never. Compliance for this study is taken to be only a
response of ‘always’.
Health condition is an illness or ailment diagnosed by
the medical practitioner, and this was taken from the
medical record of the participant. Insightfulness was
measured by the questions: 1) “Do you have a sense of
the likelihood of being ill or of when your illness is
likely to occur?” 2) Should medication only be taken
when one is experiencing symptoms of illness?” and 3)
“Will counselling on how to take medications help you
to take them better?” The options were yes, no, or unsure.
If the answers to all three questions were yes, it was
coded as insight, if no was selected in any question it
was coded as lack of insight, and unsure was coded as
undecided. Non-responses were treated as missing, and
not included in any categorization.
2.4. Ethics and Informed Consent
The survey was submitted and approved by the Univer-
sity of the West Indies Medical Faculty’s Ethics Com-
mittee. Participants and/or caregivers gave voluntary
consent to their participation in the study. In order to
ensure confidentiality, the personal information (i.e.
name, address) of the participants was taken from the
questionnaires and discarded, after which the other in-
formation was entered and stored for data-analysis.
2.5. Statistical
Percentages were used to provide background informa-
tion on the demographic characteristics of the sample,
knowledge of the medication and self-reported informa-
tion. Chi-square tests were utilized to examine whether
statistical associations existed between non-metric de-
pendent and independent variables. A P-value of < 5%
(i.e. 95% confidence interval) was used to determine the
statistical associations between the variables.
3. RESULTS
3.1. Demographic Characteristic of Sample
There were 344 participants in the sample, of which
53.7% (n = 185) came from the Day Clinic at Bellevue
Hospital and 64.2% (n = 159) came from the Hagley
Park Health Centre. Most of the sample were diagnosed

A. E. Pusey-Murray et al. / J. Biomedical Science and Engineering 3 (2010) 602-611 605
with schizophrenic disorder (58.7%, n = 202), with other
health conditions being depression (13.1%, n = 45), bi-
polar disorder (10.5%, n = 36), and drug-induced psy-
chosis (4.7%, n = 16). Sixty-two percent (n = 212) of the
sample had secondary level education; 64.8% were un-
employed; 93.0% lived with either parents, partner,
children, sibling or in a nursing home; 84.1% had family
support; 80.5% were not in a union relationship (unmar-
ried, 58.7%; separated, 13.4%: divorced, 4.6%; widowed,
3.8%); 10.5% were married; 44.7% complied with the
instruments of the medical practitioners in regards to
medication usage; and the mean age was 43.6 years (±
1.5 years).
The most common reasons among those who did not
comply with the specifications of the medications
(55.3%, n = 189) were ‘medication makes me feel too
drowsy’ – 19.0%; ‘I was out of medication’ – 18.0%; ‘I
forgot’ – 16.9%; and ‘medications make me feel worse’
– 10.1%.
The majority of the participants (76.2%, n = 262) in-
dicated that they were able to purchase the medications.
The participants indicated that if the medication was not
available at the hospital or health centre, they would
purchased them from a private pharmacy (65.1%), wait
for the next appointment in order to see if it was now
available (16.6%), and used a home remedy (0.6%).
According to Table 1, no significant statistical rela-
tionship was found between the type of medical facility
utilized and, gender, employment status, education or
age (P > 0.05); nor for compliance (P > 0.05) or in-
sightfulness (P > 0.05). However, an association did
exist between the type of medical facility utilized and
marital status (P < 0.05). Twenty-four percent of those
who utilized the Hagley Park Health Care Centre were in
intimate unions (relationships), compared to 15.6% of
those who visited the Bellevue Hospital.
Gender was the only demographic variable that was
associated with diagnosed health conditions (P < 0.05;
Table 2). Most of the participants were diagnosed with
schizophrenic disorder, with more males than females
with the illness. Females were more likely to be diag-
nosed with bipolar disorder and depression than males.
However, there were more males diagnosed with
drug-induced psychosis (5.9%) than females (2.8%).
A statistical relationship was found between gender
and compliance/non-compliance of participants (P <
0.05). Males were more likely to comply with the speci-
fications of their medication (51%) than females (35.9%).
