See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/321494401
Economic cost and quality of life of family caregivers of schizophrenic
patients attending psychiatric hospitals in Ghana
Article  in  BMC Health Services Research · November 2017
DOI: 10.1186/s12913-017-2642-0
CITATIONS READS
5 97
9 authors, including:
Yaw Nyarko Opoku-Boateng Irene A Kretchy
      University of Ghana1 PUBLICATION 5 CITATIONS   
34 PUBLICATIONS   207 CITATIONS   
SEE PROFILE
SEE PROFILE
Genevieve Cecilia Aryeetey Duah Dwomoh
School of Public Health University of Ghana University of Ghana
32 PUBLICATIONS   467 CITATIONS    6 PUBLICATIONS   31 CITATIONS   
SEE PROFILE SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Current issues in malaria management View project
Mental Health Project View project
All content following this page was uploaded by Samuel Agyei Agyemang on 05 December 2017.
The user has requested enhancement of the downloaded file.
Opoku-Boateng et al. BMC Health Services Research 2017, 17(Suppl 2):697
DOI 10.1186/s12913-017-2642-0
RESEARCH Open Access
Economic cost and quality of life of family
caregivers of schizophrenic patients
attending psychiatric hospitals in Ghana
Yaw Nyarko Opoku-Boateng1,3, Irene A. Kretchy2, Genevieve Cecilia Aryeetey3, Duah Dwomoh4, Sybil Decker5,
Samuel Agyei Agyemang3, Yesim Tozan6, Moses Aikins3 and Justice Nonvignon3*
Abstract
Background: Low and middle income countries face many challenges in meeting their populations’ mental health
care needs. Though family caregiving is crucial to the management of severe mental health disabilities, such as
schizophrenia, the economic costs borne by family caregivers often go unnoticed. In this study, we estimated the
household economic costs of schizophrenia and quality of life of family caregivers in Ghana.
Methods: We used a cost of illness analysis approach. Quality of life (QoL) was assessed using the abridged WHO
Quality of Life (WHOQOL-BREF) tool. Cross-sectional data were collected from 442 caregivers of patients diagnosed
with schizophrenia at least six months prior to the study and who received consultation in any of the three
psychiatric hospitals in Ghana. Economic costs were categorized as direct costs (including medical and non-medical
costs of seeking care), indirect costs (productivity losses to caregivers) and intangible costs (non-monetary costs
such as stigma and pain). Direct costs included costs of medical supplies, consultations, and travel. Indirect costs
were estimated as value of productive time lost (in hours) to primary caregivers. Intangible costs were assessed
using the Zarit Burden Interview (ZBI). We employed multiple regression models to assess the covariates of costs,
caregiver burden, and QoL.
Results: Total monthly cost to caregivers was US$ 273.28, on average. Key drivers of direct costs were medications
(50%) and transportation (27%). Direct costs per caregiver represented 31% of the reported monthly earnings. Mean
caregiver burden (measured by the ZBI) was 16.95 on a scale of 0–48, with 49% of caregivers reporting high
burden. Mean QoL of caregivers was 28.2 (range: 19.6–34.8) out of 100. Better educated caregivers reported lower
indirect costs and better QoL. Caregivers with higher severity of depression, anxiety and stress reported higher
caregiver burden and lower QoL. Males reported better QoL.
Conclusions: These findings highlight the high household burden of caregiving for people living with schizophrenia
in low income settings. Results underscore the need for policies and programs to support caregivers.
Keywords: Economic burden, Quality of life, Caregiving, Schizophrenia, Ghana
* Correspondence: jnonvignon@ug.edu.gh
3Department of Health Policy, Planning and Management, School of Public
Health, College of Health Sciences, University of Ghana, Legon, Ghana
Full list of author information is available at the end of the article
© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. 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.
Opoku-Boateng et al. BMC Health Services Research 2017, 17(Suppl 2):697 Page 40 of 87
Background family caregivers. In this study, we sought to address this
Mental illness accounts for about 32% of years lived gap by estimating the economic burden and quality of
with disability (YLD) and about 13% of disability- life of family caregivers of schizophrenic patients in
adjusted life-years (DALYs), a percentage that equals Ghana.
that of cardiovascular and circulatory diseases [1].
Schizophrenia is a severe mental disorder character- Methods
ized by significant distortions in thinking and percep- Study setting
tion, accompanied by an exhibition of inappropriate The study was undertaken at the three psychiatric
emotions [2, 3]. Schizophrenia alters individuals’ hospitals in Ghana: Accra, Pantang and Ankaful.
perception of reality, often making them think and Accra Psychiatric Hospital is a 600-bed hospital
act in ways that are strange or abnormal by socially commissioned in 1906 as the first psychiatric hospital
sanctioned standards. The World Health Organization in Ghana and is located in the Greater Accra Region.
(WHO) estimates that about 21 million people suffer Established in 1975 with a bed capacity of 500, the
from schizophrenia globally [4]. It is estimated that Pantang Psychiatric Hospital is situated in Pantang,
13% of Ghanaians suffer from at least one form of mental also in the Greater Accra Region. Ankaful Psychiatric
or substance abuse disorder and that about 32% of all Hospital, which was built in 1965, is a 350-bed hos-
mental health cases managed in the country’s three pital located near the coastal town of Cape Coast in
psychiatric hospitals are schizophrenic [5]. the Central Region of Ghana that serves patients from
Caregiving plays a significant role in the manage- the Central, Western and Ashanti regions of Ghana
ment of schizophrenia. For example, the WHO indi- and some neighbouring countries. All the hospitals
cates that globally up to 90% of schizophrenic are located in the southern part of the country and
patients live with their families [6]. In many low and serve the country’s population of approximately 25
middle income countries (LMICs), the social structure million [18].
