Factors associated with Antiretroviral Treatment Adherence among Adult Patients in WolaitaSoddo Hospital, Wolaita Zone, Southern Ethiopia

Abstract

Background: HIV a major challenge to the whole world since the last quarter of 20th century but has become a
member of a manageable chronic disease since the advent of ARV drugs. The non-adherence to Antiretroviral Therapy is an
emerging major challenge to AIDS care. Objective: To assess factors associated with adherence among AIDS patients
receiving Antiretroviral Therapy at Wolaita Soddo Hospital, Southern Ethiopia. Method: A cross sectional study was
carried out at Wolaita Soddo Hospital from April 15 to May 15, 2012 E.C. The collected data were entered in Epi Info
version3.5.3. Data analyses were done using SPSS for windows version 16.0. Descriptive analysis was used to describe the
data. Binary logistic regression analysis was used to measure the association between the dependent variable and
independent variables calculating odds ratio and its 95% confidence interval (CI). Statistical significance was set at α. ≤
0.05. Multivariable analyses were applied to identify the relative effect of explanatory variables on the dependent variable.
Results: Three hundred and fifty seven HIV/ AIDS patients were involved in the study. Multi-method adherence assessment
consisting of self-report, monthly dispense schedule and dietary requirements ,were used to measure adherence and the
average adherence rate was 74.4%.Multivariate logistic regression analysis showed that, with whom a subject lives
(AOR=4.943,1,(2.168-11.270)), depression(AOR=2.221,1,(1.093-4.515)), and having inadequate diet to take with ART (AOR=2.229,1,(1.034-4.807)), were independent predictors of dose adherence. Conclusions: In conclusion living with a partner, having no depression and having no food scarcity to take with ART have association with adherence to ART regimens.

Full text

Science Journal of Public Health
2014; 2(2): 69-77
Published online February 28, 2014 (http://www.sciencepublishinggroup.com/j/sjph)
doi: 10.11648/j.sjph.20140202.15
Factors associated with antiretroviral treatment
adherence among adult patients in Wolaita Soddo
Hospital, Wolaita Zone, Southern Ethiopia
Amsalu Alagaw
1
, Wanzahun Godana
1, *
, Mohammed Taha
2
, Tariku Dejene
2
1
Department of Public Health, College of Medicine and Health Sciences, Arba Minch University, Arba Minch, Ethiopia
2
School of Public Health, College of Health Sciences, Jimma University, Jimma, Ethiopia
Email address:
amsalu.alagaw@yahoo.com (A. Alagaw), wanzanati2011@gmail.com (W. Godana)
To cite this article:
Amsalu Alagaw, Wanzahun Godana, Mohammed Taha, Tariku Dejene. Factors associated with Antiretroviral Treatment Adherence
among Adult Patients in Wolaita Soddo Hospital, Wolaita Zone, Southern Ethiopia. Science Journal of Public Health.
Vol. 2, No. 2, 2014, pp. 69-77. doi: 10.11648/j.sjph.20140202.15
Abstract:
Background: HIV a major challenge to the whole world since the last quarter of 20
th
century but has become a
member of a manageable chronic disease since the advent of ARV drugs. The non-adherence to Antiretroviral Therapy is an
emerging major challenge to AIDS care. Objective: To assess factors associated with adherence among AIDS patients
receiving Antiretroviral Therapy at Wolaita Soddo Hospital, Southern Ethiopia. Method: A cross sectional study was
carried out at Wolaita Soddo Hospital from April 15 to May 15, 2012 E.C. The collected data were entered in Epi Info
version3.5.3. Data analyses were done using SPSS for windows version 16.0. Descriptive analysis was used to describe the
data. Binary logistic regression analysis was used to measure the association between the dependent variable and
independent variables calculating odds ratio and its 95% confidence interval (CI). Statistical significance was set at α. ≤
0.05. Multivariable analyses were applied to identify the relative effect of explanatory variables on the dependent variable.
Results: Three hundred and fifty seven HIV/ AIDS patients were involved in the study. Multi-method adherence assessment
consisting of self-report, monthly dispense schedule and dietary requirements ,were used to measure adherence and the
average adherence rate was 74.4%.Multivariate logistic regression analysis showed that, with whom a subject lives
(AOR=4.943,1,(2.168-11.270)), depression(AOR=2.221,1,(1.093-4.515)), and having inadequate diet to take with ART
(AOR=2.229,1,(1.034-4.807)), were independent predictors of dose adherence. Conclusions: In conclusion living with a
partner, having no depression and having no food scarcity to take with ART have association with adherence to ART
regimens.
Keywords:
Antiretroviral Treatment, Adherence, HIV/AIDS, Ethiopia
1. Background
Worldwide an estimated 33 million people are living
with HIV. Since the beginning of the HIV epidemic in 1981,
25 million people have died of AIDS globally. Every day,
there are 7 400 new HIV infections, 96% of which are in
the low-and middle-income countries. Africa
disproportionately bears the burden of the HIV/AIDS
pandemic. Although only 11% of the world's population
lives in Africa, roughly 67% of those living with
HIV/AIDS are in Africa. In Africa, there were 22.4 million
people living with HIV and 1.9 million new HIV infections
in 2008(1, 2).
