Onwujekwe OE et al. Journal of the International AIDS Society 2016, 19:20588
http://www.jiasociety.org/index.php/jias/article/view/20588 | http://dx.doi.org/10.7448/IAS.19.1.20588
Research article
Examining geographic and socio-economic differences
in outpatient and inpatient consumer expenditures
for treating HIV/AIDS in Nigeria
Obinna E Onwujekwe§,1,2,3, Ogochukwu Ibe*,3, Kwasi Torpey*,4, Stephanie Dada*,4, Benjamin Uzochukwu*,2,3,5 and
Olusola Sanwo*,4
§Corresponding author: Obinna E Onwujekwe, Department of Pharmacology and Therapeutics, College of Medicine, University of Nigeria Enugu Campus, 400001
Enugu, Nigeria. Tel: 234 803 700 7771. (obinna.onwujekwe@unn.edu.ng)
*These authors have contributed equally to the work.
Abstract
Introduction: The expenditures on treatment of HIV/AIDS to households were examined to quantify the magnitude of the
economic burden of HIV/AIDS to different population groups in Nigeria. The information will also provide a basis for increased
action towards a reduction of the economic burden on many households when accessing antiretroviral therapy (ART).
Methods: A household survey was administered in three states, Adamawa, Akwa Ibom and Anambra, from the South-East,
North-East and South-South zones of Nigeria, respectively. A pretested interviewer-administered questionnaire was used to
collect data from a minimum sample of 1200 people living with HIV/AIDS (PLHIV). Data were collected on the medical and non-
medical expenditures that patients incurred to treat HIV/AIDS for their last treatment episode within three months of the
interview date. The expenditures were for outpatient visits (OPV) and inpatient stays (IPS). The incidence of catastrophic health
expenditure (CHE) on ART treatment services was computed for OPV and IPS. Data were disaggregated by socio-economic status
(SES) and geographic location of the households.
Results: The average OPV expenditures incurred by patients per OPV for HIV/AIDS treatment was US$6.1 with variations across
SES and urban-rural residence. More than 95% of the surveyed households spent money on transportation to a treatment
facility and over 70% spent money on food for OPV. For medical expenditures, the urbanites paid more than rural dwellers.
Many patients incurred CHE during outpatient and inpatient visits. Compared to urban dwellers, rural dwellers incurred more
CHE for outpatient (p0.02) and inpatient visits (p0.002).
Conclusions: Treatment expenditures were quite high, inequitable and catastrophic in some instances, hence further
jeopardizing the welfare of the households and the PLHIV. Strategically locating fully functional treatment centres to make
them more accessible to PLHIV will largely reduce expenditures for travel and the need for food during visits. Additionally,
financial risk-protection mechanisms such as treatment vouchers, reimbursement and health insurance that will significantly
reduce the expenditures borne by PLHIV and their households in seeking ART should be implemented.
Keywords: HIV/AIDS; ART services; inequity; economic burden; catastrophic health expenditure.
Received 8 July 2015; Revised 22 December 2015; Accepted 6 January 2016; Published 1 February 2016
Copyright: – 2016 Onwujekwe OE et al; licensee International AIDS Society. This is an Open Access article distributed under the terms of the Creative Commons
Attribution 3.0 Unported (CC BY 3.0) License (http://creativecommons.org/licenses/by/3.0/), which permits unrestricted use, distribution, and reproduction in any
medium, provided the original work is properly cited.
Introduction In Nigeria, ARV drugs are free of charge to patients at
An estimated 3.1 million people are living with HIV/AIDS in designated health facilities, and the provision of free drugs has
Nigeria, and the epidemiological distribution indicates sig- apparently improved access to ARVs across the country [4].
nificant diversity across the country’s geographic settings [1]. However, because HIV/AIDS increases vulnerability to other
In 2008, the national sero-prevalence rate of HIV/AIDS was illnesses, patients often incur other expenditures, including
4.6% with a corresponding decline to 4.1 and 3.4% in 2010 and payments for opportunistic infections (OIs) and non-ARVdrugs
2013, respectively [1,2]. As with preceding years, the 2013 as well as non-routine tests, medical consultations, transpor-
report shows that the prevalence was higher among wealthier tation, food and hospital stays, that are seldom covered by
Nigerians (3.7%) than among poorer Nigerians (2.9%). There any risk-pooling mechanism or government programme [5].
were similar findings when the rural dwellers (3.6%) were Non-ARV drugs include antibiotics, antivirals, blood tonics,
compared with the urban dwellers (3.2%) [2]. Based on and drugs for other OIs. Non-routine tests include x-rays,
projected HIV estimates, 220,394 new HIV infections occurred haemoglobin level and ultrasound.
in 2013, a total of 210,031 died from AIDS-related causes and Some studies have found that expenditures on treat-
an estimated 1,476,741 required antiretroviral (ARV) drugs in ment for those infected with HIV are potentially catastro-
2013 [3]. phic, even within existing free ARV treatment programmes.
1
Onwujekwe OE et al. Journal of the International AIDS Society 2016, 19:20588
http://www.jiasociety.org/index.php/jias/article/view/20588 | http://dx.doi.org/10.7448/IAS.19.1.20588
These expenditures include substantial costs incurred for non- matic actions that will decrease the economic burden of
ARV drugs, non-routine tests, medical consultations and HIV/AIDS. The findings of this study are expected to high-
hospital stays, which match or sometimes exceed the costs light potential areas for intervention that will ensure
of the ARV drugs alone [68]. There is evidence that universal financial risk protection in access to holistic HIV/
expenditures incurred by households with people living with AIDS treatment services.