of families and economic hardship impede the use of
paid caregivers. Other health systems challenges, such Study population and sample size
as inadequate health personnel and poor infrastruc- The study population comprised family (primary) care-
ture, further increase the burden of caring for people givers of schizophrenic patients reporting to the out-
with mental disorders. patient department at each study site. Based on Cochran
A key provision of Ghana’s 2012 Mental Health Act [19], the sample size was calculated using the formula
(Act 846) is to de-institutionalize mental health care (i.e., below:
provide community-based care for people with mental 
 
disorders) in order to decongest the three psychiatric 2Z∝ pð1−pÞ
hospitals. The role of caregivers in this community-based n0 ¼ ð 2Deff Þ 2
approach is crucial since family interactions influence e
treatment outcomes and relapse rates [7]. However, as in Where n0 is the minimum required sample size; Z
2 is an
other LMICs, family caregivers of schizophrenic patients abscissa of the curve that cuts off an area ∝ at the tail (1
in Ghana bear significant economic, psychological and so- – α equals the desired confidence level, i.e., 95%); e is
cial burdens, which are often unaccounted for in interven- the desired level of precision; p is the estimated propor-
tions [8–15]. For instance, the time required to care for tion of schizophrenic patients that is present in the
schizophrenic patients affects the productivity of care- population, which was assumed to be 50% since the
givers, often limiting their participation in the labor force, current proportion is unknown; and Deff is the design
which has economic implications for the family. Families effect. For a 95% confidence interval, Z∝ is 1.96 and the
also incur substantial medical and non-medical costs asso- 2level of precision e (margin error for the study)±5%.
ciated with the management of the condition, which may Assuming a design effect of 1.03, we computed a mini-
lead to financial impoverishment. In addition, caring mum sample size of.
people with mental disorders places significant psycho-
logical burden on caregivers and has been shown to nega- 1:962  0:5 0:5
tively affect caregivers’ quality of life (QoL) [16, 17]. Other n0 ¼ 1:03 2 ¼ 395:684≈396:0:05
social burdens that families bear include stigma, which
affects the social support that is much needed for such Assuming a 90% response rate, we estimated a mini-
patients and their families. mum sample size of 440 patients. We computed the
Although Ghana has a high burden of mental illness, sample size required in each facility using probability
few studies have examined the burden of mental illness proportionate to size (PPS) of number of schizophrenic
in general—and schizophrenia in particular—on unpaid patients (Table 1).
Opoku-Boateng et al. BMC Health Services Research 2017, 17(Suppl 2):697 Page 41 of 87
Table 1 Sample size for study sites calculated by adding the total number of hours spent
Hospital Number of patients Sample size Percentage travelling to and from the hospital. Waiting time consti-
Accra 11,256 228 51.8 tuted the total time spent waiting for and receiving treat-
Pantang 6780 138 31.4 ment (i.e., from the time the caregiver and patient
arrived at the hospital to the time they left the hospital).
Ankaful 3660 74 16.8
The time used for other caregiving activities, such as
Total 21,696 440 100.0 household activities and leisure, was also calculated and
included in the indirect costs. Total indirect cost was
In each hospital, folders of patients diagnosed with calculated as the product of the sum of total time spent
schizophrenia at least six months preceding the on personal care for patient, transport and waiting for
study were sorted and reviewed using criteria health care, other household activities for the patient (in
outlined in the 10th revision of the International hours) and the respective (hourly) wage rate. Total
Statistical Classification of Diseases and Related economic cost constituted the sum of direct and indirect
Health Problems (ICD-10), and primary caregivers costs.
who attended the hospital with the patients identi- Intangible costs (sometimes referred to as caregiver
fied and interviewed. Accompanying relatives who burden) were assessed using the 12-item Zarit Burden
were not the primary caregivers were excluded. Interview (ZBI) tool developed by Bedard et al. [20],
Given that many patients presenting for outpatient which comprises items assessing stress, pain, anxiety and
services were unaccompanied, accompanying primary depression. The intangible cost was analysed by
caregivers were recruited until the required sample summing the scores for all 12 items for each respondent.
size for the particular facility was met. The overall score ranges between 0 and 48. Scores were
then categorized as low burden (0–16) or high burden
Data collection and tools (17–48).
Data were collected using a structured, interviewer- Quality of life was assessed using the abridged WHO
administered questionnaire with closed- and open-ended Quality of Life (WHOQOL-BREF) tool. The WHOQOL-
questions. The questionnaire elicited information on care- BREF comprises 26 items: two items assessing the over-
givers’ sociodemographic characteristics, direct health care all quality of life and general health; and 24 assessing
costs (medical and non-medical costs) and indirect costs satisfaction in four main domains – seven items on
(i.e. productivity losses) to caregivers. physical health, six items on psychological health, three
items on social relationships, and eight items on envir-
Variables onmental health. Each item was rated on a scale of 1 to
We used a cost of illness analysis approach. Costs were 5 with 1 being the lowest and 5 being the highest QoL.
analysed from the caregiver perspective and included The mean score of items within each domain was used
costs incurred during the month preceding the data to calculate the domain mean. The mean score for each
collection. All costs were computed in United States domain was then multiplied by 4 (four) to make it com-
Dollars (US$) (3.9 Ghanaian Cedi (GHS) ≈ 1 US$). Costs parable with the full WHOQOL tool (WHOQOL-100)
were not adjusted for inflation. Costs were categorized [21]. The scores were scaled in a positive direction so
into direct, indirect and intangible costs. Direct costs that a higher score indicated a higher QoL.