According to the most recent data from the Joint United
Nations Program on HIV/AIDS and National projections
estimate approximately 1.1 million Ethiopians are living
with HIV and prevalence increased slightly to 2.3 percent
by 2009(3). In 2009 alone, 43,130 new HIV patients were
started on ART. Likewise the total number of people ever
enrolled to chronic HIV care reached to 435,150 by the end
of 2009(1). With the advent of ART, HIV/AIDS is
becoming a chronic disease; therefore adherence to HIV
medication is an extremely important but complicated
process that needs dedication from patients, sympathetic
support from the society, and high level of care from health
care professionals (4).

70 Amsalu Alagaw et al.: Factors associated with Antiretroviral Treatment Adherence among Adult Patients in
Wolaita Soddo Hospital, Wolaita Zone, Southern Ethiopia
Worldwide, regardless of the illness or treatment many
people do not take their medications correctly (5). A study
in Brazil showed that the cumulative incidence of non
adherence to be 36.9% (6). Adherence among patients in
Soweto, South Africa was 88% In Cape Town; 63% of
patients maintained adherence levels of 90% (7). Consistent
factors for poor adherence include, stress, substance use,
regimen complexity, self efficacy for medication taking and
depression Social support have also been consistently
associated with decreased adherence and a patient who
doesn’t have social support is less likely to continue their
treatment with optimal requirement(8).
On aggregate, non-adherence to ART is estimated at
between 50-80% in different social and cultural settings (9).
With optimal adherence rates, studies by Lewis, Rao and
others have demonstrated that ART can suppress the viral
load to undetectable levels, boost the immune system by
increasing the number of CD-4 cells, and improve the
quality of life of PLWHAS(9). Since its introduction about
10 years ago, ART has progressed from a single therapy, to
double therapy, and finally to the current combination
therapy commonly known as Highly Active Antiretroviral
Therapy (HAART). This development was necessitated by
the desire to reduce the number of pills taken by the patient
in order to foster adherence. This was borne out of the
realization that adherence rates were inversely proportional
to the number of pills, capsules or tablets necessary for
treatment (10).
Despite patients’ understanding the consequences of non-
adherence to medication, adherence rates were sub optimal
(6). Long-term adherence interventions are needed for
durable effect, particularly in chronic diseases such as HIV.
Antiretroviral therapy lowers viral load only when
treatment regimen is fully adhered to. Human immuno
deficiency virus (HIV) poses a unique challenge due to its
rapid replication and mutation rates hence very high levels
of adherence (greater than 95%) are required to achieve
long-term suppression of viral load (11).
A study conducted in the antiretroviral therapy unit of
Jimma University Specialized Hospital on predictors of
ART adherence among HIV infected patients Self-reported
dose adherence in the study area was 94.3%. The rate
considering the combined indicator (dose, time and food)
was 75.7 %.( 12). Many studies have observed that self-
reporting over-estimates adherence, as patients may report
to be perfectly adhering when, in actual sense, they are not
(11,12, 13).
Barriers related to daily schedules: disruptions in routine
or having a chaotic schedule, finding HAART too
inconvenient or difficult to incorporate and difficulties
coordinating adherence with work, family or care giving
responsibilities at home. A patient’s behavior is the critical
link between a prescribed regimen and treatment outcomes.
Consequently, the most important factors influencing
adherence are patient related and under the patient’s control,
so attention to them is a necessary and important step in
improving adherence (14).
2. Methods and Materials
2.1. Study Area and Period
The study was conducted in Wolaita Soddo hospital,
which is one of hospitals in Wolaita soddo town from April
15/2012_May15/2012. The town is located in SNNPR
378km south of Addis Ababa. Wolaita Soddo Hospital
currently is in transition period to be Wolaita Soddo
University hospital is serving in four major clinical fields
i.e. Internal medicine, pediatrics, surgery and gynecology.
There are also some minor specialized fields like dental
care service. The Hospital serving thousands of HIV/AIDS
patients since 2005 from rural and urban areas. There were
2400 HIV AIDS patients on follow up among which 1162
started ART treatment and on follow up.
2.2. Study Design
A cross sectional study was utilized with an internal
comparison of adherence status among predictor variables.
2.3. Study Variables
2.3.1. Dependent
Adherence status to ARV treatment.
2.3.2. Independent
Sex, age, income, with whom client lives, marital status,
household size, level of education, occupation, food source,
treatment regimen and co-management of co-infections,
socio-cultural factors like avoiding people because of HIV
status, religion, ethnicity, psychiatric problems like
depression, health care facility and health care providers,
patient behavioral factors like HIV status disclosure and
substance abuse.
2.3.3. Target Population
All HIV sero positive subjects in the study area.
2.3.4. Source Population
All adult HIV and AIDS patients who started
antiretroviral treatment and follow up in the Hospital
during the study period.
2.3.5. Study Population
All adult HIV sero-positive subjects started on ARV
treatment and on follow up who were recruited in the study
during the study period.
2.3.6. Sample Size Determination
According to 2010 Ethiopia national progress report a
significant number of patients (28%) had been lost to
follow up as of December 2008. Patients alive and on
treatment has significantly declined as the duration of
treatment increased using two population proportion and
assumptions sample size was 361.