HIV/AIDS (PLHIV) on non-ARV-related costs, which are often
borne through out-of-pocket payments, very often discourage Methods
people from using services or cause them to postpone health Study setting
checks [9]. The high level of direct and indirect costs could The study was undertaken in three states selected from three
lead to catastrophic health expenditures (CHEs) [10]. geopolitical zones in Nigeria. The three states (Adamawa
The evidence on the incidence of economic burden, from the North-East zone, Akwa Ibom from the South-South
especially occurrence of CHE on access to treatment for zone and Anambra from the South-East zone) were chosen to
HIV/AIDS in the context of out-of-pocket spending, is still obtain an approximately nationally representative view and
limited in many low- and middle-income countries, including to enable the estimation of the incidence of CHE in different
Nigeria. Only a few studies have focused on the equity geopolitical zones and geographic places of residence in
dimension of catastrophic health spending [1115]. CHE Nigeria. In each state, an urban and rural local government
occurs when out-of-pocket payments for healthcare exceed area (LGA) were selected. In 2009 the prevalence of HIV was
the estimated threshold share of household expenditure, 3.8, 10.9 and 8.7% in Adamawa, Akwa Ibom and Anambra
over which a household is forced to sacrifice other basic states, respectively. The 2006 census put the estimated
needs, sell assets, incur debt or be impoverished [16,17]. population of the three states at 3,178,950; 3,902,051; and
The impact of CHE on households and individuals are well 4,177,828 for Adamawa, Akwa Ibom and Anambra states,
documented in literature [6,12,13,1821]; apart from redu- respectively.
cing utilization of health services, CHE can also lead to poverty ARV therapy (ART) was fully subsidized in the three study
or re-ranking of socio-economic status (SES) [22]. states by the government and development partners. How-
The existence of CHE signifies the failure of a health system ever, patients still paid for their laboratory investigations and
to protect its citizens from financial consequences of health- any incident expenditures on co-morbidity. In Adamawa,
care [23]. Therefore, protection against CHE is considered some facilities also received fully subsidized treatment of
a role that government should strive to perform [23,24]. OIs, whereas others charged a fee. In Akwa Ibom State, all
Out-of-pocket payments are still the major payment mecha- patients were routinely charged a fee for treatment of OIs,
nism for healthcare in Nigeria, as in many less developed whereas treatment for OIs was fully subsidized in Anambra
and some developing countries. Evidence shows a positive State. Patients bore the costs of co-morbidities. Co-morbidities
correlation between the percentage of households caught in are incident illnesses that are not necessarily due to an
catastrophic payments and out-of-pocket-payments [2529]. individual being HIV positive. The common illnesses consi-
In addition, although CHE can occur for both the rich and dered in this study were malaria and other febrile illnesses.
poor, the consequences are often more disastrous for the OIs are those conditions that are likely due to diminished
poor, whose resources are limited [20]. immunity of the individual suffering from HIV/AIDS.
As a means of characterizing the level of economic burden
Study design
of diseases, several thresholds for measuring CHE have been
A descriptive cross-sectional household survey was under-
proposed by different researchers in different settings [14],
taken from June to September 2013. A minimum sample
and the CHE thresholds have ranged from 5 to 40% of
size of 1200 was calculated using a power of 80 and 95%
non-food expenditure [11]. However, it is important to use
confidence interval, assuming a maximum catastrophic in-
thresholds that are context-specific, since the levels of wealth
cidence level of 10%. However, in order to cover for refusals
across different geographic regions differ. Some authors used
and incomplete data, the sample size was increased by 20% of
a threshold of 40% of capacity to pay, which was defined as
the minimum calculated sample size. The target respondents
‘‘income after subsistence needs are met,’’ which in practice
were patients 18 years and above, living with HIV/AIDS.
amounts to income minus food expenditure [11]. With more
The patients were identified through support groups of the
than two-thirds of Nigerians living below the US$1.25-a-
Association of PLHIV to avoid undue exposure of the HIV status
day poverty head count ratio [30], any payment for health
of respondents, which could arise in a typical household
services in the country may be catastrophic and present a
survey. All eligible patients were included in the study, but
barrier to access [31].
the data on expenditures were from those patients who had
This paper provides new information on the level of medi-
been on ART and sought treatment within the three months
cal and non-medical expenditures incurred on outpatient preceding the survey. Informed consent was sought before
visits (OPV) and inpatient admissions by patients in seeking determining patient’s eligibility to participate in the survey.
treatment for HIV/AIDS. The paper also provides information
and the resultant incidence of CHE from such expenditures Data collection
to patients from different socio-economic groups and geo- Trained data collectors administered a pretested question-
graphic locations. The information will increase the under- naire to a sample of eligible respondents with HIV/AIDS.
standing of the level of expenditures borne by HIV/AIDS Patients were interviewed at home or at a location of
patients and will be important to guide policy and program- their choice. Information was obtained on the most recent
2
Onwujekwe OE et al. Journal of the International AIDS Society 2016, 19:20588
http://www.jiasociety.org/index.php/jias/article/view/20588 | http://dx.doi.org/10.7448/IAS.19.1.20588
treatment sought by patients for HIV/AIDS within the three food expenditure [11]. Other thresholds have been used
months preceding the survey. The questionnaire contained in- in the literature [15]; however we explored two scenarios,
formation on the demographic details of respondents, treat- which were monthly health expenditure as a share of
ment seeking for HIV/AIDS (including outpatient and inpatient monthly non-food expenditure greater than 40% and greater
visits) and associated expenditures. The categories of expen- than 10% for inpatient and OPVs. These two thresholds were
ditures included both medical expenditures and non-medical adopted in consideration of the high poverty levels in the
expenditures. The medical expenditures included expenses context of the study, where more than 70% live on less than
for laboratory tests, drugs, payment for hospital cards/ $1/day, and for the findings to be comparable to those in the
registration and payments made before patient could be literature. All expenditures are presented in US dollars using
seen by a provider (consultation). The non-medical expendi- a 2014 exchange rate of US$1 to 160 Nigerian naira.
tures were expenditures such as transport, food and caregiver
Ethical considerations
and accommodation expenses, where applicable. Information
All project staff completed the FHI 360 online ethics training
was also collected on weekly household food and non-food
before undertaking the surveys. Initial consent was obtained
expenditures [5,14,32,33] and ownership of key household
from the association of PLHIV in each location, and individual
assets identical to those in the National Demographic and
written consent was obtained before interviews were con-
Health Survey conducted in Nigeria in 2008 [34] Annex 1.