were further grouped into medical and non-medical Other variables used in the analysis included Age (in
costs. Medical costs included the costs of drugs, consul- years) as at last birthday; Sex measured as a dummy, 1
tations, laboratory investigations and diagnostics and for male and 0 for female; Marital status: Married (1),
other therapies. Non-medical costs included the cost of single (0); Proximity of residence from health facility
travel to and from hospital, food costs incurred by the visited; 1 for far (living more than 30 min from facility,
caregiver during the treatment period, accommodation by typical mode of transport used) and 0 for close;
costs if the caregiver travelled with the patient away Caregiver’s highest level of education: no education (0),
from home, and miscellaneous costs such as telephone primary/junior high school (i.e. basic education or
costs related to medical care. The sum of medical and equivalent) (1), secondary (2) and tertiary (3); Caregiver’s
non-medical costs constituted the direct costs. employment status: unemployed (0), self-employed (1),
Indirect costs were estimated using the national daily private sector (2), public sector (3), student/apprentice
minimum wage of US$ 2.0 per day (May 2016) for care- (4); Caregivers were asked whether other members of
givers in the formal sector and a local daily agricultural the family spent time taking care of the patient (other
wage rate of US$ 4.50 was used for caregivers working family support): yes (1), no (0); and Caregiver’s mental
in the informal sector. Unemployed caregivers were con- health status (i.e. depression, anxiety, stress). Mental
sidered part of the informal sector. Travel time was health status was assessed using the Depression Anxiety
Opoku-Boateng et al. BMC Health Services Research 2017, 17(Suppl 2):697 Page 42 of 87
Stress Scale (DASS-21), which comprises of three sub- Table 3 Socio-demographic characteristics of caregivers
scales: Depression (DASS-D), Anxiety (DASS-A), and (N = 444)
Stress (DASS-S). Responses on each item ranged from 0 Characteristic Number (%)
(did not apply to me at all) to 3 (applied to me very Sex
much). Male 193 43.5
Reliability of the QoL and Zarit burden scale was
Female 251 56.5
assessed using Cronbach’s alpha. With the exception of
the depression sub-scale of the DASS-21 and the social Age
sub-scale of the WHOQOL, the Cronbach alpha values < 20 3 0.7
fell within the acceptable range of alpha values (0.70 to 20–29 51 11.5
0.95) [22–24] (Table 2). 30–39 95 21.4
40–49 103 23.2
Data analysis
50–59 89 20.0
Descriptive statistics (mean, standard deviation) of
study variables are presented in Table 3. Two sets of 60–69 76 17.1
multivariate linear regression models were fitted to > 69 27 6.1
assess factors associated with economic costs and Marital Status
QoL. The first set of models had direct and indirect Married 294 66.2
costs as dependent variables, and demographic and Single 150 33.8
socioeconomic factors (e.g., age, sex, marital status,
Religion
proximity of residence to facility, highest education
level and employment status) as explanatory variables. Christian 393 88.5
The second set of models had the four domains of Muslim 45 10.1
QoL as dependent variables. Explanatory variables Traditionalist 2 0.5
comprised direct costs, indirect costs, Zarit burden Other 4 0.9
scores, time and duration of care provided, mental Educational Level
health indices (anxiety, depression, stress). The
No education 50 11.3
models also controlled for caregivers’ sociodemo-
graphic characteristics. A final model was fitted to in- Primary 146 32.8
vestigate the covariates of caregiver burden (measured Secondary 149 33.6
by the ZBI score). Tertiary-Graduate/Post Graduate 99 22.3
Employment Status
Results Self employed 240 54.2
Background characteristics of caregivers
Private sector 70 15.8
Caregivers’ sociodemographic characteristics are summa-
rized in Table 3. Caregivers’ mean age was 47 years (45% Public sector 41 9.2
were aged 30–49 years and 37% aged 50–69 years). Unemployed 78 17.6
Fifty-seven percent of caregivers were female and 66% Student/Apprentice 14 3.2
Table 2 Internal consistency of the mental health status, quality were married. Twenty-two percent of the caregivers had
of life, and Zarit burden indices tertiary education: 11% had university degrees, and 11%
Domain Number of items Cronbach’s alpha had certificate, diploma or post-diploma qualifications.
Mental health status About 34% of the caregivers had secondary education
Depression 7 0.67 and 11% had no formal education. In terms of employ-
ment status, 54% of caregivers were self-employed, 16%
Anxiety 7 0.80
were working in the private sector, 9% were employed in
Stress 7 0.83 the public sector, and 18% were unemployed. About 3%
Quality of life of the caregivers were students or apprentices.
Physical 7 0.82
Psychological 6 0.77 Economic costs of caregiving
Social 3 0.56 The average caregiving cost per month was US$273.28,
with indirect costs being $242.95 and direct costs being
Environmental 8 0.81
$30.36 (Table 4). Total cost for the study sample was
Zarit burden index 12 0.83 $76,839.54, with indirect costs accounting for about
Opoku-Boateng et al. BMC Health Services Research 2017, 17(Suppl 2):697 Page 43 of 87
Table 4 Economic costs to caregiving for schizophrenia
Cost component N Total costa (US$) Standard deviation Cost profileb (%) Average cost (US$)
Direct costs
Direct medical cost
Consultation 444 1411.02 7.02 1.8 3.17
Drugs 444 6668.49 22.59 8.7 15.01
Lab/Diagnostics 442 428.20 5.38 0.5 0.97
Other 442 144.64 4.20 0.2 0.33
Sub total 8652.36 11.2 19.50
Direct non- medical costs
Transportation 444 3675.57 9.62 4.8 8.28
Meals 442 843.15 5.56 1.1 1.90
Lodging 442 153.20 5.05 0.2 0.35
Miscellaneous 442 142.69 1.20 0.2 0.33
Sub total 4814.62 6.3 10.85
Total direct costs 13,466.97 17.5 30.36
Indirect costs
Formal sector 131 9118.97 40.26 11.9 69.61
Informal sector 313 54,253.59 88.38 70.6 173.33
Total indirect costs 444 63,372.56 82.5 242.95
Grand Total Cost 76,839.54 100 273.28
aUS$ 1.00 equivalent to GHS 3.9 (Bank of Ghana average monthly interbank exchange rate, June 2016)
bPercentages computed by dividing the total cost for the item by the grand total cost and multiplying the result by 100
82.5%. The key components of direct costs were drug Tertiary education was significantly associated with
(about 50%) and transportation (27%) costs. indirect costs. Compared to those with no formal
education, indirect cost for caregivers with tertiary
Intangible costs of caregiving education were lower by GHS 216 (US$55.38, − 95% CI:
The mean caregiver burden score was 16.95 (SD 8.82) -329.41, −102.10, p-value = 0.0007). A unit increase in
out of a maximum of 48. About 51% of the caregivers age, resulted in a GHS 3.3 increase in indirect cost (US$
reported low burden (i.e., ZBI score of >16). The analysis 0.75, 95% CI: 0.99, 5.52, p < 0.05). Indirect cost for males
also revealed that the burden in caregiving was signifi- was GHS 61.1 lower compared to females (US$11.86,
cantly higher for females (61%) than for males (39%) 95% CI: -121.77, −0.41, p < 0.05).