2.3.7. Sampling Procedure
Patients on follow up at ART treatment and care clinic
meeting the inclusion criteria were selected by using

Science Journal of Public Health 2014; 2(2): 69-77 71
patient’s register as a sampling frame and adequate sample
size was collected by using systematic random sampling
method until the required sample size obtained. The first
case selected by lottery method and every 3rd subject was
taken as study subject. For absent and non responding cases
the subject who came next was taken as a replacement case.
2.3.8. Inclusion Criteria
The inclusion criteria comprised of AIDS patients who
were on ART treatment for at least one month and willing
to participate in the study.
2.3.9. Exclusion Criteria
HIV/AIDS patients on regular follow up but who did not
start ART and those patients who were in critical medical
and mental illness
2.3.10. Data Collection Methods
Data were collected from the study population using a
pre- tested structured questionnaire through face to face
interviews in the local language.
2.3.11. Data Quality Assurance
Properly designed and pre-tested questionnaire was used.
Interviewers and data clerks were trained and closely
supervised during data collection and entry; and double
data entry was used to ensure data quality.
2.3.12. Data Analysis
The data were entered into a pre-drafted coding sheet on
Epi info software, version 3.5.3.by two different data clerks.
Binary logistic analysis with conditional method
calculating odds ratios (OR) and 95% confidence intervals
(CI) was used to estimate the association between the
dependent variable and independent variables. Statistical
significance was set at α. ≤ 0.05. In an attempt to identify
the relative effects of explanatory variables on the outcome
variable multivariable analyses was applied.
2.4. Ethical Approval
The ethical approval and clearance for this research
study was obtained from Jimma University College of
Health Sciences Institutional Research Ethics Review
Committee. At all levels, officials were contacted and
permission from administrators was secured. All the
necessary explanation about the purpose of the study and
its procedures was explained with the assurance of
confidentiality. Both written and verbal consent from the
study participants was also secured.
3. Results
3.1. Socio-Demographic and Economic Characteristics
Out of total study subjects 54.3 %( 194) were females
and 45.7 % (163) were males. The respondents’ ages
ranged from 18 to70 years and those with in the age group
50-59 and above 60 years old were 6.4 % and 1.7%
respectively. More than 91% of the subjects were below the
age of 50 years .This finding indicated that majority of
patients were in age between 30-39 years and the mean age
of respondents was 34.7 with SD of 8.8years.
Two hundred and sixteen (60.4%) of clients were
protestant by religion followed by orthodox 34.5%, Muslim
2.8% and 1.7% catholic. Concerning ethnicity majority of
respondents (78.2%) were Wolaita, 10.9% Amhara,
3.1%Oromo, 1.7% Gurage and the remaining were other
ethnic groups.The mean family size of the study population
was 4.5 with SD of 2 persons. About 73.7% of the subjects
had less than 6 persons in their families.
Of the 300 subjects who had formal education, the
majority (42.0%) attended some primary education
(grade1-8). Thirty one percent of subjects attended
secondary level education including preparatory (grade9-
12). About eleven (11%) of clients attended higher
education while Sixteen percent of subjects were illiterate.
ART adherence was highest among those who educated to
secondary level (85.6%).
One hundred eighty nine (52.9%) of subjects were living
with their partners the remaining 47.1% living with non
partner members of the family or alone. Majority of
respondents (59.7%) were married, 20.7% were widowed,
8.7% were single, and 10.9% were divorced. Two hundred
and fifty four (71.1%) of subjects were employed, but one
hundred and three (28.9%) were unemployed. Of employed
subjects 65.2% were involved in self or private business.
Among unemployed housewives have higher proportion
(56.3%). Ninety two percent of respondents mainly
purchased food for their consumption, 3.4 % got their food
from household farm and 3.4 % from both household farm
and by purchasing, 0.8% from NGO/welfare. Despite low
income and purchasing food for consumption most of
subjects (79.6%) were able to afford three meals in a day.
Those who could afford two and one meal in a day were
20.2% and 0.3 % respectively (Table 1)
The study subjects for income were only 285(79.83% of
the study subjects) the remaining 20.16% had no monthly
income. The median income of the study subjects was
500.00Et.birr per-month, with standard deviation of
5.28birr per month (bpm). The minimum monthly income
was 100.00bpm while maximum income was 4000.00bpm.