ducted, for those who volunteered to participate in the
Data analysis survey. Interviews were carried out discreetly to ensure
Data were analyzed from the patient’s perspective for the minimal exposure of the respondents.
whole sample. The main variables analyzed were patients’
demographic characteristics, medical and non-medical ex- Results
penditures for outpatient and inpatient visits, and treatment- Respondent characteristics
seeking behaviour. The components (sub-items) of medical Data analysis was based on data from 1409 respondents
expenditures were registration; consultation; and tests and with complete information. Table 1 shows that more than
drugs. Similarly, the components of non-medical expendi- two-thirds (74%) of the respondents were females and
tures were transport; food; accommodation; caregiver and 49.6% were monogamously married. The mean age of the
others. The frequency distributions of categorical variables respondents was 37. Overall, about 91% had some form of
were calculated and the means calculated for non-categorical education. The average number of years spent in schooling
variables. Average costs were computed on the whole was 10.6. The major source of income for a majority (45.2%)
sample, since everybody incurred an expenditure in either was petty trading. About half of the respondents (53%) were
one or all the expenditure components. The KruskalWallis resident in rural areas. The average weekly expenditures
non-parametric test, which reports a chi-square statistic, was and the per capita weekly expenditures on food were 6556.3
used to compare differences in means. naira (US$41) and 1514.6 naira (US$9.5), respectively.
For estimating the SES, an SES index was developed using The average monthly non-food expenditure was 10,926 naira
principal component analysis (PCA). The input into the PCA (US$68.3).
was information on households’ ownership of key assets, Number of people that spent money on different items
such as a car, electricity, radio, television, phone, fan, electric for OPV by SES and geographic place of residence
iron and so on, and per capita weekly household expendi- Analyses of expenditures were limited to 1392 respondents
tures on food. The index was used to divide the individuals who had sought treatment for HIV/AIDS in the three months
into five SES groups (quintiles), namely Q1 (poorest), Q2 preceding the survey. It was found that 99% of respondents
(second), Q3 (third), Q4 (fourth) and Q5 (wealthiest). The chi- sought treatment in the three months preceding the survey.
square for trend analysis was undertaken for all disaggrega- The most recent OPV was in the month directly preceding the
tion of key dependent variables by SES quintiles. In addition, survey for a majority of respondents across the three states
the equity ratio (Q1:Q5) was computed. (x2 119.6, p0.00). More than two-thirds (73%) had been
Data were disaggregated by SES and geographic location; on ART for more than a year but frequency of check-up
mean and standard deviation were reported for the main varied (x2 46.4, p0.00). Overall, there were 35 admissions
outcome measures, which were the patient’s medical and (less than 3% of respondents) within the three months
non-medical expenditures, and incidence of CHE were preceding the survey, with an average of one admission in
obtained and compared across urban-rural and SES groups. the period. Most admissions (51%) were in public facilities
Significance was assessed at 5% (p-valueB0.05). The most (x2 1.1, p0.90), and more than two-thirds (77%) of
recent outpatient or inpatient visit in the three months admissions were for the treatment of OIs (x2 1.3, p0.51)
preceding the survey was used. Thus reported expendi- (not shown in table).
tures were for one inpatient or OPV, as the case may be. For themost recent OPV, the treatment receivedwasmostly
The three-month recall period was deemed appropriate to routine ARV drugs; for example over 90% of respondents in
capture substantial inpatient events, given that they occur each state received routine treatment (x2 14.0, p0.00).
less frequently than OPVs. However, 79.4% of respondents in Anambra received treat-
The method proposed by Xu et al. [11] was used to esti- ment for other OIs, compared to 25 and 45% of respondents
mate CHE, which by definition, refers to treatment-related from Akwa Ibom and Adamawa states, respectively (x2 282.5,
expenditures exceeding 40% of a household’s monthly non- p0.00).
3
Onwujekwe OE et al. Journal of the International AIDS Society 2016, 19:20588
http://www.jiasociety.org/index.php/jias/article/view/20588 | http://dx.doi.org/10.7448/IAS.19.1.20588
Table 1. Demographic characteristics of respondents Table 2 shows that majority of patients reported spending
money on transport (97.8%) and food (72.8%). The propor-
Variable n (%) N 1409 tions of respondents that spent money on other items are
shown in Table 2.
Gender, n (%)
Female 1048 (74.3) Average treatment expenditures for OPVs and inpatient
Status in household, n (%) stays
Male head 316 (22.4) Table 3 shows that medical expenditures and non-medical
expenditures contributed to 34.3 and 65.7%, respectively, of
Female head 376 (26.7)
total expenditures for OPVs. Similarly, the table shows that
Son/daughter 228 (16.1)
medical expenditures and non-medical expenditures contrib-
Wife 489 (34.7)
uted to 38.7 and 61.3%, respectively, of total expenditures
Marital status, n (%)
for inpatient stays (IPS). Table 3 also shows that for OPVs
Married monogamous 699 (49.6) the main drivers for medical expenditures were the expen-
Married polygamous 38 (2.7) ditures on drugs, at 271 naira (US$1.70), and tests, at 37
Single 262 (18.5) naira (US$0.22). Average non-medical expenditure was 647
Divorced 58 (4.1) naira (US$4.02). The most significant non-medical expendi-
Separated 75 (5.3) ture component was transport, with an average of 489 naira
Widowed 275 (19.5) (US$3.05), followed by average expenditure on food, which
Average number of all household residents, was 144 naira (US$0.90). For IPS drug expenditure was the
mean (SD) 5 (2.7) single most significant medical expenditure component, with
 an overall average of 4693 naira (US$29.30).Adults 18 3 (1.8)
13 to 17 years 1 (1.0) Differences in average outpatient expenditures by
12 and less 1.4 (1.4) SES and geographic location
Age of respondents, mean (SD) 37 (9.9) The result in Table 4 is based on OPVs, since there were
Attended school, n (%) 1279 (90.8) few people that incurred inpatient expenditures. There were
Highest level of education, n (%) significant urban-rural differences in average medical and
None 7 (0.5) non-medical expenditures per visit for HIV/AIDS treatment
Primary education 400 (31.2) (pB0.05), and across SES and urban-rural residence, non-
JSS 149 (11.6) medical expenditures (US$4.00) were about twice the medical
expenditures (US$2.10). For medical expenditures the urban
SSCE 496 (38.8)
dwellers paid more (US$2.20 compared to US$2.00), whereas
Tertiary 111 (8.7)
the reverse was true for non-medical expenditures. The
NCE 87 (6.8)
expenditures on treatment of co-morbidities was more than
Other 30 (2.3)
50% of total medical expenditures with significant urban-rural
Years spent schooling, mean (SD) 10.6 (3.8) differences (p 0.00). The expenditures on food (US$0.90)
Major source of income, n (%) and transport (US$3.00) were much higher than the other
Unemployed 182 (12.9) categories of non-medical expenditures, and they contributed
Farmer 128 (9.0) to 15 and 50% of total outpatient expenditures on HIV/AIDS
Artisan/petty trader 638 (45.2) treatment, respectively. For transport, the rural dwellers paid
Government worker 130 (9.3) more (US$3.40, p0.00).