among caregivers reporting high burden. With respect to predictors of ZBI score, anxiety, de-
pression and stress were related to Zarit burden score,
Caregivers’ quality of life as presented in Table 7. The burden of caregivers with
The mean quality of life scores by domain and back- severe anxiety was approximately 4% (95% CI: 1.02, 6.59;
ground characteristics are summarized in Table 5. The p = 0.0022) higher than those with no anxiety disorders,
average overall QoL was 28.2 (SD 12.0). Average scores controlling for other covariates.
in the four domains were 19.6 (physical), 29.1 (psycho- Table 8 shows that for a unit increase in the Zarit bur-
logical), 29.2 (social), and 34.8 (environmental). QoL den score, QoL in relation to the physical domain de-
scores across all domains were, on average, higher creased by 0.25 (95% CI: -0.38, −0.13, p = 0.0001).
among males (29.6) than females (27.1) (p < 0.05). Similar results were obtained for the psychological QoL
Caregivers who were married had higher QoL scores domains. The coefficients for the Zarit burden score,
than those who were unmarried. The mean score across taken for all the four QoL domains together, were statis-
all domains increased as the caregiver’s educational level tically significant (F = 5.30, p = 0.0004). There was no
increased. statistically significant relationship between the direct
and indirect cost and the four QoL domains. The QoL
Predictors of costs and QoL in relation to the physical and psychological domains
The multivariate linear multiple regression results for was higher among caregivers who reported family
covariates of economic cost are shown in Table 6. support. The QoL with reference to the psychological
Opoku-Boateng et al. BMC Health Services Research 2017, 17(Suppl 2):697 Page 44 of 87
Table 5 Mean quality of life scores, by domain and background characteristics
Physical Psychological Social Environmental Total
Mean (SD) p-value Mean (SD) p-value Mean (SD) p-value Mean (SD) p-value Mean (SD) p-value
Agea 0.303 0.257 <0.05 0.434 0.511
< 30 20.1 (17.5) 31.6(24.1) 35.2(27.0) 33.6(23.3) 30.1(13.3)
30–39 20.4 (17.6) 32.2(25.7) 30.2(24.6) 31.1(24.3) 28.5(11.7)
40–49 19.7 (15.9) 26.2(22.2) 27.6(24.7) 37.2(25.9) 27.8(11.3)
50–59 18.8 (14.4) 25.9(21.1) 23.4(21.2) 38.1(25.1) 26.6(12.1)
60+ 19.3 (15.5) 30.6(24.9) 31.6(24.6) 33.6(24.3) 28.8(12.2)
Sex b <0.001 <0.01 <0.05 <0.05 <0.05
Male 20.0 (16.1) 30.8(25.0) 31.0(26.3) 36.4(26.1) 29.6(12.6)
Female 19.3 (16.1) 27.9(22.7) 27.7(22.9) 33.6(23.8) 27.1(11.5)
Marital statusb
Married 20.0(15.7) 30.0(24.1) 29.1(23.6) 33.6(24.7) 28.1(11.7)
Single 19.0(16.4) 0.270 28.1(23.1) 0.330 29.4(26.1) 0.071 37.2(24.9) 0.436 28.4(12.5) 0.863
Religionc <0.05 0.872 0.486 0.270 0.972
Christian 19.1(15.4) 28.8(23.4) 29.6(24.6) 35.1(24.9) 28.2(11.9)
Muslim 23.9(20.3) 31.6(26.6) 25.8(23.4) 32.5(24.4) 28.513.0)
Educationa <0.01 <0.001 <0.001 <0.001 <0.001
None 19.1(16.4) 22.3(17.5) 23.2(19.4) 27.7(21.3) 23.1(8.4)
Primary 19.5(17.9) 26.2(20.2) 23.5(22.1) 30.6(24.7) 24.9(11.5)
Secondary 20.0(16.2) 28.3(22.5) 30.3(24.7) 35.4(23.7) 28.5(12.0)
Tertiary 19.1(14.8) 36.1(29.1) 31.9(26.5) 38.9(28.3) 31.5(12.7)
Employmenta 0.703 0.140 0.911 0.247 0.558
Unemployed 21.4(19.0) 31.6(25.6) 32.0(25.2) 32.7(23.8) 29.4(13.8)
Self-employed 19.6(15.7) 27.3(22.2) 29.0(24.6) 35.2(25.3) 27.8(11.3)
Private sector 17.4(12.5) 32.2(26.4) 27.7(24.4) 38.0(26.1) 28.8(11.8)
Public sector 19.6(16.3) 28.3(23.1) 26.6(22.5) 31.5(21.7) 26.6(12.3)
Total 19.6 (16.0) 29.1(23.7) 29.2(24.5) 34.8(24.8) 28.2(12.0)
aKruskal-Wallis test used in determining significance of difference
bWilcoxon Rank Sum test used in determining the significance of difference
cOther religions constituted 1% and was not considered in the analysis due to its insignificance
Opoku-Boateng et al. BMC Health Services Research 2017, 17(Suppl 2):697 Page 45 of 87
Table 6 Multivariate linear multiple regression results for covariates of economic cost
Cost
Direct Indirect Joint effect
Covariates β (95% CI) β (95% CI) F-test statistic p-value
**
0.71 3.25
Age in years (−0.17,1.59) (0.99,5.52) 10.18 0.0001***
Sex *
Female Ref Ref
Male 12.06 −61.09 3.65 0.0269*
(−11.58,35.70) (−121.77,-0.41)
Marital status
Single Ref Ref
Married 19.30 53.42 3.93 0.0203*
(−5.58,44.17) (−10.43,117.28)
Proximity *
Close Ref Ref
Far 32.33 0.84 2.91 0.0554
(5.52,59.15) (−68.00,69.67)
Education ***
None Ref Ref
Primary/junior high school 12.27(−26.56,51.11) −64.17(−163.87,35.52) 18.61 0.0001***
Secondary 16.19(−22.77,55.16) −79.76(−179.78,20.27)
Tertiary 39.84(−4.43,84.11) −215.75(−329.41,-102.10)
Employment status ***
None Ref Ref
Private 1.07(−40.83,42.96) −327.68(−435.22,-220.14)
Public 5.58(−41.69,52.86) −302.55(−423.91,-181.19) 2.36 0.0953
Self-employed 7.35(−25.48,40.17) −45.51(−129.79,38.76)
Student/apprentice 66.56(−6.26,139.38) −19.64(−206.58,167.30)
R2 50.8% 64.2%
β is the estimated effect of covariate; R2 is the adjusted coefficient off determination; CI is confidence interval; ref. is the reference category
*p < 0.05, ** p < 0.01, *** p < 0.001
domain was about 3% (95% CI: 0.68, 5.12; p = 0.0060) (95% CI: -0.28, −0.06; p = 0.0030), controlling for
higher for participants who reported family support other covariates in the model. Overall QoL was 17%
compared to those that did not. Depression was associ- lower for caregivers with depression compared to
ated with all the four domains of QoL (F = 19.20, those with no depression.