In addition, significant statistical associations were
found between family support and compliance/non-
compliance (P < 0.05), and family support and gender of
respondents (P < 0.05). Males (59.7%) had more family
support compared with females (40.3%), and more of
Table 1. Demographic characteristics of sample.
Hagley
Park Clinic
(n = 158)
Outpatient
Day clinic at
Bellevue
(n = 186)
Va riab le
n (%) n (%)
P
Gender NS
Male 92 (58.2) 110 (59.1)
Female 66 (41.8) 76 (40.9)
Marital status < 0.05
Married 16 (10.1) 20 (10.8)
Common-law 22 (13.9) 9 (4.8)
Widowed 8 (5.1) 5 (2.9)
Divorced 6 (3.8) 10 (5.4)
Separated 16 (10.1) 30 (16.1)
Unmarried 90 (57.0) 112 (60.0)
Employment status NS
Employed 48 (28.5) 54 (29.0)
Unemployed 112 (70.9) 128 (68.8)
Other 1 (0.6) 4 (2.2)
Education NS
Primary or below 36 (30.4) 66 (35.5)
Secondary
(including vocational) 100 (63.3) 112 (60.2)
Tertiary 10 (6.3) 8 (4.3)
Compliance NS
Yes 65 (41.1) 88 (47.8)
No 93 (58.9) 96 (52.2)
Insightfulness NS
Insight 83 (52.5) 83 (44.9)
Undecided 3 (1.9) 15 (8.1)
Lack insight 72 (45.6) 87 (47.0)
Age Mean (SD) 42.1 years
(1.4)
44.1 years
(1.2) NS
NS not significant
those with family support complied with the specifica-
tions of the medications (53.3%).
Approximate thirty seven percent (37.3%) of the par-
ticipants were knowledgeable about the medication they
were taking. Sixty-two percent of those who were
knowledgeable about the medication indicated that they
received the information from their medical doctors,
16.9% from pharmacists and 15.4% from nurses. Con-
currently, when they were asked to state who was re-
sponsible for controlling their illness, one-half (50%)
indicated their medical doctor was responsible for con-
trolling their illness, nurse 10.5%, pharmacists 9.0%,
God 9.0%, self 4.1%, and other 3.5%.
The majority of the participants indicated that apart
from medication and other health care professionals,
other factors that controlled their illness were: praying
(52.4%), cigarettes (11.0%), smoking marijuana (10.5%),
fasting (5.05), leaves from sour-sop tree (3.5%) and
Obeah (0.3%). A cross-tabulation between gender and
insightfulness revealed a significant statistical associa-
tion (P < 0.05). Lacking insightfulness was greatest for
males (68.6%) and insightfulness was greatest among
f
emales (50.0%).

606 A. E. Pusey-Murray et al. / J. Biomedical Science and Engineering 3 (2010) 602-611
Table 2. Diagnosed mental illness by demographic characteristic
Health condition
Schizophrenia
n = 202
Bipolar
Disorder
n = 36
Depression
n = 45
Drug
Induced
psychosis
n = 16
Other
n = 7
Undiagnosed
n = 38
P
Characteristic
n (%) n (%) n (%) n (%) n (%) n (%)
Gender < 0.05
Male 125 (61.9) 19 (9.4) 22 (10.9) 12 (5.9) 2 (1.0) 22 (10.9)
Female 77 (54.2) 17 (12.0) 23 (16.2) 4 (2.8) 5 (3.5) 16 (11.3)
Age group NS
≤ 20 years 1 (0.5) 1 (2.8) 12 (26.7) 1 (6.2) 0 (0.0) 0 (0.0)
21 – 39 years 90 (44.6) 18 (50.0) 23 (51.1) 12 (75.0) 4 (57.1) 10 (26.3)
40 – 59 years 87 (43.1) 17 (47.2) 10 (22.2) 3 (18.8) 2 (28.5) 23 (60.5)
60+ years 24 (11.9) 0 (0.0) 0 (0.0) 0 (0.0) 1 (14.2) 5 (13.2)
Marital status NS
Married 17 (8.4) 5 (13.9) 9 (20.0) 3 (18.7) 2 (28.6) 1 (2.6)
Common-law 14 (6.9) 4 (11.1) 8 (17.8) 2 (12.5) 1 (14.3) 1 (2.6)
Widowed 8 (4.0) 1 (2.8) 1 (2.2) 0 (0.0) 0 (0.0) 2 (5.3)
Divorced 10 (4.9) 2 (5.6) 2 (4.4) 1 (6.3) 0 (0.0) 2 (5.3)
Separated 28 (13.9) 4 (11.1) 5 (11.2) 2 (12.5) 1 (14.3) 6 (15.8)
Unmarried 125 (61.9) 20 (55.5) 20 (44.4) 8 (50.0) 3 (42.8) 26 (68.4)
Employment Status NS
Employed 48 (48.5) 14 (14.1) 13 (13.1) 6 (6.1) 2 (2.0) 16 (16.2)
Unemployed 150 (62.5) 21 (8.8) 32 (13.3) 10 (4.2) 5 (2.1) 22 (9.2)
Other 4 (80.0) 1 (20.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0)
Education NS
Primary or below 73 (36.1) 6 (16.7) 16 (35.6) 3 (18.8) 3 (42.8) 13 (34.2)
Secondary 119 (58.9) 27 (75.0) 27 (60.0) 11 (68.8) 4 (57.1) 24 (63.2)
Tertiary 10 (5.0) 3 (8.3) 2 (4.4) 2 (12.5) 0 (0.0) 1 (2.6)