Table 1. Basic Socio-demographic characteristics of PLWHA involved in the study, Wolaita Soddo Hospital, South Ethiopia, May 2012
Variable Adherent (n=312) Non adherent(n=45) Total
Frequency (%) Frequency (%) Frequency (%)
Sex Male 147(41.17) 16(1.68) 163(45.66)
Female 165(46.22) 29(8.12) 194(54.34)
Age group 18-19 1(0.28) 0(0) 1(0.28)

72 Amsalu Alagaw et al.: Factors associated with Antiretroviral Treatment Adherence among Adult Patients in
Wolaita Soddo Hospital, Wolaita Zone, Southern Ethiopia
Variable Adherent (n=312) Non adherent(n=45) Total
Frequency (%) Frequency (%) Frequency (%)
20-29 78(21.85) 13(3.64) 91(25.49)
30-39 144(40.34) 21(5.88) 165(46.22)
40-49 64(17.92) 7(1.96) 71(19.89)
50-59 20(5.6) 3(0.84) 23(6.44)
60+ 5(1.4) 1 (0.28) 6(1.68)
Income n=285 <=500 132(46.32) 17(5.96) 149(52.28)
>500 117(41.05) 19(6.67) 136(47.71)
Family size <=5 226(63.3) 37(10.36) 263(73.67)
>=6 86(24.09) 8(2.24) 94(26.33)
Marital status
Married 190(53.22) 23(6.44) 213(59.66)
Single 24(6.72) 7(1.96) 31(8.68)
Divorced 35(9.8) 4(1.12) 39(10.92)
Widowed 63(17.64) 11(3.08) 74(20.72)
Live with Non Partner 132(36.97) 36(10.08) 168(47.06)
Partner 180(50.42) 9(2.52) 189(52.94)
Educational status
Illiterate 48(13.44) 9(2.52) 57(15.96)
Primary 136(38.09) 14(3.92) 150(42.01)
Secondary 95(26.61) 16(4.48) 111(31.09)
Higher 33(9.24) 6(1.68) 39(10.92)
Occupation Unemployed 91(25.49) 12(3.36) 103(28.85)
Employed 221(61.9) 33(9.24) 254(71.15)
Meal number
One 1(0.28) 0(0) 1(0.28)
Two 62 (17.36) 10(2.8) 72(20.18)
>=three 249(69.74) 35(9.80) 284(79.55)
Consumption food
source
Purchase 286(80.11) 44(12.34) 330(92.43)
Household farm 12(3.36) 0(0) 12(3.36)
Welfare/NGO 3(0.84) 0 3(0.84)
Purchase +household farm 11(3.08) 1(0.28) 12(3.36)
3.2. Treatment Related, Personal Behavior, Clinical, and
Health Service Related Characteristics
The three adherences (self report of missed doses,
monthly dispensing schedule and dietary adherences) were
assessed in the study to get a combined adherence indicator
(Table2)). Accordingly, 312 (87.4%) of the study subjects
were adherent based on self-report of missed doses (dose
adherence) in a one-week recall. Three hundred thirty two
(93%) of the study subjects were adherent to monthly
dispensing schedules. One hundred and fifty three (42.9%)
subjects had resources to follow instructions related to
dietary requirement (food adherence). Hence, the rate of
adherence in the study area based on the combined
indicators of the three adherences was 74.4% indicating
that ART adherence rate was sub-optimal (<95%).
Most common management mentioned by subjects for
inadequate diet was taking drug in empty stomach (96.1 %
of cases). But 3.4% avoid taking drugs and 0.5% by
extending the time of taking drugs.
Table 2. Percentage of different adherences in PLWHA involved in the
study, Wolaita Soddo Hospital, South Ethiopia, May 2012,
Type of adherence Number Percent
Dose adherence 312 87.4
Dispense adherence 332 93
Dietary adherence 153 42.9
Average adherence 74.4
Twenty nine percent of respondents were taking a FDC
known as Triomune 30 {Stavudine (dt4), Lamivudine (3TC)
and Nevirapine (NVP)}. Triomune 30 regimen was
recommended for respondents who were below 60 kg body
weight. Only two (0.6%) of clients were taking Fixed Dose
Combination of ARV regimen D4T (30) + 3TC + NVP (1a
(30)) known as Nevilast 40 {Stavudine (dt4), Lamivudine
(3TC) and Nevirapine (NVP)}. Nevilast 40 regimen was
recommended for respondents who were above 60 kg body
weight. Currently the recommendation is to shift patients
on Nevilast 40 to triammune30. Nineteen percent of clients
were taking TDF+3TC+EFV, 16.5% were taking
AZT+3TC+NVP,13.2% were taking AZT+3TC
+EFV,13.2% were taking TDF+3TC+NVP,6.2% were
taking D4T+3TC+EFV. Relatively higher proportion of
patients who missed ART was among the groups on
D4T+3TC+EFV (18.2%). Fourteen (14.5%) missed TDF-
3TC-EFV, thirteen (12.5%) missed AZT-3TC- NVP).
Out of 12.6% of respondents who failed to adhere to self
reported dose, the main reason for missing a drug dose was
forgetting (48.9%). Stock finished accounted for 15.6%,
11.1% said they missed because of lack of transport cost,
being too ill and lack of food to take with drugs each
accounted for 2.2% of reasons for missing to take ARV
drugs, and other reasons (11.1%). This showed that
respondents had various reasons that made them miss
treatment doses. Majority of respondents 99.4% said that
government health facilities are the only source of ARV
drugs. Two patients (0.7%) said they got ART drugs from
friends, relatives and from private health facility in case
their ARV drugs get finished .This finding indicated that
majority of respondents were aware of where they should
get their ARV medication timely.
Most of the client (88.5%) disclosed their HIV status.

Science Journal of Public Health 2014; 2(2): 69-77 73
Non adherence among those who disclosed their status was
11.07% while among who did not disclose was 24.4%.