Self-employed 160 (11.3) Across SES, there were no significant differences in
Employed in private sector 104 (7.3) expenditure categories except for those on food (p0.01),
Other 66 (4.6) where those in the higher SES spent more. The rural-urban
Place of residence equity ratios show that medical expenditures were pro-poor
Urban 664 (47.1) (0.91), whereas for non-medical costs the rural dwellers paid
more (1.11). The ratios for all expenditure categories across
Rural 745 (52.9)
SES suggest that the poor pay less compared to the rich
Weekly food expenditure, mean (SD) 6556.3 (3650.8)
except for caregiver expenditures. Few respondents paid for
Per capita weekly food expenditure, mean (SD) 1515 (992.0)
consultation and the result is not significantly different across
Average monthly non-food expenditure: mean (SD) 10,926 (10022.0)
SES and the rural-urban divide (Table 4). The median for
SES distribution of respondents
total expenditures on OPVs was US$3.90 and the range was
Quintile 1 (poorest) 282 (20.04) $0 to $100.10.
Quintile 2 (second) 281 (19.97)
Quintile 3 (third) 282 (20.04) SES and geographic differences in level of CHEs
Quintile 4 (fourth) 281 (19.97) Table 5 shows that overall, at a threshold of 40%, about
Quintile 5 (wealthiest) 281 (19.97) 8 and 94% of patients incurred catastrophic expenditures
on outpatient and inpatient visits, respectively. At a thres-
SES, socio-economic status. hold of 10%, the corresponding numbers were 40 and
4
Onwujekwe OE et al. Journal of the International AIDS Society 2016, 19:20588
http://www.jiasociety.org/index.php/jias/article/view/20588 | http://dx.doi.org/10.7448/IAS.19.1.20588
5
Table 2. Proportion of people that spent money on different items for outpatient visit, by SES and geographic place of residence
Outpatient visits
Urban % Rural % Chi-square Q1 poorest % Q2  second % Q3  third % Q4  fourth % Q5 wealthiest % Chi-square for trend Total n (%)
Variables n 654 n 738 (p-value) n 279 n 275 n277 n279 n280 (p-value) n 1392
Registration 20.8 42.9 76.8 (0.00) 36.9 28.8 30.9 32.4 33.0 0.34 (0.55) 452 (32.5)
Consultation 1.5 0.9 0.54 (0.45) 1.1 1.4 1.1 1.1 1.4 0.00 (0.93) 14 (1.0)
Tests 3.4 3.2 0.01 (0.90) 4.3 6.0 1.8 2.1 2.5 5.16 (0.02) 46 (3.3)
Drugs 13.3 6.7 16.59 (0.00) 9.1 7.1 8.2 13.5 11.0 2.89 (0.08) 136 (9.8)
Transport 97.1 97.4 3.39 (0.06) 96.8 95.0 95.7 97.2 98.2 0.52 (0.47) 1361 (96.6)
Food 71.8 70.9 0.26 (0.60) 69.5 75.1 69.9 73.3 69.0 0.34 (0.55) 1005 (71.3)
Accommodation 1.1 0  0.7 0.7 0.4 0 0.4 1.70 (0.19) 7 (.5)
Caregiver 1.2 2.3 1.72 (0.19) 4.3 1.1 1.4 1.8 0.4 8.61 (0.00) 25 (1.8)
Co-morbidities 6.6 11.1 7.74 (0.00) 8.1 9.6 9.6 10.3 6.8 0.35 (0.54) 124 (8.9)
Other expenditures 2.3 1.2 1.76 (0.18) 0.7 1.8 0.35 1.1 4.3 6.66 (0.00) 24 (1.7)
SES, socio-economic status.
Table 3. Average treatment expenditures for outpatient visits and inpatient stays
Outpatient visits Mean Inpatient stays Mean Percentage of expenditure components Percentage of expenditure components
Variable total in naira (US$) total in naira (US$) of outpatient visits in total expenditure of in-stays in total expenditure
Medical expenditures 338 ($2.11) 5712 ($35.70) 34.3 38.7
Registration 28 ($0.17) 181 ($1.13) 2.7 1.2
Consultation 3 ($0.02) 183 ($1.14) 0.3 1.2
Tests 37 ($0.22) 656 ($4.10) 3.8 4.4
Drugs 271 ($1.70) 4693 ($29.30) 27.5 31.8
Non-medical expenditures 647 ($4.02) 9055.4 ($56.59) 65.7 61.3
Transport 489 ($3.05) 833 ($5.20) 49.6 5.6
Food 144 ($0.90) 2291 ($14.32) 14.9 15.5
Accommodation 2 ($0.01) 2183 ($13.64) 0.2 14.8
Caregiver 4 ($0.02) 806 ($5.03) 0.4 5.5
Other expenditures 5 ($0.03) 2943 ($18.39) 0.5 19.9
Average total expenditure 985 ($6.10) 14,767.0 ($92.30) 100 100
Note: US$1 160 Nigerian naira.