NS not significant
4. DISCUSSION
In this study, the majority of the participants were diag-
nosed with schizophrenia, followed by depression, bipo-
lar disorder and drug induced psychosis which is consis-
tent with other findings in the medical literature [1,2].
The majority (84.1%) of the participants have family
support, but of this amount only 54% complied with
taking the medication as prescribed. This suggests that
family support is a significant determinant of medication
compliance. This finding corroborates that of Garcia and
colleagues, that the support from family caregivers pre-
dicts the use of psychiatric medication [25]. According
to Hickling [26] family members of mentally ill patients
in Jamaica take on the ongoing tasks of monitoring the
patient and supervising his or her medication program.
Along with the patient and the community mental health
service providers, family members participate in formu-
lating and carrying out plans for treatment and voca-
tional rehabilitation. Although such family involvement
is not unique, the extent to which the Jamaican commu-
nity mental health service has to rely on the extended
family is noteworthy.
Most (80.5%) of the participants were not in a stable
intimate partner relationship, a fact which could have
influenced the amount and the quality of social support
they received. Possible explanations could be the nega-
tive stigma associated with mental disorders in Jamaica
and their inability to provide financially in such a rela-
tionship, since 64.8% of the participants were unem-
ployed. However, in Jamaica, the fear and stigma asso-
ciated with mental illness has been greatly reduced,
mental illness is openly discussed in the media, and pa-
tients are comfortable receiving treatment for a variety
of psychiatric conditions in public and private commu-
nity treatment facilities [27].
Medication compliance among the participants in our
study was very low with only 44.7% taking their medi-
cation. Among those participants who did not comply
with the medication specifications (55.3%), 19% of this
group indicated they experienced drowsiness, and 10.1%
indicated that they felt worse. According to Feuertein
et al. [28] estimates of noncompliance ranges between
4% and 92%, with average from 30 to 35 percent. The
reason for non- compliance among the participants in
this study could be due to a number of factors such as:
discomfort resulting from treatment (side effects of
medication), expense of treatment, problems of filling a
prescription, decision based on personal value, judgment
or religious or cultural beliefs about the advantages and

A. E. Pusey-Murray et al. / J. Biomedical Science and Engineering 3 (2010) 602-611 607
disadvantages of the proposed treatment, maladaptive
personality, traits or coping style (for example the denial
of illness), the stigma attached to drugs for mental health
conditions and lack of understanding about the nature of
the illness [29].
According to Demyttenwere, the dropout rate of psy-
chiatric patients is attributed to various factors, such as
illness and patient’s characteristics, time taken to im-
prove or poor doctor-patient relationship [30]. Mann
indicated that the variance between what the doctors
prescribe and what the patients take can be reduced by
the doctor listening more carefully to the patients and
addressing their concerns about the side effects of the
medication [31]. In addition, educating the patients
about their disorder, pharmacological management and
addressing their concerns and fears of being addicted to
the drug increases their compliance [32]. In our study,
18% among the noncompliant participants suggested that
their medication was finished, and 16.9% indicated that
they forgot to take their medication. These patients can
be instructed about life skill techniques, cueing, about
Table 3. Compliance by demographic characteristic and health
condition.