Three hundred forty four (96.4 %)of subjects were free
of any substance abuse but 1.7% use kcat,0.8%
alcohol,0.3% cigarette and 0.8% use the above substances
in some combination. Twelve (92.3%) of substance users
were males and only 7.7% of users was female. There was
no significant association between substance use and ART
adherence status.
Most respondents 65.8 % were undergoing treatment of
both HIV and antibiotics, 3.6% were undergoing co-
treatment of HIV and TB, 0.6% was on treatment with
appetizers and 0.3 % was being treated with painkillers. As
most patients were on prophylactic antibiotic which is once
daily dose, pill burden was not complained by study
subjects. There was no statistically significant association
between ART adherence status and pill burden (co-
treatment of other problems.
One hundred and six (29.7%) of clients had depression,
70.3% no depression. Adherence among depressed groups
was 79.24% but among non depressed groups was 90.83%.
Three hundred and fifty four (99.2%) of clients received
counseling, 0.8% said they were not counseled. All the
clients said counseling is important for ART adherence and
all the subjects satisfied that privacy was maintained during
clinic visit and during special consultation of health
professionals. One hundred and ninety nine (55.7%) of
clients have some supporter for ART adherence, 44.3%
didn’t have supporter of adherence.
Table 3. Treatment related characteristics of PLWHA involved in the study, Wolaita Soddo Hospital, South Ethiopia, May, 2012
Variables Adherent(312) Non-adherent(45) Total (357)
Frequency (%) Frequency (%) Frequency (%)
ART drug used for
Cure 23(6.44) 7(1.96) 30(8.4)
Reducing pain 162(45.37) 23(6.44) 185(51.82)
Reduce progression of HIV 73(20.45) 5(1.4) 78(21.85)
Don’t know 0(0) 1(0.28) 1(0.28)
Reducing pain and progression of HIV 54(15.12) 9(2.52) 63(17.64)
Type of ART used
D4T-3TC-NVP(30) 90(25.2) 14(3.92) 104(29.13)
D4T-3TC-NVP(40) 2(0.56) 0 2(0.56)
D4T-3TC-EFV(1b30) 18(5.04) 4(1.12) 22(6.16)
AZT-3TC-NVP(1C) 53(14.8) 6(1.68) 59(16.52)
AZT-3TC-EFV(1D) 40(11.2) 7(1.96) 47(13.17)
ABC-DDI-LPV/t(2a) 2(0.56) 0 2(0.56)
TDF-DDI-LPV/R(2c) 1(0.28) 0 1(0.28)
TDF-DDI-NFV(2d) 1(0.28) 0 1(0.28)
TDF-3TC-NVP 41(11.48) 7(1.96) 48(13.44)
TDF-3TC-EFV 62(17.37) 7(1.96) 69(19.32)
TDF-3TC-LPV/r 2(0.56) 0 2(0.56)
Reason for missing ART Toxicity 0(0) 4(1.12) 4(1.12)
Forgetting 0(0) 22(6.16) 22(6.16)
Too ill 0 1(0.28) 1(0.28)
Stock finished 0 7(1.96) 7(1.96)
Lack of transport cost 0 5(1.4) 5(1.4)
Lack of food to take with drugs 0 1(0.28) 1(0.28)
Others 0 5(1.4) 5(1.4)
Non ART drugs used by
subjects
Pain killers 0 1(0.28) 1(0.28)
Appetizers 2(0.56) 0 2(0.56)
Anti TB drugs 11(3.08) 2(0.56) 13(3.64)
Antibiotics 209(58.54) 26(7.28) 235(65.83)
Others 1(0.28) 1(0.28) 2(0.56)
Table 4. personal behavior factors
Variable Adhered Non-adhered Total
Frequency (%) Frequency (%) Frequency (%)
Substance abuse
Alcohol 2(0.56) 1(0.28) 3(0.84)
Kcat 5(1.4) 1(0.28) 6(1.68)
Cigarette 1(0.28) 0 1(0.28)
Combined 3(0.84) 0 3(0.84)
Disclosed HIV status Yes 281(78.71) 35(9.8) 316(88.52)
No 31(8.68) 10(2.8) 41(11.48)
Inadequate diet to take
with ART
Yes 171(47.9) 33(9.24) 204(57.14)
No 141(39.5) 12(33.6) 153(42.86)
How food scarcity
managed
Opinion to ART
By avoiding taking drugs 5(1.4) 2(0.56) 7(1.96)
Prolonging time of taking drugs 167(46.78) 30(8.4) 197(55.2)
Approve 308(86.27) 44(12.32) 352(98.6)
Disagree 4(1.12) 0 4(1.12)
Undecided 0 1(0.28) 1(0.28)

74 Amsalu Alagaw et al.: Factors associated with Antiretroviral Treatment Adherence among Adult Patients in
Wolaita Soddo Hospital, Wolaita Zone, Southern Ethiopia
Variable Adhered Non-adhered Total
Frequency (%) Frequency (%) Frequency (%)
Subject supported for
adherence
Yes 186(52.1) 13(3.64) 199(55.74)
No 126(35.3) 32(8.96) 158(44.26)
Depression Yes 84(23.53) 22(6.16) 106(29.7)
No 228(63.86) 23(6.44) 251(70.3)
Who supported for
adherence
Spouse 91(25.49) 4(1.12) 95(26.6)
Immediate member of family 88(24.65) 9 97(27.17)
Nurse 4(1.12) 0 4(1.12)
Doctor 2(0.56) 0 2(0.56)
Other 1(0.28) 0 1(0.28)
3.3. Bivariate and Multivariate Analysis
Non-adherent patients and those who had been adherent
were compared on key demographic, health, mental health,
and social factors. Results of bivariate logistic regression
analyses showed that from socio-demographic variables
with whom a subject lives i.e. with a partner or non partner
was statistically significantly associated with adherence
status (OR= 1.916(1.017-3.608) P<0.001. Other socio-
demographic variables with a p-value of 0.25 to be
included in multivariate analysis include sex of the subject
(OR 1.615(0.843-3.092) P=0.148), family size of the
subject (OR 1.760 (0.788-3.931), P = 0.168), and
consumption food source (OR =0.250 (0.033-1.889)
P=0.179).