Onwujekwe OE et al. Journal of the International AIDS Society 2016, 19:20588
http://www.jiasociety.org/index.php/jias/article/view/20588 | http://dx.doi.org/10.7448/IAS.19.1.20588
100%, respectively. At the 40% threshold, rural dwellers
(p0.00) and the poorest quintile (p0.00) were more
likely to incur catastrophic expenditures for OPVs; there were
no significant differences for inpatient visits. At the 10% thres-
hold for OPVs, again rural dwellers (p0.00) and the poorest
(p0.00) were significantly more likely to incur catastrophic
costs. For inpatient visits at the 10% threshold, everyone
incurred catastrophic expenditure. There were no significant
differences in catastrophic expenditures by state (Table 5).
Discussion
The findings show that the public provision of free drugs is
not enough to eliminate the high and sometimes inequitable
economic burden of HIV/AIDS on households. Many adjunct
treatments and expenditures on diagnostics that are not
covered by free ART programmes still predispose patients to
incurring CHE. In addition, some non-medical expenditures
that are incurred by patients, such as transport and feeding
during treatment visits, are also substantial contributors to
the high level of economic burden of HIV/AIDS to house-
holds. Inpatient visits particularly led to a high level of CHE.
Irrespective of the free provision of ARV drugs at several
facilities across Nigeria, patients still need to pay out-of-
pocket for other medical expenditures, such as OIs and co-
morbidities, as well as non-medical expenditures. Other
studies have similarly reported that treatment seeking could
still remain unaffordable despite the availability of free ARVs
due to other care components associated with HIV/AIDS [6,7].
In Ghana, total outpatient expenditure on ART was found to
be up to US$55 depending on how far the patients had to
travel to get to the nearest ART centre and how long they had
to wait at the ART facility [7]. Similar findings were reported
by other studies elsewhere [8,35,36]. Rosen et al. found that
91% of patients paid for transport to attend ART clinics and
60% of patients purchased non-prescription medicines or
special food at considerable cost [35]. It was also found in
Kenya that patients made an average payment of US$7 for
ART [36]. These findings were similar to the high incidence
and levels of expenditures on transport that was found in
our study.
It was revealing to find that non-medical expenditures
were much higher than medical expenditures and that the
two most significant expenditure components were food and
transport to treatment facilities. Hence, it does appear that
policy interventions such as decentralizing treatment centres
by bringing them nearer to people that can significantly
lower these expenditures to PLHIV and their households.
Such measures will subsequently improve health seeking for
PLHIV and adherence to ART [37]. In the long term, they are
expected to significantly improve the welfare of the patients
and households. Moon et al. show that expenditures for
transport may pose significant barriers of access to ARVs
even where they are free [6].
The findings show that rural dwellers, who are usually
poorer than the urbanites as reflected by their lower SES,
suffered greater economic burden in accessing ART services
and treatment for other HIV/AIDS-related conditions. Overall,
there were geographic differences in medical and non-
medical expenditures. The average treatment expenditure
6
Table 4. Differences in average outpatient expenditures by SES and geographic location
Total N1392 Rural Mean Urban Mean Chi-square Q1poorest Q2second Q3third Q4fourth Q5wealthiest Chi-square Q1:Q5 ratio
Variables Mean (SD) (SD) N654 (SD) N738 (p-value) N279 N275 N277 N279 N280 (p-value) (equity ratio)
Medical expenditures 2.1 (6.6) 2.0 (6.2) 2.2 (6.9) 5.9 (0.01) 1.7 (7.2) 2.5 (7.2) 2.2 (7.9) 1.8 (4.5) 2.4 (7.7) 1.98 (0.73) 0.7
Registration 0.2 (0.4) 0.2 (0.3) 0.0 (0.4) 41.2 (0.00) 0.1 (0.3) 0.2 (0.4) 0.1 (0.3) 0.2 (0.4) 0.2 (0.4) 1.61 (0.81) 0.5
Consultation 0.02 (0.3) 0.0 (0.0) 0.0 (0.4) 0.0 (0.84) 0.0 (0.0) 0.0 (0.3) 0.0 (0.2) 0.0 (0.2) 0.0 (0.4) 0.01 (1.00) 0
Tests 0.2 (1.7) 0.2 (1.8) 0.2 (1.5) 0.0 (0.93) 0.2 (1.3) 0.4 (1.7) 0.1 (1.1) 0.1 (1.5) 0.3 (2.4) 1.12 (0.89) 0.6
ARV drugs 0.6 (2.6) 0.4 (2.4) 0.9 (2.8) 4.6 (0.00) 0.5 (2.3) 0.5 (2.3) 0.5 (2.4) 0.7 (2.6) 0.9 (3.4) 1.96 (0.74) 0.6
Co-morbidities 1.1 (5.4) 1.1 (5.1) 1.0 (5.7) 7.9 (0.00) 0.8 (3.9) 1.4 (6.2) 1.4 (7.2) 0.7 (2.5) 1.0 (6.0) 0.66 (0.95) 0.8
Non-medical expenditures 4.0 (4.1) 4.2 (4.3) 3.8 (3.9) 9.0 (0.00) 3.9 (3.7) 3.8 (3.0) 4.1 (4.6) 3.6 (3.1) 4.6 (5.7) 5.39 (0.24) 0.8
Transport 3.0 (3.8) 3.4 (4.1) 2.7 (3.3) 28.8 (0.00) 3.1 (3.4) 2.8 (2.7) 3.3 (4.4) 2.6 (2.7) 3.5 (4.9) 5.72 (0.22) 0.9
Food 0.9 (1.0) 0.8 (0.8) 0.9 (1.2) 2.2 (0.14) 0.7 (1.0) 0.9 (1.0) 0.8 (0.7) 1.0 (1.0) 1.0 (1.2) 11.99 (0.01) 0.7
Accommodation 0.0 (0.2) 0 0.0 (0.3) 0.1 (0.73) 0.0 (0.1) 0.0 (0.1) 0.0 (0.0) 0 (0) 0.0 (0.4) 0.03 (0.99) 0
Caregiver 0.02 (0.2) 0.03 (0.3) 0.0 (0.3) 0.1 (0.73) 0.1 (0.5) 0.0 (0.1) 0.0 (0.2) 0.0 (0.2) 0.0 (0.1) 0.75 (0.94) 0.1
Other expenditures 0.03 (0.5) 0.0 (0.3) 0.0 (0.6) 0.1 (0.73) 0.0 (0.1) 0.0 (0.1) 0.0 (0.0) 0.0 (0.1) 0.1 (1.0) 0.83 (0.93) 0
Total OPV expenditures 6.1 (8.0) 6.3 (8.1) 5.9 (7.9) 5.9 (0.01) 5.6 (6.6) 6.3 (8.0) 0.6 (9.6) 5.3 (5.5) 7.1 (7.1) 1.98 (0.73) 0.7
Note: The median for total expenditures on outpatient visit was $3.90 and the range was $0 to $100.10.