Compliance
n = 153
Non-compliance
n = 189 Va riab le
n (%) n (%)
P
Gender <
0.05
Male 102 (66.7) 98 (51.9)
Female 51 (33.3) 91 (48.1)
Age group NS
≤ 20 years 1 (0.7) 2 (1.1)
21 – 39 years 59 (38.6) 87 (46.0)
40 – 59 years 68 (44.4) 85 (45.0)
60+ years 25 (16.3) 15 (7.9)
Marital status NS
Married 11 (7.2) 28 (14.8)
Common-law 16 (10.5) 12 (6.4)
Widowed 8 (5.2) 5 (2.6)
Divorced 6 (3.9) 10 (5.3)
Separated 17 (11.1) 28 (14.8)
Unmarried 95 (62.1) 106 (56.1)
Employment
status
NS
Employed 40 (26.1) 59 (31.2)
Unemployed 112 (73.2) 127 (67.2)
Other 1 (0.7) 3 (1.6)
Education NS
Primary or below 56 (36.6) 56 (29.6)
Secondary 88 (57.5) 124 (65.6)
Tertiary 9 (5.9) 9 (4.8)
Health condition NS
Schizophrenia 92 (60.1) 108 (57.1)
Bipolar disorder 14 (9.1) 22 (11.6)
Depression 20 (13.1) 25 (13.2)
Drug-induced
psychosis
5 (3.3) 11 (5.8)
Other 5 (3.3) 2 (1.2)
Undiagnosed 17 (11.1) 21 (11.1)
NS not significant
Table 4. Family support by demographic characteristic and
health condition.
Family support
Yes
n = 290
No
n = 37
Va riab le
n (%) n (%)
P
Gender < 0.05
Male 173 (59.7)
15
(40.5)
Female 117(40.3)
22
(59.5)
Age group NS
≤ 20 years 3 (1.0) 11 (29.7)
21 – 39 years 129 (44.5) 20
(54.1)
40 – 59 years 126 (43.5) 6 (16.2)
60+ years 32 (11.0) 0 (0.0)
Marital status NS
Married 26 (9.0) 2 (5.4)
Common-law 36 (12.4) 2 (5.4)
Widowed 15 (5.2) 2 (5.4)
Divorced 11 (3.8) 2 (5.4)
Separated 31 (10.7) 5 (13.5)
Never married 171 (58.9) 24
(64.9)
Employment status NS
Employed 74 (25.5)
18
(48.6)
Unemployed 213 (73.5)
18
(48.7)
Other 3 (1.0 1 (2.7)
Education NS
Primary or below 95 (32.8) 13
(35.1)
Secondary 181 (62.4)
22
(59.5)
Tertiary 14 (4.8) 2 (5.4)
Health condition NS
Schizophrenia 173 (59.7)
15
(40.5)
Bipolar disorder 31 (10.7) 4 (10.8)
Depression 36 (12.4) 8 (21.6)
Drug-induced psychosis 13 (4.5) 3 (8.2)
Other 7 (2.4) 0 (0.0)
Undiagnosed 30(10.3) 7 (18.9)
Compliance
Yes 154 (53.3)
10
(27.8) <0.05
No 135 (46.7)
26
(72.2)
NS not significant
the medication data displayed on the cap, and about
audiovisual feedback as this kind of instruction increases
medication compliance among psychiatric patients [33].
Interventions to increase medication compliance through
patient education in Jamaica are necessary because only
a minority (37.3%) of the participants reported that they
were knowledgeable about the medication they were
taking, and the majority (76.2%) of the participants in
dicated that they were able to buy their medication, so
the cost of the drug was not a prohibitive factor. Medical
doctors can play a major role in this regard because
62.0% of the participants who were knowledgeable

608 A. E. Pusey-Murray et al. / J. Biomedical Science and Engineering 3 (2010) 602-611
about their medication stated that they received informa-
tion from their doctor.