p = 0.0001). The QoL was 17% (95% CI: -22.92, −11.87;
p-value = 0.0001) lower for participants with severe Discussion
forms of depression compared to those with no depres- The aim of this study was to estimate the economic
sion. The results further showed that more highly costs and QoL of family caregivers of schizophrenic
educated caregivers reported higher QoL (7–15% more) patients attending three psychiatric hospitals in Ghana
than those who were uneducated, and male caregivers and to assess the factors associated with these two out-
reported higher QoL (2–6% more) than female comes. The average costs incurred by caregivers
caregivers. amounted to approximately US$273 per month, with
The results from the multiple linear regression about 82% of these costs being indirect costs (in terms
analysis in Table 9 indicated that a unit increase in of lost productivity). The mean caregiver burden (mea-
Zarit burden score reduced caregivers’ QoL by 0.17 sured by the ZBI) was 16.95 on a scale of 0–48, with
Opoku-Boateng et al. BMC Health Services Research 2017, 17(Suppl 2):697 Page 46 of 87
Table 7 Multiple linear regression results of total caregiver 49% of caregivers reporting high burden. Overall, care-
burden (ZBI) givers reported low QoL (28.2 out of 100).
Covariates β 95% CI The estimated mean monthly direct cost per month
Age in years −0.01 (−0.07, 0.05) (US$30) represented about 18% of the total cost in-
Sex volved in caregiving. This proportion is lower than
that obtained by Addo et al. [25] (21%) in their study
Female Ref
on the cost of caregiving for mental illness (not
Male −0.42 (−1.89, 1.05) schizophrenia specifically as in the current study) in
Marital status: Ghana. Our finding is also lower than that found by
Single Ref Zhai et al. [3] (33%) in their study on the cost of
Married −0.09 (−1.64, 1.46) caregiving for schizophrenia in China. Although Zhai
Proximity et al. [3]’s study was specific to schizophrenia, differ-
ences in the study contexts may have accounted for
Close Ref
the differences in the proportion accounted for by
Far 0.90 (−0.78, 2.58) direct costs, implying that access to treatments and
Education study contexts could influence the proportion of
None Ref overall costs of caring for vulnerable populations. The
Primary/junior high school 0.10 (−2.34, 2.55) relative contribution of direct costs to the total
Secondary −4.43 (−2.92, 2.06) economic costs further implies that caregivers bear
significant indirect costs which must be taken into ac-
Tertiary −1.01 (−3.86, 1.84)
count in efforts to ameliorate the economic burden of
Employment status** mental health on families.
None Ref Similar to Knock et al. [26], the current study’s re-
Private 3.47 (0.74, 6.19) sults demonstrate that caregivers of people with
Public 6.09 (3.10, 9.07) schizophrenia bear levels of other types of psycho-
Self-employed 1.79 (−0.30, 3.88) logical and social burdens that are difficult to quan-
tify. The current study found that caregivers of
Student/apprentice 0.17 (−4.37, 4.71)
schizophrenics reported lower QoL compared to pre-
Direct cost 0.01 (−0.001, 0.01) vious studies [16, 27] conducted in Chile and France.
Indirect cost 0.002 (−0.002, 0.01) Factors such as caregiver burden and depression were
Duration of care given 0.002 (−0.01, 0.01) associated with lower QoL, whereas support from
Daily care given time in hours −0.10 (−0.32, 0.13) other family members and education were associated
Other family with higher QoL. Sex was also significantly associated
with QoL, with males reporting higher QoL than
No Ref
females. Further, 61% of caregivers who reported
Yes 0.42 (−1.03, 1.87) higher burden were females and 39% were males.
Anxiety** Ohaeri [28] argues that caregiving roles are primarily
Normal Ref undertaken by female family members who bear
Moderate 2.82 (1.05, 4.59) much of the burden (e.g., psychological effect) of care
Severe 3.80 (1.02, 6.59) in Nigeria. Similarly, Papastavrou et al. [29] found
that females were more burdened than males and
Depression***
experienced chronic stress because of caregiving
Normal Ref responsibilities for schizophrenics in Cyprus. Sex
Moderate 3.24 (1.54, 4.94) differences in the burden of care have also been re-
Severe 8.25 (4.71, 11.80) ported in studies examining elderly care [30]. Such
Stress*** findings reflect broader societal realities and require
Normal Ref more in-depth investigation.