SES, socio-economic status; OPV, outpatient visit.
Onwujekwe OE et al. Journal of the International AIDS Society 2016, 19:20588
http://www.jiasociety.org/index.php/jias/article/view/20588 | http://dx.doi.org/10.7448/IAS.19.1.20588
Table 5. Differences in the level of catastrophic health expenditures in different population groups
Total outpatient Total inpatient expenditure Total outpatient Total inpatient expenditure
expenditure 40% of non- 40% of non-food expenditure 10% of non- 10% of non-food
food expenditure, n (%) expenditure, n (%) food expenditure, n (%) expenditure, n (%)
Combined 107 (7.7) 33 (94.3) 561 (40.3) 35 (100)
Urban-rural differences
Urban 37 (5.7) 10 (90.9) 188 (28.7) 11 (100.0)
Rural 70 (9.5) 23 (95.8) 373 (50.5) 24 (100.0)
Chi2 (p-value) 7.20 (0.00) 0.33 (0.53) 68.8 (0.00)
Socio-economic status
differences
Poorest 42 (15.0) 10 (100.0) 161 (57.7) 10 (100.0)
Second 20 (7.3) 4 (80) 126 (45.8) 5 (100)
Third 22 (7.9) 6 (100) 115 (41.5) 6 (100)
Fourth 11 (3.9) 6 (100) 87 (31.1) 6 (100)
Wealthiest 12 (4.3) 7 (80) 72 (25.7) 8 (100)
Chi2 (p-value) 31.5 (0.00) 3.91 (0.34) 73.4 (0.00)
Differences by state
Adamawa 396 (92.1) 17 (80.9) 175 (40.7) 20 (95.2)
Akwa Ibom 426 (93.6) 4 (100.0) 173 (38.0) 4 (100.0)
Anambra 463 (91.3) 10 (100.0) 213 (42.0) 10 (100.0)
Chi2 (p-value) 1.84 (0.39) 1.63 (0.44) 1.17 (0.55) 0.68 (0.71)
for an OPV was higher for the rural dwellers, but it was not the states. However, it is possible that since all the states
clear why there was no difference in some of the expendi- are guided by the national guidelines of ART provision, they
tures across the three states. It is possible that the states are are not expected to be programmatically different at the
not programmatically different from each other. However, macro level.
this line of inquiry could be a subject for future studies. Our findings confirm that hospitalization greatly increases
It was found that the households’ expenditures on HIV/AIDS the possibility of incurring catastrophic medical expenditures
were to a large extent catastrophic and that the magnitude among households. Although the proportion of people hav-
was significantly higher for the rural dwellers and those from ing inpatient visits was low, almost all expenditures were
lower SES. The magnitude of CHE in the study was less than catastrophic regardless of SES group and place of residence
reported in a previous study, which found quite high levels when the threshold was lowered to 10% and more than half
of CHE for patients regardless of geographic locations, sex or at 40%. The finding that OIs account for most hospitalizations
SES [5], possibly due to increases in the number of ART is in line with studies elsewhere that have demonstrated the
facilities in the region. However, it should be noted that more role of OIs in HIV/AIDS morbidity and mortality [38]. In less-
than 70% of Nigerians live below the poverty line and they developed countries where there are limited risk-pooling
usually spend all their money on food; any other expenses are mechanisms to cover individuals from medical expenditures,
potentially catastrophic. Hence, the use of 40% or even 10% hospitalizations are associated with high levels of expendi-
non-food expenditure thresholds may be misleading in the tures [39], especially when they occur at private facilities
context of the study. Non-availability of money when a person where patients receive few or no subsidies for treatment.
needs healthcare is a major barrier to accessing healthcare A limitation of the study was the restriction of survey
services in Nigeria [4]. respondents to those who belonged to a support group
The greater differences in expenditures and incidence of for PLHIV. This implies that the sample eliminated people
CHE by geographic location of the households rather than by that did not belong to the groups and was potentially self-
SES implies that rural-urban inequity is a more significant selected. However, recruiting respondents from only sup-
problem than SES differences in Nigeria within the context of port groups was done to avoid exposing the HIV status of
treatment for HIV/AIDS. This inequity is addressable, but the respondents if they were selected from a random household
intervention may have to adopt a multisectoral approach to survey. Another limitation was that data on frequency of
address the multifaceted problems impoverishing PLHIV and visits were not collected, since some patients visit clinics
their households. Such an approach may involve the devel- often within one month, and such multiple visits may lead
opment and implementation of some income-generating to underestimation of CHE if not captured. Moreover,
interventions. Policy options could be explored to support there was no qualitative component that could have been
the provision of a full subsidy for payment of OIs. It is not used for deeper exploration of some of the issues and no
easy to understand the non-difference in CHE between external validation of expenditure data from other sources.
7
Onwujekwe OE et al. Journal of the International AIDS Society 2016, 19:20588
http://www.jiasociety.org/index.php/jias/article/view/20588 | http://dx.doi.org/10.7448/IAS.19.1.20588
These could be researched in future studies. Furthermore, 2. Federal Ministry of Health (2013). National HIV & AIDS and Reproductive
future studies could explore the impact of different funding Health Survey, 2012 (NARHS Plus). Abuja, Nigeria: Federal Ministry of Health.