There is a significant relationship between marital
status and the clinic used, because 24.0% of the partici-
pants used the Hagley Park Health Centre Clinic com-
pared to 15.6% for the Bellevue Clinic. This is a sur-
prising finding which requires more research. One pos-
sible explanation for this finding is that the Bellevue
Hospital is the oldest and only major psychiatric hospital
in Jamaica. It is heavily stigmatized given the very nega-
tive perception Jamaicans have toward mental disorders
and people with mental disorders. Therefore, it is possi-
ble that the married participants attended the Hagley
Park Clinic to avoid the very negative stigma associated
with the Bellevue Clinic.
Many chronically ill patients take less of their medi-
cation than has been prescribed, owing to cost concerns,
especially those patients with low incomes, multiple
chronic health problems, or no prescription drug cover-
age [34]. The consequences of cost-related medication
underuse include increased emergency department visits,
psychiatric admissions, nursing home admissions, as
well as decreased health status [35]. In Jamaica there is
government assistance for low-income patients. Ac-
cording to McKenzie, few patients ask for it. In his study,
in one clinic, none of those who were prescribed medi-
cation (at least 35) asked for assistance, despite the fact
that none of them were working. It is customary for
families to pay for their relatives’ medication. Caregiv-
ers considered a lack of money and the need to pay for
prescriptions as a deterrent for attending the clinic [11].
There is a significant relationship between gender and
psychiatric disorder which is an unexpected finding. A
majority of males and females were diagnosed with
schizophrenia, with a higher incidence among males.
More males than females were diagnosed with drug-
induced psychosis, while more females were diagnosed
with depression and bipolar disorder than males. Further
research is required to understand the nuances of the
relationship between gender and mental disorder which
is an under-researched area. In addition, there is a related
significant relationship between gender and medication
compliance, in which males were more likely to comply
with their medication. This finding may be explained by
the significant relationship between insightfulness about
medication and gender, where males account for 68.0%
of the participants who lacked insightfulness. This find-
ing suggests that it is possible that there is greater medi-
cation compliance among males, because they are less
insightful about their medication and its side effects than
females. It is also possible that since the majority (59.7%)
of participants with family support are males, the social
support increases medication compliance in this group [25].
In the present study, non-compliance of medication-
taking among those with bipolar disorder and depression
was over 80%. Lack of insight [36] is a factor associated
with poor drug compliance. Hummer and Fleischacker
[37] explained that non-compliance with medication is
owing to the patients’ perception that the illness is not
serious to enough to warrant treatment compliance.
Conversely, a study by Cramer and Rosenheck found
that 58% of those diagnosed with psychosis and 65% of
those with depression complied with the treatment pre-
scription [38], which is significantly higher than a simi-
lar group in Jamaica. One researcher admitted that the
side effects of neuroleptics are real, but can be managed
[39], which clearly is accepted by those diagnosed with
drug psychosis and bipolar disorder in Jamaica
Patients’ belief in their ability to control their illness is
very important for medication compliance. However, the
participants in our study believed in a range of external
factors apart from medication that could be used to con-
trol their disorder. These external factors are: praying,
use of Obeah (Jamaican witchcraft), fasting, smoking
marijuana or use of sour-sop leaves, and smoking ciga-
rettes. There are also some participants (69.5%) who
believed that doctors, nurses and pharmacists were re-
sponsible for controlling their disorder. Thus the major-
ity of the participants in our study displayed an external
locus of control (rather than an internal locus of control)
which downplays the belief that they themselves have
major control over what happens in their lives and their
wellbeing. This can clearly has an effect on medication
compliance. These findings are corroborated by the
findings of Volis and colleagues, that the locus of con-
trol is important in medication compliance because it
mediates the relationship between medication compli-
ance and social support [40,41].