There are limitations to the study that are worth
Moderate 5.08 (2.91, 7.24)
noting. First, it was not possible to determine whether
Severe 7.70 (4.70, 10.70) the caregiver received any assistance, in cash or in
R2 51.9 kind from other members of the family or the patient;
β is the estimated effect of covariate; R2 is the adjusted coefficient off thus, the cost burden may not be entirely attributed
determination; CI is confidence interval; ref. is the reference category
*p < 0.05, **p < 0.01, ***p < 0.001 to the main caregiver. Second, it was not possible to
conduct more nuanced analysis to assess possible
Opoku-Boateng et al. BMC Health Services Research 2017, 17(Suppl 2):697 Page 47 of 87
Table 8 Multivariate multiple regression results for covariates of quality of life domains
Domains measuring quality of life
Physical Psychological Social Environment Joint effect
Covariates β (95% CI) p-value β (95% CI) p-value β (95% CI) p-value β (95% CI) p-value F-test statistic p-value
−0.14 −0.09 −0.06 −0.05
Age in years (−0.22, −0.06) (−0.18, 0.002) (−0.17, 0.04) (−0.13, 0.04) 3.19 0.0133*
Sex *** ** *** *
Female Ref Ref Ref Ref
Male 4.08(2.15, 6.01) 2.93(0.68, 5.17) 5.66(3.09, 8.22) 2.22(0.16, 4.28) 6.14 0.0001***
Marital status: ***
Single Ref Ref Ref Ref
Married 0.45(−1.59, 2.48) −1.73(−4.10, 0.63) −5.68(−8.38, −2.98) −2.18(−4.35, −0.02) 5.08 0.0005***
Proximity *** * * **
Close Ref Ref Ref Ref
Far −5.88(−8.10, −3.66) −3.01(−5.59, −0.43) −3.57(−6.52, −0.63) −3.66(−6.02, −1.30) 6.68 0.0001***
Education *** *** * ***
None Ref Ref Ref Ref
Primary/junior high school 2.20(−1.00, 5.40) 3.59(−0.13, 7.31) 6.79(2.54, 11.05) 5.24(1.83, 8.65)
Secondary 3.02(−0.24, 6.29) 5.12(1.33, 8.91) 5.23(0.90, 9.57) 6.53(3.06, 10.01)
Tertiary 7.05(3.32, 10.79) 11.04(6.69, 15.38) 9.81(4.84, 14.77) 14.55(10.57, 18.53) 12.03 0.0001***
Employment status
None Ref Ref Ref Ref
Private 2.41(−1.18, 6.01) 0.07(−4.11, 4.25) −1.61(−6.39, 3.16) 1.49(−2.35, 5.32)
Public 1.12(−2.86, 5.11) −1.79(−6.42, 2.84) −5.02(−10.32, 0.27) −1.26(−5.50,2.99)
Self-employed 2.81(0.06, 5.56) 1.57(−1.63, 4.76) −0.90(−4.55, 2.75) 2.86(−0.07, 5.79)
Student/apprentice −2.91(−8.85, 3.03) −1.61(−8.51, 5.30) −5.36(−13.25, 2.53) 1.51(−4.82, 7.84) 1.38 0.2412
Direct cost −0.003(−0.01, 0.005) −0.005(−0.014, 0.004) −0.0008(−0.01, 0.01) −0.003(−0.01, 0.01) 0.35 0.8468
Indirect cost 0.0001(−0.005, 0.005) −0.007(−0.01, −0.001) −0.001(−0.01, 0.01) −0.004(−0.01, 0.001) 1.46 0.2151
*** ***
Zarit −0.25(−0.38, −0.13) −0.27(−0.42, −0.12) −0.06(−0.23, 0.11) −0.10(−0.23, 0.04) 5.30 0.0004***
Duration of care given −0.003(−0.013, 0.007) −0.004(−0.02, 0.01) 0.01(−0.01, 0.02) 0.002(−0.01, 0.01) 0.85 0.4931
Daily care given time in hrs 0.14(−0.16, 0.44) 0.17(−0.18, 0.52) 0.002(−0.40, 0.40) 0.27(−0.05, 0.59) 0.63 0.6409
Other family * **
No Ref Ref Ref Ref
Opoku-Boateng et al. BMC Health Services Research 2017, 17(Suppl 2):697 Page 48 of 87
Table 8 Multivariate multiple regression results for covariates of quality of life domains (Continued)
Domains measuring quality of life
Physical Psychological Social Environment Joint effect
Covariates β (95% CI) p-value β (95% CI) p-value β (95% CI) p-value β (95% CI) p-value F-test statistic p-value
Yes 2.38(0.47, 4.29) 2.90(0.68, 5.12) −2.33(−4.87, 0.20) 0.56(−1.48, 2.59) 6.24 0.0001***
Anxiety **
Normal Ref
Moderate 0.13(−2.22, 2.47) 1.29(−1.43, 4.01) 0.07(−3.04, 3.18) 0.74(−1.75, 3.24)
Severe −2.80(−6.48, 0.87) 5.12(0.84, 9.39) −0.18(−5.06, 4.70) 3.60(−0.31, 7.52) 4.06 0.0030**
Depression *** *** *** ***
Normal Ref Ref Ref Ref
Moderate −2.73(−5.00, −0.47) −6.02(−8.66, −3.39) −3.30(−6.31, −0.29) −5.11(−7.52, −2.70)
Severe −8.39(−13.15, −3.64) −17.39(−22.92, −11.87) −12.29(−18.60, −5.97) −16.40(−21.47, −11.34) 19.20 0.0001***
Stress **
Normal Ref Ref Ref Ref
Moderate 2.46(−0.45, 5.36) 3.04(−0.34, 6.41) 0.30(−3.56, 4.15) 2.34(−0.76, 5.44)
Severe −2.44(−6.48, 1.60) −1.26(−5.96, 3.44) −10.43 (−15.80, −5.06) −1.33(−5.64, 2.988) 3.73 0.0054**
R2 68.5 69.8 64.3 67.2
β is the estimated effect of covariate, R2 is the adjusted coefficient off determination, CI is confidence interval; ref. is the reference category, Phy (Physical domain), Psy (Psychological domain), Soc (Social domain), Env
(Environmental domain)
* p < 0.05, ** p < 0.01, *** p < 0.001
Opoku-Boateng et al. BMC Health Services Research 2017, 17(Suppl 2):697 Page 49 of 87
Table 9 Multiple regression results for covariates of overall Table 9 Multiple regression results for covariates of overall
quality of life quality of life (Continued)
Covariates Overall quality of life score Covariates Overall quality of life score
β 95% CI β 95% CI
Age in years −0.08 (−0.15, −0.1) Severe −3.87 (−7.48, −0.25)
Sex*** R2 70.1
Female Ref β is the estimated effect of covariate, R2 is the adjusted coefficient off
determination, CI is confidence interval, ref. is the reference category
Male 3.72 (1.99, 5.45) *p < 0.05, **p < 0.01, ***p < 0.001
Marital status*
Single Ref underlying reasons (e.g., coping styles) for some of
Married −2.29 (−4.11, −0.47) the differences observed. Nonetheless, the current
Proximity*** study adds to the sparse literature on the economic
and social costs of mental disorders in sub-Saharan
Close Ref
Africa.