3. Federal Republic of Nigeria (2014). Global AIDS Response; Country progress
arrangements on catastrophic expenditures. Finally, relying
report. Abuja, Nigeria: National Agency for the Control of AIDS (NACA).
on patient’s recall of expenditures on services may have 4. National Population Commission (NPC) [Nigeria] and ICF Macro (2014).
affected the accuracy of information and therefore repre- Nigeria Demographic and Health Survey 2013. Abuja, Nigeria: ICF International,
sents a potential source of bias in this study [40]. An alter- Rockville, Maryland, USA.
5. Onwujekwe O, Dike N, Chukwuka C, Uzochukwu B, Onyedum C, Onoka C,
native approach could be to explore the differences in cost of
et al. Examining catastrophic costs and benefit incidence of subsidized
treatment for patients on ART versus those not on ART. antiretroviral treatment (ART) programme in south-east Nigeria. Health Policy.
2009;90:2239.
6. Moon S, Van Leemput L, Durier N, Jambert E, Dahmane A, Jie Y, et al. Out-of-
Conclusions pocket costs of AIDS care in China: are free antiretroviral drugs enough? AIDS
All in all, households’ expenditures for their members that Care. 2008;20(8):98494.
are living with HIV/AIDS (PLHIV) to receive treatment for 7. Apanga S, Punguyire D, Adjei G. Estimating the cost to rural ambulating HIV/
the disease was quite high, inequitable and catastrophic in AIDS patients on Highly Active Antiretroviral Therapy (HAART) in rural Ghana: a
pilot study. Pan Afr Med J. 2012;12(21).
some instances, hence further jeopardizing the welfare of
8. Beauliere A, Toure S, Alexandre PK, Kone K, Pouhe A, Kouadio B, et al.
the household as a whole, as well as the PLHIV. The fact that The financial burden of morbidity in HIV-infected adults on antiretroviral
a greater share of treatment expenditures were from the therapy in Cote d’Ivoire. PLoS One. 2010;5(6):e11213.
transport and food expenditure categories suggests that 9. WHO (2010). The World Health Report 2010. Health system financing: the
locating treatment centres closer to PLHIV and deploying path to universal coverage. Geneva: World Health Organization.
10. Van Doorslaer E, O’Donnell O, Rannan-Eliya R, Somanathan A, Adhikari S,
more health personnel to the treatment centres will reduce
Garg C, et al. Catastrophic payments for health care in Asia. Health Econ.
travel expenditures, improve adherence to treatment and 2007;16(11):115984.
lessen the need to spend a lot of time during care visits, 11. Xu K, Evans D, Kawabata K, Zeramdini R, Klavus J, Murray C. Household cata-
which necessitates expenditures on feeding at the ART strophic health expenditure: a multicountry analysis. Lancet. 2003;362:1117.
facilities. In addition, financial risk-protection mechanisms 12. Chuma J, Maina T. Catastrophic health care spending and impoverishment
in Kenya. BMC Health Serv Res. 2012;12:413.
should be implemented that will significantly eliminate the
13. Bredenkamp C, Mendola M, Gragnolati M. Catastrophic and impoverishing
expenditures borne by PLHIV and their households in order to effects of health expenditure: new evidence from the Western Balkans. Health
receive ART services. In particular, subsidies of expenditures Policy Plan. 2011;26(4):34956.
on transport in the form of vouchers or reimbursement 14. Onwujekwe O, Hanson K, Uzochukwu B. Examining inequities in incidence
systems are good financial protection mechanisms. Enhancing of catastrophic health expenditures on different healthcare services and health
facilities in Nigeria. PLoS One. 2012;7(7):e40811.
the income of PLHIV and their households can reduce the 15. Onoka C, Onwujekwe O, Hanson K, Uzochukwu B. Examining catastrophic
incidence of CHE, since CHE increased as SES decreased. health expenditures at variable thresholds using household consumption
Finally, universal financial risk protection within the sphere expenditure diaries. Trop Med Int Health. 2011;16(10):133441.
of universal health coverage should be the ultimate goal of 16. Xu K, Evans D, Carrin G, Aguilar-Rivera A, Musgrove P, Evans T. Protecting
household from catastrophic health spending. Health Aff. 2007;26(4):97283.
HIV/AIDS treatment services, so as to protect all households
17. Wagstaff A, van Doorslaer E. Catastrophe and impoverishment in paying
against CHEs. for health care: with applications to Vietnam 19931998. Health Econ. 2003;
12(11):92134.
Authors’ affiliations 18. Boing AC, Bertoldi AD, Barros AJ, Posenato LG, Peres KG. Socioeconomic
1Department of Pharmacology and Therapeutics, College of Medicine, inequality in catastrophic health expenditure in Brazil. Rev Saude Publica.
University of Nigeria Enugu Campus, Enugu, Nigeria; 2Department of Health 2014;48(4):63241.
Administration and Management, College of Medicine, University of Nigeria 19. Daneshkohan A, Karami M, Najafi F, Matin BK. Household catastrophic
Enugu Campus, Enugu, Nigeria; 3Health Policy Research Group, College of health expenditure. Iran J Public Health. 2011;40(1):949.
Medicine, University of Nigeria Enugu Campus, Enugu, Nigeria; 4Department of 20. Kavosi Z, Rashidian A, Pourreza A, Majdzadeh R, Pourmalek F, Hosseinpour
Health Systems, FHI 360, Abuja, Nigeria; 5Department of Community Medicine, AR, et al. Inequality in household catastrophic health care expenditure in a
College of Medicine, University of Nigeria Enugu Campus, Enugu, Nigeria low-income society of Iran. Health Policy Plan. 2012;27(7):61323.
21. Lee WY, Shaw I. The impact of out-of-pocket payments on health care
Competing interests inequity: the case of national health insurance in South Korea. Int J Env Res
The authors declare that they have no competing interests Public Health. 2014;11(7):730418.
22. Ichoku H, Fonta W. The distributive effect of health care financing in
Authors’ contributions Nigeria. Quebec, Canada: University of Laval; 2006.