In bivariate analysis of treatment related, personal
behavior, clinical, and health service related characteristics
of the study subjects were associated with ART adherence
status. Variables statistically significantly associated
adherence status include HIV status disclosure (OR
2.590(1.17-5.73) P=0.019)), depression (OR 2.596(1.375-
4.903) P=0.003)), having inadequate diet to take with ART
(OR= 2.268(1.129, 4.554) P=0.021)), and having adherence
support (OR 3.634(1.835-7.195) P=0.001)).
Variables entered in multivariate analysis were:
adherence supporter, sex of the subject, with whom a
subject lives, depression, food source of the subject,
inadequate diet to take with ART, family size, and HIV
disclosure status of the subject.
Multivariate logistic regression showed that inadequate
diet to take with ART (AOR=2.229(1.034-4.807)),
depression, and with whom a subject lives were statistically
significantly associated with ART adherence (Table 5).
Disclosure of HIV status, food source of the subject, having
or not having adherence support which were significant in
bivariate analysis were not statistically significant in
multivariate analysis which may be due to confounding
effect of other variables.
Table 5. Summary of the logistic regression analysis of the effects of socioeconomic, demographic, clinical and behavioral factors that affect ART
adherence, Wolaita Sddo Hospital, south Ethiopia, May, 2012.
Variable Missed ART in the last
week OR(95% CI)
Sex of client
No Yes Crude Adjusted
Male 147 16 1.615(0.843-3.092) 1.058(0.509-2.196)
Female 165 29
HIV disclosure No 31 10 2.590(1.17-5.73) 0.647(0.247-1.698)
Yes 281 35
Depression status No 228 23
Yes 84 22 2.596(1.375-4.903) 2.221(1.093-4.515)
Supported for adherence No 126 32 3.634(1.835-7.195) 0.523(0.233-1.171)
Yes 186 13
Subject live with Partner 182 19
Non partener 130 26 1.916(1.017-3.608) 4.943(2.168-11.27)
Food source Only purchase 286 44
Others 26 1 0.250(0.033-1.889) 4.561(0.558-37.259)
Food scarcity to take with
ART
Yes 171 33 2.268(1.129-4.554) 2.229(1.034-4.807)
No 141 12
Family size <=5 226 37
>=6 86 8 1.76(0.788-3.931) 1.046(0.436-2.507)
4. Discussions
In this study, the adherence to ART is good. The
following are associated with ART adherence. Food
scarcity to take with ART, source of food for consumption,
with whom subject lives and depression status.
In most studies, researchers assess adherence by
indicators that measure the rate of follow up of dose of the
drugs prescribed. But successful treatment with ART also
includes adhering to dietary instructions that accompany
many antiretroviral drugs (15). In this study adherence was
assessed by two indicators of dose adherence i.e. self report
of missing of the drug doses and regular following of
monthly dispense schedule. Follow up of dietary instruction
was used as additional independent type of adherence. A

Science Journal of Public Health 2014; 2(2): 69-77 75
combined indicator was made to determine the rate of
adherence in the study area. Participants' self-reports of
adherence in this study indicated sub-optimal dose
adherence (87.5%) and an acceptable degree of monthly
dispense adherence(93%), while very low adherence to
dietary instructions(42.9%) in comparison with the required
95% adherence. Subjects were more adherent to monthly
dispensing appointments this may be because they have
remainder cards for monthly appointments. But were least
adherent to dietary requirements may be because of low
socioeconomic status (median monthly income was
500.00Etbpm with SD of 5.28bpm).
As mentioned above the rate of dose adherence in the
study area was (87.5%) which is higher than that reported
in Addis Ababa (81.2% and 82.8%) (16, 17). This may
suggest that adherence in regional setting is better than
central settings and that patients of low socioeconomic
status were able to achieve excellent rates of adherence
with ART. Only one quarter of the Addis Ababa study
population reported income of less than 500 birr per month
(16) whereas more than 50% of this study population
earned less than 500 birr per month. Consistent findings
were also documented in comparable studies in resource
limited settings in the sub-Saharan Africa (18).