Far −4.03 (−6.01, −2.05)
Education*** Conclusions
None Ref The study findings highlight the significant burden of
Primary/junior high 4.46 (1.59, 7.32) caring for people with schizophrenia on family care-
school givers in Ghana. As the country considers deinstitu-
Secondary 4.98 (2.06, 7.89) tionalizing mental health care, it is important that
Tertiary 10.61 (7.27,13.96) measures to alleviate the direct costs on caregivers
are taken into account. Currently, the National Health
Employment status
Insurance Scheme (NHIS) does not cover services for
None Ref
schizophrenia or other mental disorder, deepening the
Private 0.59 (−2.63, 3.81) burden of families. One possible avenue to alleviate
Public −1.74 (−5.30, 1.83) the costs incurred by caregivers of people with mental
Self-employed 1.59 (−0.87, 4.05) illness would be to extend social protection programs,
Student/apprentice −2.09 (−7.41, 3.22) such as the Livelihood Empowerment Against Poverty
− − (LEAP) program (which provides cash transfers toDirect cost 0.003 ( 0.01,0.004)
vulnerable groups), to caregivers to cushion the ef-
Indirect cost −0.003 (−0.01, 0.001)
fects of the shocks of the direct costs. Study findings
Zarit −0.17 (−0.28, −0.06) also underscore the significant non-quantifiable bur-
Duration of care given 0.0002 (−0.01, 0.01) dens, such as emotional stress, that are borne by
Daily Care given time 0.15 (−0.12, 0.41) caregivers and that affect their quality of life. The
in hours quality of life of caregivers of patients with mental
Other family disorders should be considered in health policies re-
No Ref lated to mental illness.
Yes 0.88 (−0.83, 2.58) Abbreviations
Anxiety DALYs: Disability-adjusted life-years; DASS: Depression anxiety stress scale;
GHS: Ghanaian Cedi; ICD-10: International Classification of Disease, 10th
Normal Ref edition; LEAP: Livelihood empowerment against poverty; LMICs: Low and
Moderate 0.56 (−1.54, 2.65) middle income countries; NHIS: National Health Insurance Scheme;
PPS: Probability proportionate to size; QoL: Quality of Life; UK: United
Severe 1.43 (−1.85, 4.72) Kingdom; US$: United States Dollar; WHO: World Health Organisation;
Depression*** WHOQOL: WHO Quality of Life; WHOQOL-BREF: abridged WHO Quality of
Life; YLD: Years lived with disability; ZBI: Zarit Burden Interview
Normal Ref
Moderate −4.29 (−6.32, −2.26) Acknowledgements
The authors acknowledge the following for their support in implementing
Severe −13.62 (−17.87, −9.37) the study: Joseph Osafo and Gloria Odue, both at the Psychology
Department, University of Ghana; the Mental Health Authority of Ghana; and
Stress** the management and staff of the study facilities. The authors also thank all
Normal Ref the study participants.
Moderate 2.03 (−0.57, 4.63) Funding
This research was funded by an African Doctoral Dissertation Research
Fellowship (ADDRF) re-entry grant award to JN and IAK offered by the
Opoku-Boateng et al. BMC Health Services Research 2017, 17(Suppl 2):697 Page 50 of 87
African Population and Health Research Center (APHRC) in partnership with 9. Bevans MF, Sternberg EM. Caregiving burden, stress, and health
the International Development Research Centre (IDRC). Publication was effects among family caregivers of adult cancer patients. JAMA. 2012;
funded by the International Development Research Center (Grant Number 307(4):398–403.
107508–001) and the John D. and Catherine T. MacArthur Foundation (Grant 10. Dwyer JW, Lee GR, Jankowski TB. Reciprocity, elder satisfaction, and
Number 14–107495-000-INP). caregiver stress and burden: the exchange of aid in the family caregiving
relationship. J Marriage Fam. 1994;56(1):35–43.
Availability of data and materials 11. Knapp M, Mangalore R, Simon J. The global costs of schizophrenia.
Data supporting the findings of the study will be made available upon Schizophr Bull. 2004;30(2):279–93.
request. 12. Kreisman DE, Joy VD. Family response to the mental illness of a relative: a
review of the literature. Schizophr Bull. 1974;10:34–57.
About this supplement 13. Robison J, Fortinsky R, Kleppinger A, Shugrue N, Porter M. A broader view of
This article has been published as part of BMC Health Services Research family caregiving: effects of caregiving and caregiver conditions on
Volume 17 Supplement 2, 2017: Research for health systems strengthening depressive symptoms, health, work, and social isolation. J Gerontol B
in Africa: studies by fellows of the African Doctoral Dissertation Research Psychol Sci Soc Sci. 2009;64(6):788–98.
Fellowship (ADDRF) program. The full contents of the supplement are 14. Salama RAA, El-Soud FAA. Caregiver burden from caring for impaired
available online at https://bmchealthservres.biomedcentral.com/articles/ elderly: a cross sectional study in rural lower Egypt. Ital J Public Health.
supplements/volume-17-supplement-2. 2012;9(4). doi: http://dx.doi.org/10.2427/8662.