This study has contributed to the literature by un-
earthing a range of factors that are significantly related
to medication compliance among psychiatric patients in
Jamaica, which increases our understanding of this very
important health issue. There are some limitations to our
study. While our sample was taken from the two public
mental health clinics that treat the majority of psychiatric
cases in the country, these clinics are located in the met-
ropolitan region of Kingston. Therefore, it is possible
that our sample did not capture psychiatric patients from
other urban centres and rural areas. Therefore only cau-
tious generalizations should be made. In addition, there
is the possibility of social desirability bias, where some
of the patients told the interviewers what they wanted to
hear to get their approval. It is also possible that the pa-
tients’ mental disorders affected the accuracy of their
self-reports. Despite the aforementioned fact, medication
non-compliance among schizophrenic patients in this
research was low and in keeping with another in a non-
Caribbean nation which found that it ranges between

A. E. Pusey-Murray et al. / J. Biomedical Science and Engineering 3 (2010) 602-611 609
24-90% [42]. This work also concurs with the findings
of the Jamaican Ministry of Health, which found that
medication compliance among schizophrenia patients in
Jamaica was 69% [43].
6. CONCLUSIONS
Medication compliance among males is about average,
but is extremely low among females. The majority of the
participants with an average age of 43.6 years from the
Bellevue and Hagley Park mental health outpatient clin-
ics do not comply with their medication regimen for
schizophrenia, bipolar disorder, depression and psycho-
sis among other disorders. This non-compliance can be
explained by three significant factors. These are gender,
which is related to the factor of reduced insightfulness
about taking medication and family support. Non- com-
pliance may also be explained by the locus of control
which was not tested in this study, but the majority of
the noncompliant participants believe that factors exter-
nal to them have greater control over their illness than
they do. The current study highlights the challenge for
public health practitioners and policy makers in ad-
dressing this high non-compliance of medication among
the mentally ill in Jamaica. A pertinent finding of this
study is the fact that the level of education did not
change compliance (or non-compliance) among mentally
ill patients, suggesting that there is a need for more re-
search to unearth the range of factors that influence
medication compliance, the belief system of mentally ill
patients, an examination of alternative approaches to the
treatment of mental illness, and a social intervention
programme that is geared towards holistic education
strategies in patient care. What is evident from the study
is the fact that medication compliance can be explained
by 1) the perception of the severity of the illness and the
usefulness of the relevant medication, and 2) the per-
ceived side-effects of prescribed medications.
Medication non-compliance places mentally ill pa-
tients at great risk of exacerbation of their symptoms,
homelessness and interruptions in their daily lives, so
much so that this has become a public health concern
which must be addressed with urgency and care. Family
support emerged as a positive determinant of medication
compliance, suggesting that public health practitioners
must begin to explore the role of social support in treat-
ing mentally ill individuals, as well as aiding in the drive
for increased medication compliance among these indi-
viduals.
REFERENCES
[1] Royes, K. (1962) The incidence and features of psycho-
sis in a Caribbean community. Journal of Social Psy-
chiatry, 2, 1121-1125.
[2] Burke, A.W. (1974) First admissions and planning in
Jamaica. Journal of Social Psychiatry, 9, 39-45.
[3] Hickling, F.W. (1991) Psychiatric hospital admission
rates in Jamaica, 1971 and 1988. British Journal of Psy-
chiatry, 159, 817-821.
[4] Kane, J.M. and Marder, S.R. (1993) Psychopharma-
cologic treatment of schizophrenia. Schizophr Bull, 19,
287-302.
[5] Crane, G.E. (1973) Clinical psychopharmacology in its
20th year: Late, unanticipated effects of neuroleptics may
limit their use in psychiatry. Science, 181, 124-128.
[6] Davis, J.M. and Gierl, B. (1984) Pharmacological treat-
ment in the care of schizophrenic patients. In: Bellack,
A.S., Ed. Treatment and care for schizophrenia. Grune &
Stratton, Orlando.
[7] Johnson, D.A. (1988) Drug treatment of schizophrenia.
In: Bebbington, P. and McGuffin, P., Eds. Schizophrenia:
The major issues. Heinemann Professional, Oxford, 158-
171.
[8] Ellenbroek, B.A. (1993) Treatment of schizophrenia: A
clinical and preclinical evaluation of neuroleptic drugs.
Pharmacology & Therapeutics, 57, 1-78.
[9] Kane, J., Honigfeld, G., Singer, J., et al. (1988) Clozap-
ine for the treatment-resistant schizophrenic. A dou-
ble-blind comparison with chlorpromazine. Archives of
General Psychiatry, 45, 789-796.