The overall rate of adherence in the study area based on
the combined indicators of the three adherence rates was
74.4% which is comparable to a finding in Jimma
University Specialized hospital(75.7%)(13).Adherence rate
in most developed countries, ranged from 50% to 70% (1,
18). This was lower than the finding in this study
confirming that patients in developing countries can
achieve good adherence despite limited resources. Orrell
and others also found that low socio economic status was
not a predictor of adherence for patients with fully
subsidized therapy and concluded that adherence in
developing countries has been found to be at least as good
as adherence in developed countries (19). The dose
adherence rate (87.5%) found by this study was less than
that found by Sarna and others at Mombasa Kenya, which
was greater than 95%. The finding by Sarna was for
patients on ART who were under Directly Administered
Antiretroviral Therapy (DAART) program at Mombasa
(20).
Among female respondents 85% adhered to ART while
among the male respondents 90.18% adhered to treatment
by direct interview indicating that there is slight difference
in ART adherence with higher rate in males(Table 1). This
may be because females have burden of routine daily
household activities which contributed to forgetting. Also
study done on Canadian injection drug users showed that
after adjusting for clinical characteristics as well as drug
use patterns measured longitudinally throughout follow-up,
female gender was independently associated with a lower
likelihood of being 95% adherent to ART.
Depression was statistically significantly associated with
adherence. Subjects with no depression have 2.596 times
more adherent than those with depression. In other studies,
in Kenya and South Africa, depression has been an
important predictor of adherence (19, 21). These findings
support a role for HIV/AIDS providers/counselors in
screening for depression and providing treatment when
appropriate, either directly or through collaboration with
mental health professionals.
Patients who reported to have adherence support were
3.634 times more likely to be adherent than those who do
not have adherence supporter. This finding suggests that
lack of adequate diet influenced the qualitative aspect of
ART adherence as most respondents managed their
inadequate diet by taking drugs even when they didn’t have
diet. Inadequate diet to take with ART was significantly
associated with dose adherence; similarly a study in South
Africa showed that food supplement and higher CD4 cell
count have statistical association with adherence (20).
With whom a subject lives was an independent predictor
of adherence also studies in south Africa found that
patients who were likely to be non-adherent were
unmarried or lived alone, had poor social support, were
reluctant to disclose and consumed alcohol (21).
The use of family members and peers to enhance ART
adherence has emphasized the importance of social support
in the treatment of HIV patients. Social support may
enhance adherence through encouragement, reassurance,
reinforcement, systematic cues, boastering of competence,
and motivation, or by masking the effect of stress, anxiety,
and depression (16)
Other variables like knowledge of ART benefit,
knowledge on ART, and adherence knowledge which were
significant predictors of adherence in other studies did not
show association with adherence in this study. This may be
because despite patients knowing the benefits of ART and
the importance of adherence there were other factors like
fear of stigma, low economic status that may mask the
effect of the knowledge. This finding was supported by a
study on HIV-patients in Lusaka Zambia who used to skip
treatment doses due to lack of food (18)
In this study patients have high level of acceptance
towards ART and there was low level of unacceptable
behaviors for ART adherence like substance abuse which
may contributed to an acceptable level of adherence found
in the study. Also other studies shown that patient’s
behavior is the critical link between a prescribed regimen
and treatment outcomes. Consequently, the most important
factors influencing adherence are patient related and under
the patient’s control, so attention to them is a necessary and
important step in improving adherence (14).
The limitation of the study, Recall bias- was the
possible limitation of this self reported study to minimize it
recent short duration(one week) period used. Over
estimation of adherence during self report was another
limitation and Absence of prior studies to estimate the
proportion of adherent subjects which was approximated by
national reports and figures from other studies in similar
setting.
In conclusion Level of adherence among the study

76 Amsalu Alagaw et al.: Factors associated with Antiretroviral Treatment Adherence among Adult Patients in
Wolaita Soddo Hospital, Wolaita Zone, Southern Ethiopia
participants was good. This could be explained by the fact
that majority of the patients were positive about the aspects
of care related to the ART clinic, interpersonal relationship
with their care providers and their treatment plan plus being
knowledgeable about HIV/AIDS.
On top of their infection which predispose to different
infections and low productivity, the study subjects have low
income with a need to involve them in capacity building
activities through collaboration with different stakeholders.
Psychosocial factors were the predominant factors
affecting ART adherence. The findings emphasized the
importance of multiple periodic assessments of adherence
problems especially reasons for forgetting to take ART
drugs. Timely detection of non-adherence behaviors and
appropriate monitoring of patients' difficulties with
HAART could potentially help patients to maintain
adherence and therefore improve the treatment outcome.
Finally the results suggested that psychosocial and medical
interventions aimed at increasing adherence of HAART-
treated patients should integrate the dynamic dimensions of
adherence behaviors.
Authors' Contributions
AA conceived of and designed the study, supervised data
collection, analyzed the data, drafted the paper and
approved the final version. MT and TD contributed to the
conception, designing, data analysis, drafting and approval
of the manuscript. WG participated in data analysis,
interpretation of findings, contributed to the drafting of the
paper, writing and revising the manuscript critically for
intellectual content and updated the manuscript. All authors
read and approved the manuscript.