15. Yusuf AJ, Nuhu FT, Akinbiyi A. Caregiver burden among relatives of patients
Authors’ contributions with schizophrenia in Katsina, Nigeria. S Afr J Psychiatr. 2009;15(2).
JN and IAK conceived the study. YNOB, GCA, SAA, SD, DD, YT, MA 16. Margetic BA, Jakovljevic M, Furjan Z, Margetic B, Marsanic VB. Quality of life
contributed to design of the study and data collection.. YNOB led the of key caregivers of schizophrenia patients and association with kinship.
collection of data. YNOB, JN, IAK, DD, SAA, GCA, MA contributed to analysis Cent Eur J Public Health. 2013;21(4):220–3.
and interpretation of results. YNOB and JN drafted the manuscript. The 17. Rawat P. Poster #M258 Quality of life and depression among caregivers of
manuscript was reviewed by all the authors for intellectual content. The final schizophrenics. Schizophr Res. 2014;153(Supplement 1):S283–S85.
version of the manuscript was approved by all authors. 18. GSS. Population and housing census 2010: summary report of final results.
Ghana Statistical Service: Accra; 2012.
Ethics approval and consent to participate 19. Cochran WG. Sampling techniques. 2nd ed. New York: John Whiley and
Ethical approval for the study was granted by the Ghana Health Service Sons Inc; 1963.
Ethical Review Committee. 20. Bédard M, Molloy DW, Squire L, Dubois S, Lever JA, O'Donnell M. The zarit
burden interview: a new short version and screening version. Gerontologist.
Consent for publication 2001;41(5):652–7.
Not applicable 21. The WHOQOL Group. Whoqol-Bref: introduction, administration, scoring and
generic gersion of the assessment. Programme on Mental Health. 1996;16:1-12.
Competing interests 22. Bland JM, Altman DG. Statistics notes: Cronbach’s alpha. BMJ. 1997;
The authors declare that they have no competing interests. 314(7080):572.
23. MacKenzie SB, Podsakoff PM, Podsakoff NP. Construct measurement and
validation procedures in MIS and behavioral research: integrating new and
Publisher’s Note existing techniques. MIS Q. 2011;35(2):293–334.
Springer Nature remains neutral with regard to jurisdictional claims in published 24. Nunnally JC, Bernstein I. Psychometric therapy. New York: McGraw-Hill; 1994.
maps and institutional affiliations. 25. Addo R, Nonvignon J, Aikins M. Household costs of mental health care in
Ghana. J Ment Health Policy Econ. 2013;16(4):151–9.
Author details
1 2 26. Knock J, Kline E, Schiffman J, Maynard A, Reeves G. Burdens and difficultiesNational Health Insurance Authority, Accra, Ghana. Department of experienced by caregivers of children and adolescents with schizophrenia-
Pharmacy Practice and Clinical Pharmacy, School of Pharmacy, College of spectrum disorders: a qualitative study. Early Interv Psychiatry. 2011;5(4):349–54.
Health Sciences, University of Ghana, Legon, Ghana. 3Department of Health 27. Boyer L, Caqueo-Urízar A, Richieri R, Lancon C, Gutiérrez-Maldonado J,
Policy, Planning and Management, School of Public Health, College of Health Auquier P. Quality of life among caregivers of patients with schizophrenia: a
Sciences, University of Ghana, Legon, Ghana. 4Department of Biostatistics, cross-cultural comparison of Chilean and French families. BMC Fam Pract.
School of Public Health, College of Health Sciences, University of Ghana, 2012;13(1):42.
Legon, Ghana. 5Pantang Psychiatric Hospital, Accra, Ghana. 6College of 28. Ohaeri JU. Caregiver burden and psychotic patients’ perception of social support
Global Public Health, New York University, New York, NY, USA. in a Nigerian setting. Soc Psychiatry Psychiatr Epidemiol. 2001;36(2):86–93.
29. Papastavrou E, Charalambous A, Tsangari H, Karayiannis G. The cost of caring:
Published: 4 December 2017 the relative with schizophrenia. Scand J Caring Sci. 2010;24(4):817–23.
30. Nortey ST, Aryeetey GC, Aikins M, Amendah D, Nonvignon J. Economic
References burden of family caregiving for elderly population in southern Ghana: the
1. Vigo D, Thornicroft G, Atun R. Estimating the true global burden of mental case of a peri-urban district. Int J Equity Health. 2017;16(1):16.
illness. Lancet Psychiatry. 2016;3(2):171–8.
2. WHO. Burden of mental and behavioural disorders. The world health report
2001-mental health: new understanding, new hope. Geneva: World Health
Organisation; 2001. Submit your next manuscript to BioMed Central 
3. Zhai J, Guo X, Chen M, Zhao J, Su Z. An investigation of economic and we will help you at every step:
costs of schizophrenia in two areas of China. Int J Ment Health Syst.
2013;7(1):26. •  We accept pre-submission inquiries 
4. WHO. Mental disorders. Geneva: Switzerland World Health Organisation; 2016. •  Our selector tool helps you to find the most relevant journal
5. WHO: WHO-AIMS report on mental health system in Ghana. Geneva: World •  We provide round the clock customer support 
Health Organization; 2011.
6. WHO. Integrating mental health into primary care: a global perspective. •  Convenient online submission
Geneva: World Health Organization; 2008. •  Thorough peer review
7. Sariah AE, Outwater AH, Malima KI. Risk and protective factors for relapse •  Inclusion in PubMed and all major indexing services 
among individuals with schizophrenia: a qualitative study in Dar es Salaam,
Tanzania. BMC Psychiatry. 2014;14(1):240. •  Maximum visibility for your research
8. Basheer S, Niazi RS, Minhas FA, Najam N. Depression and anxiety in the
caregivers of mentally ill patients. J Pak Psychiatr Soc. 2005;2(1):27. Submit your manuscript at
www.biomedcentral.com/submit
View publication stats