[10] Tuunainen, A., Wahlbeck, K. and Gilbody, S.M. (2002)
Newer atypical antipsychotic medication in comparison
to clozapine: A systematic review of randomised trials.
Schizophrenia Research, 56, 1-10.
[11] McKenzie, K. Jamaica: Community mental health services.
http://www.publications.paho.org/english/Jamaica_CD_1
83. pdf
[12] Bruer, J.T. (1982) Methodological rigor and citation
frequency in patient compliance literature. American
Journal of Public Health, 72, 911-1123.
[13] Razali, M.S. and Yahya, H. (1995) Compliance with
treatment in Schizophrenia: A drug intervention program
in a developing country. Acta Psychiatrica Scandinavica,
91, 331-335.
[14] Montgomery, L.A. (2002) The relationship between
health belief model constructs and medication compli-
ance in the treatment of bipolar disorder. Dissertation
Abstracts International: Section B: The Sciences and En-
gineering, 62(12-B), 5974.
[15] Cohen, N.L., Parikh, S.V. and Kennedy, S.H. (2000)
Medication compliance in mood disorders: Relevance of
the Health Belief Model and other determinants. Primary
Care Psychiatry, 6(3), 101-110.
[16] Lund, V.E. (2000) Perceived quality of life for persons
with bipolar disorder: The role of medication compliance,
family and health stress, level of coping, and health locus
of control. Dissertation Abstracts International: Section
B: The Sciences and Engineering, 60(7-B), 3198.
[17] Buckalew, L.W. and Sallis, R.E. (1986) Patient compli-
ance and medication perception. Journal of Clinical
Psychology, 42(1), 49-53.
[18] Thompson, C., Peveler, R.C., Stephenson, D. and McKen-
drick, J. (2000) Compliance with antidepressant medica-
tion in the treatment of major depressive disorders in

610 A. E. Pusey-Murray et al. / J. Biomedical Science and Engineering 3 (2010) 602-611
primary care: A randomized comparison of fluoextine
and a tricyclic antidepressant. The American Journal of
Psychiatry, 157(3), 338-343.
[19] Diaz, E., Levine, H.B., Sullivan, M.C., et al. (2001) Use
of the Medication Events Monitoring System to estimate
medication compliance in patients with schizophrenia.
Journal of Psychiatry and Neuroscience, 26(4), 325-329.
[20] Janssen, B., Gaebel, W., Haerter, M., Komaharadi, F.,
Lindel, B. and Weinmann, S. (2006) Evaluation of fac-
tors influencing medication compliance in patient treat-
ment of psychotic disorders. Psychopharmacology, 187
(2), 229-236.
[21] Hayward, P., Chan, N., Kemp, R. and Youle, S. (1995)
Medication self-management: A preliminary report on an
intervention to improve medication compliance. Journal of
Mental Health, 4(5), 511-517.
[22] Seltzer, A., Roncari, I. and Garfinkel, P.E. (1980) The
effect of patient education on medication compliance.
Canadian Journal of Psychiatry, 25(8), 638-645.
[23] Cramer, J.A. and Rosenheck, R. (1999) Enhancing medi-
cation compliance for people with serious mental illness.
Journal of Nervous and Mental Disease, 187(1), 53-55.
[24] (1993) Demographic Statistics, Jamaica 1992. Statistical
Institute of Jamaica, Kingston.
[25] Garcia, J.I., Chang, C.L., Young, J.S., Lopez, S.R. and
Jenkins, J.H. (2006) Family support predicts psychiatric
medication use among Mexican American individuals
with schizophrenia. Social Psychiatry and Psychiatric
Epidemiology, 41(8), 624-631.
[26] Hickling, F.W. (1975) An analysis of the 1873 Mental
Hospital Law in Jamaica and its effect on mental treat-
ment facilities. Jamaica Law Journal, 64-69.
[27] Hickling, F.W. (1994) Community psychiatry and dein-
stitutionalization in Jamaica. Hospital & Community
Psychiatry, 45, 1122-1126.
[28] Feuertein, M., Labbe, E.E. and Kuegmierezyk, A.R.
(1986) Health psychology: A psychobiological perspec-
tive. Pleneum Press, New York.