References
[1] Annual Performance Report of HSDP-III, Ministry of
Health, Federal Democratic Republic of Ethiopia, EFY2001,
(2008/2009)
[2] Health and Health Related Indicators, Federal Ministry of
Health, Planning and Programming Department: 2007/8
[3] AIDS Epidemic Update. UNAIDS, World Health
Organization. Geneva. WHO 2003
[4] Samuel Sendagala ,Factors affecting the adherence to
antiretroviral therapy by hiv positive patients treated in a
community based hiv/aids care programme in rural Uganda
A case of tororo district ,University Of South
Africa ,November 2010
[5] Adherence to HIV Treatment. Geneva, Switzerland. World
Health Organization 2004
[6] Bonolo, Palmira de Fae, Casar, Cibele C b, Acurio,
Fransisco A ac, et al. Non-adherence among patients
initiating antiretroviral therapy: a challenge for health
professionals in Brazil. JAIDS. 2006; 19 Sppl4: S5-S13.
[7] Nachega J. B, Stein D M, Lahnan D.A, et al. Adherence to
Antiretroviral Therapy in HIVInfected Adults in Soweto,
South Africa, AIDS Research and Human Retroviruses.
2004;20(10):1053-1056.
[8] Adriana A, Paola TM, Rita M et al. Correlates and
predictors of adherence to Highly Active Antiretroviral
Therapy. JAIDS 2002;31 supplement 3:s123-s127.
[9] Eliud Wekesa. ART adherence in resource poor settings in
sub-Saharan Africa: a multidisciplinary review
[10] HIV / AIDS IN ETHIOPIA -AN EPIDEMIOLOGICAL
SYNTHESIS.Ethiopia HIV/AIDS Prevention & Control
Office (HAPCO) and Global HIV/AIDS Monitoring and
Evaluation Team (GAMET) april 2008.available at:
www.worldbank/aids > publications
[11] Nyambura Anthony Wanjohi. Factors That Influence Non–
Adherence To Antiretroviral Therapy Among Hiv And Aids
Patients In Central Province, Kenya April 2009.
[12] Alemayehu Amberbir, Kifle Woldemichael, Sofonias
Getachew, Belaineh Girma,and Kebede Deribe
.
Predictors
of adherence to antiretroviral therapy among HIV-infected
persons: a prospective study in Southwest Ethiopia,july
2008
[13] Yared Mekonnen, Eduard Sanders, Tsehaynesh Messele,
Dawit Wolday,Wendelien Dorigo-Zestma, Ab Schaap,
Walelgne Mekonnen, Hailu Meless1,Wude Mihret, Arnaud
Fontanet, Roel A. Coutinho and Nicole H.T.M. Dukers,
Prevalence and Incidence of, and Risk Factors for,HIV-1
Infection among Factory Workers in Ethiopia, 1997-2001
[14] Chesney, MA. 2000. Factors affecting adherence to
antiretroviral therapy. Clinical Infectious Diseases
30(2):171-176.
[15] Arnsten, JH, Demas,PA, Farzadegan, H, Grant, RW,
Gourevitch, MN, Chang, CJ, Buono, D, Eckholdt, H,
Howard, AA, & Schoenbaun, EE .2001. Antiretroviral
therapy adherence and viral suppression in HIV-infected
drug users: comparison of self-report and electronic
monitoring. Clinical Infectious Diseases 33:1417-1423.
[16] Tadios Y, Davey G. Retroviral drug adherence & its
correlates in Addis Ababa, Ethiopia; Ethiop Med J 2006;
44:237-244.
[17] Chishimba S and Zulu F. The 3x5 HIV and AIDS Treatment
Plan; Challenges for Developing Countries from Zambian
Perspective. Int Conf AIDS. (2004). 15: Abstract no.
B11132
[18] Orrell C, Bangsberg DR, Badri M, Wood R. Adherence is
not a barrier to successful antiretroviral therapy in South
Africa. AIDS. 2003;17:1369–75. doi: 10.1097/00002030-
200306130-00011.
[19] Sarna A., Luchter S., Giebel S., Munyau P., Kaai S., Shikely
K., Mandaliya K., Hawken M., Van Dam J., and
Temmerman M. “Promoting adherence to antiretroviral
therapy through a directly administered antiretroviral
therapy (DAART) strategy in Mombassa Kenya, Horizons
Research Update. Nairobi: Population Council. 2005.
[20] Poppa, A., Davidson, D., Deutsch, J., “British HIV
Association (BHIVA)/British Association for sexual health
and HIV (N‘BASHH) guidelines on provision of adherence
support to individuals receiving antiretroviral therapy”
HIVMedicine 2004 ,5:S 46-60.

Science Journal of Public Health 2014; 2(2): 69-77 77
[21] Ekanem Esu Williams,A Study on Social Support and ART
Adherence at Carletonville Hospital and Zola Clinic in
Gauteng Province.South Africa,2007
[22] Michaels, D, Darder, M; Boulle A; et al; Adherence to
antiretroviral treatment: the Experience of patients on ART
for longer than 24 months. Oral presentation, 2nd South
Africa AIDS Conference Durban, June 2005.