The study was conceived by OO, OI and KT. OO, OI, SD, BU and OS participated 23. WHO (2000). The World Health Report 2000. Health systems: improving
in data collection. All the authors participated in data analysis. OO drafted the performance. Geneva: World Health Organization.
paper. All the authors contributed to revising the paper until it was submitted 24. Abolhallaje M, Hasani S, Bastani P, Ramezanian M, Kazemian M.
for publication. Determinants of catastrophic health expenditure in Iran. Iran J Public Health.
2013;42(Supple1):15560.
Acknowledgements 25. Amaya Lara JL, Ruiz Gomez F. Determining factors of catastrophic health
This study is made possible by the generous support of the American people spending in Bogota, Colombia. Int J Health Care Finance Econ. 2011;11(2):
through the US Agency for International Development (USAID), with technical 83100.
support from Family Health International (FHI 360). The contents are the 26. Brinda EM, Andres RA, Enemark U. Correlates of out-of-pocket and
responsibility of the authors and do not necessarily reflect the views of FHI catastrophic health expenditures in Tanzania: results from a national house-
360, USAID or the United States government. We thank Nkem Chineme for her hold survey. BMC Int Health Hum Rights. 2014;14:5.
help in editing the paper. 27. Li X, Shen JJ, Lu J, Wang Y, Sun M, Li C, et al. Household catastrophic
medical expenses in eastern China: determinants and policy implications. BMC
References Health Serv Res. 2013;13:506.
1. Federal Government of Nigeria (2009). National policy on HIV/AIDS. Abuja, 28. Li Y, Wu Q, Xu L, Legge D, Hao Y, Gao L, et al. Factors affecting catastrophic
Nigeria: Federal Government of Nigeria. health expenditure and impoverishment from medical expenses in China:
8
Onwujekwe OE et al. Journal of the International AIDS Society 2016, 19:20588
http://www.jiasociety.org/index.php/jias/article/view/20588 | http://dx.doi.org/10.7448/IAS.19.1.20588
policy implications of universal health insurance. Bull World Health Org. 35. Rosen S, Ketlhapile M, Sanne I, DeSilva MB. Cost to patients of
2012;90(9):66471. obtaining treatment for HIV/AIDS in South Africa. S Afr Med J. 2007;97(7):
29. Gotsadze G, Zoidze A, Rukhadze N. Household catastrophic health 5249.
expenditure: evidence from Georgia and its policy implications. BMC Health 36. Zachariah RI, Harries AD, Manzi M, Gomani P, Teck R, Phillips M, et al.
Serv Res. 2009;9:69. Acceptance of anti-retroviral therapy among patients infected with HIV and
30. World Bank. Indicators by country 2011. 2011. Available from: http://data. tuberculosis in rural Malawi is low and associated with cost of transport. PLoS
worldbank.org/indicator/SI.POV.DDAY/countries One. 2006;27(1):e121.
31. Lagarde M, Palmer N. The impact of user fees on health service utilization 37. Sansom SL, Anthony MN, Garland WH, Squires KE,Witt MD, Kovacs Andrea
in low and middle income countries: how strong is the evidence. Bull World A, et al. The costs of HIV antiretroviral therapy adherence programs and impact
Health Org. 2008;86(11):83948. on health care utilization. AIDS Patient Care STDs. 2008;22(2):1318.
32. Onwujekwe O, Hanson K, Uzochukwu B, Ichoku H, Ikeh E, Onwughalu B. 38. Xiao J, Gao G, Li Y, Zhang W, Tian Y, Huang Y, et al. Spectrums of
Are malaria treatment expenditures catastrophic to different socio-economic opportunistic infections and malignancies in HIV-infected patients in tertiary
and geographic groups and how do they cope with payment in southeast care hospital, China. PLoS One. 2013;8(10):e75915.
Nigeria? Trop Med Int Health. 2010;15(1):1825. 39. Mukherjee S, Haddad S, Narayana D. Social class related inequalities in
33. Onwujekwe O, Hanson K, Ichoku H, Uzochukwu B. Financing incidence household health expenditure and economic burden: evidence from Kerala,
analysis of household out-of-pocket spending for healthcare: getting more south India. Int J Equity Health. 2011;10:1.
health for money in Nigeria? Int J Health Plan Manage. 2013;29(2):e174e185. 40. Onwujekwe O, Chukwuogo O, Ezeoke U, Uzochukwu B, Eze S. Asking
34. National Population Commission (NPC) [Nigeria] and ICF Macro (2008). people directly about preferred health seeking behavior yields invalid res-
Nigeria Demographic and Health Survey 2008. Abuja, Nigeria: ICF International, ponse: an experiment in South east Nigeria. J Public Health. 2011;33(1):
Rockville, Maryland, USA. 93100.
Annex 1. Sample questions used to collect expenditure data
E8. Please can you tell me about the cost you incurred in your last outpatient visit for anti-retroviral treatment?
(Read out the different categories and record the cost categories the respondent tells you. If category applies, insert amount in naira.
If category of cost does not apply leave blank; if the category is not listed, use ‘‘other’’ and specify the cost.)
Category of cost Amount in naira and kobo
E8a Transport going to facility TrFacE8H [ ]
E8b Transport return from facility TrRtnE8H [ ]
E8c Cost of health card/registration RegE8H [ ]
E8d Consultation fee ConsE8H [ ]
E8e Cost of tests: TstE8eH [ ]
 Lab. [ ]
 X-ray [ ]
 Other [ ]
[ ________________________ ]
E8f Cost of drugs DrgE8fH [ ]
E8g Cost of any food or drink FdcE8gH [ ]
E8h Cost of accommodation AccE8hH [ ]
E8i Caregiver cost (if any) CareE8iH [ ]
E8j Other costs (1) OthE8jH [ ]
[ ________________________ ]
E8k Other costs (2) OthE8kH [ ]
[ ________________________ ]
E8l Other costs (3) OthE8lH [ ]
[ ________________________ ]
E8m Total cost (interviewer add up) TotE8mH [________________ ]
9