[29] Adewuya, A.O., Ola, B.A., Mosaku, S.K., Fatoye, F.O.
and Eegunranti A.B. (2006) Attitude towards antipsy-
chotics among outpatients with schizophrenia in Nigeria.
Acta Psychiatrica Scandinavica, 113, 207-211.
[30] Demyttenwere, K. (1997). Compliance during treatment
with antidepressants. Journal of Affective Disorder, 43,
27-39.
[31] Mann, J.J. (1986) How medication compliance affects
outcome. Psychiatric Annals, 16(10), 567-570.
[32] Seltzer, A., Roncari, I. and Garfinkel, P.E. (1980) The
effect of patient education on medication compliance.
Canadian Journal of Psychiatry, 25(8), 638-645.
[33] Cramer, J.A. and Rosenheck, R. (1999) Enhancing medi-
cation compliance for people with serious mental illness.
Journal of Nervous and Mental Disease, 187(1), 53-55.
[34] Safran, D.G., Neuman, P., Schoen, C., et al. (2002) Pre-
scription drug coverage and seniors: how well are states
closing the gap? Health Affairs, W253-W268.
[35] Tamblyn, R., Laprise, R., Hanley, J.A., et al. (2001) Ad-
verse events associated with prescription drug cost-
sharing among poor and elderly persons. Journal of the
American Medical Association, 285, 421-429.
[36] Kane, J.M. (1985) Compliance issues in outpatient treat-
ment. Journal of Clinical Psychopharmacology, 5, 22-27.
[37] Hummer, M. and Fleischhacker, W.W. (1999) Ways of
improving compliance. In: Lader, M. and Naber, D., Eds.,
Difficult clinical problems in psychiatry. Martin Dunitz,
London, 229-238.
[38] Cramer, J.A. and Rosenheck, R. (1998) Compliance with
medication regimens for mental and physical disorders.
Psychiatric Services, 49, 196-201.
[39] Chen, A.M. (1991) Non-compliance in community psy-
chiatry: A review of clinical interventions. Hospital and
Community Psychiatry, 42(3), 282-287.
[40] Voils, C.I., Steffens, D.C., Flint, E.P. and Bosworth, H.B.
(2005) Social support and locus of control as predictors
of adherence to antidepressant medication in an elderly
population. The American Journal of Geriatric Psychia-
try, 13(2), 157-165.
[41] Meehan, A.J. (1995) From conversion to coercion: The
police role in medication compliance. Psychiatric Quar-
terly, 66(2), 163-184.
[42] Perkins, D.M. (2002) Predictors of non-compliance in
patients with schizophrenia. Journal of Clinical Psychia-
try, 63(12), 1121-1181.
[43] (2008) Mental health and substance abuse unit. Annual
Report, Ministry of Health, Jamaica (MOH), Kingston.

A. E. Pusey-Murray et al. / J. Biomedical Science and Engineering 3 (2010) 602-611 611
Appendix
Question Particular
Ques 1-8 Demographic characteristics
Gender; age at last birth day; educational level; marital status;
Employment status; source of income; whom do you live with
Ques 9 Do they assist or support you in taking care of your conditions?
Ques 10 What is your diagnosis?
Ques 11 Do you know what medication(s) you are taking?
Ques 12 How many different types of medication do you take daily?
One; Two; Three; Four; Five; Six or more
Ques 13 Are you on monthly injection for psychiatric illness?
Ques 14 How often are you to take your medication?
Daily, less than three times per week; more than three times per week
Ques 15 I take my medication as prescribed by my doctor
Always; most times; Sometimes; Rarely; Never
Ques 16 What are the reasons for not taking your medication as is prescribed?
Ques 17 Did you take the medication this morning? Yes or no
Ques 18 I know when I will become sick? Strongly agree; agree; unsure; disagree; strongly disagree
Ques 19 Medication should only be taken if you are experiencing a symptom of mental illness? Strongly agree;
agree; unsure; disagree; strongly disagree
Ques 20 Were you told about the medication that you are taking? Yes or no
Ques 21 If yes (Ques 20), whom? Doctor; Nurses; Pharmacist; Other
Ques 22-30 Questions on medication availability; health care insurance coverage;
dispensary of medication; waiting time for medication dispensary
Ques 31 Will counselling on how to take medications help me to take them better?