Investigación original / Original research
Social position, gender role, and treatment
adherence among Colombian women living
with HIV/AIDS: social determinants of
health approach
Marcela Arrivillaga,1 Michael Ross,2 Bernardo Useche,3
Martha Lucia Alzate,4 and Diego Correa1
Suggested citation
Arrivillaga M, Ross M, Useche B, Alzate ML, Correa D. Social position, gender role, and treatment ad-
herence among Colombian women living with HIV/AIDS: social determinants of health approach. Rev
Panam Salud Publica. 2009;26(6):502–10.
ABSTRACT
Objective.
To assess and analyze the associations between adherence to treatment and so-
cial position in women living with HIV/AIDS.
Method.
A cross-sectional, descriptive, correlational study among 269 Colombian women
was conducted. Participants completed three questionnaires: a socio-demographic and clinical
characteristics survey, a treatment adherence scale, and a social position survey.
Results.
Women of low social position had a significantly higher probability of low treat-
ment adherence (OR = 5.651, P < 0.0001), and the majority of social position variables mea-
sured had a significant effect on adherence. A general model considering the variables “type of
national health care plan” (“contributive,” “subsidized,” or, in the case of vinculadas or the
uninsured, “none”); “having HIV-positive children”; and “level of viral load” was statisti-
cally reliable in predicting study participants’ treatment adherence. Membership in the subsi-
dized plan or being uninsured had a greater effect on the probability of low adherence than
membership in the contributive plan (OR = 3.478, P < 0.0001). Univariate regression analy-
ses confirmed that women with HIV-positive children and a viral load ? 400 copies/ml were
more likely to have low adherence than women without those characteristics (OR = 2.395,
P = 0.0274 and OR = 2.178, P = 0.0050, respectively).
Conclusions. Improving women’s adherence to HIV/AIDS treatment in Colombia would re-
quire eliminating barriers to national health care system and comprehensive health care services
and implementing programs that take into account women’s role as maternal caregivers The
findings underscore the need to integrate variables related to gender inequality and social posi-
tion in treatment adherence analysis, as advocated in the social determinants of health approach.
Key words
Patient compliance; HIV; AIDS; gender identity; Colombia.
1
Department of Social Sciences, Pontificia Universi-
The relationship between socio-eco-
(1–5). In Latin America, conceptual mod-
dad Javeriana, Cali, Colombia. Send correspon-
nomic-political factors and health out-
els studying the link between macro so-
dence and reprint requests to: Marcela Arrivillaga,
Investigadora, Departamento de Ciencias Sociales,
comes has been recognized since the be-
cioeconomic structures and growing
Pontificia Universidad Javeriana–Cali, Calle 18
ginning of the history of public health
health inequities have been used in so-
No. 118-250, Cali,Valle del Cauca, Colombia; tele-
cial medicine and public health since the
phone: +57-2 321-8200; fax: +57-2 555-2550; e-mail:
marceq@javerianacali.edu.co
3
School of Public Health, University of Texas, Hous-
1970s (5).
2
Center for Health Promotion and Prevention Re-
ton, Texas, United States of America.
4
It is also well established that the most
search, School of Public Health, University of
School of Nursing, Universidad Nacional de
Texas, Houston, Texas, United States of America.
Colombia, Bogotá, Colombia.
valuable resources of a society—finan-
502
Rev Panam Salud Publica/Pan Am J Public Health 26(6), 2009

---

Arrivillaga et al. • Social position, gender role, and HIV/AIDS treatment adherence in Colombian women
Original research
cial benefits, and the mechanisms of so-
amount of the insurance premium), re-
economic resources, gender, and social
cial control—are unequally distributed
ceive restricted benefits (about half of
position. Therefore, the population most
or assigned, not only among different so-
those provided under the POS-C) subsi-
affected by barriers to care and treatment
cial positions but also by gender. This
dized by the government. This segment
is largely composed of poor women,
social and gender differentiation results
of the population obtains health care sub-
whose limited ability to pay for services,
in institutionalized imbalances between
sidies once they are registered with the
status as women, and low social position
men and women in access to and control
System for Identification of Government
block their access to health care (17, 18).
of resources (6, 7), and produces impor-
Subsidies Recipients (SISBEN). Left out
This, in turn, diminishes their ability to
tant differences in health outcomes (8).
of this two-part system is a third cate-
obtain and sustain treatment (19). In ad-
In Latin American countries, policies
gory of Colombians known as vinculados
dition, poor women with HIV/AIDS
based on gender norms restrict women’s
or the uninsured. In spite of being below
face other challenges, such as unemploy-
access to productive resources such as
the poverty line, and due to the policies
ment or informal employment (working
land, income, education, technology,
of the SISBEN, vinculados have not yet
in the informal job sector); low income;
and credit. For example, in Colombia, it
achieved access to the national health
low education; and inadequate nutrition
is estimated that housing ownership is
care system or the supply-driven re-
and housing conditions—all of which
15% higher for men; unemployment is
sources originally targeted toward this
imply difficulties in obtaining sufficient
6.8% higher in women; 52% of married
population. Consequently, about 10 mil-
resources to enable them to follow any
women do not have a personal income;
lion citizens (approximately 25% of the
kind of ART satisfactorily (20).
38% of employed women work in the
Colombian population) do not have any
Traditionally, adherence to treatment
most precarious informal sector; 30% of
type of health care coverage (15) and
has been studied using both a biomed-
households are headed by a woman, and
must rely on health care services pro-
ical and a psychosocial approach, based
65% of those households are families liv-
vided by local governments (primary
on analysis of general categories of vari-
ing in conditions of poverty (9–10).
care interventions, high-level or emer-
ables related to the patient (e.g., age, ed-
Colombia has the third-highest num-
gency services, and some intermediate-
ucation level, and coping mechanisms);
ber of HIV/AIDS cases in Latin America.
level services), which often entail ex-
health care personnel (e.g., quality of
The country’s level of incidence has
tremely high out-of-pocket expenses.
health care and communication skills);
shown a linear trend since 1983 and cur-
In the SGSSS, HIV/AIDS is considered
the illness or condition (e.g., state of the
rently falls between 0.6% and 0.7% in
a high-cost condition and is therefore not
immune system); treatment (e.g., side ef-
women 15–49 years old (11, 12). How-
covered for approximately 66% of sero-
fects); and health care systems (e.g.,
ever, it is estimated that for every case
positive men and women, effectively
quality of health care and geographical
registered in the national HIV/AIDS epi-
blocking their access to antiretroviral
barriers) (21–25). Several studies have
demiological surveillance system there
therapy (ART). Data by gender on sero-
shown that compared to male samples
are about seven non-registered cases.
positives without access to ART are not
women with HIV/AIDS face major treat-
Heterosexual transmission of HIV is the
available. Because access to ART and
ment adherence barriers, including de-
main mode of transmission among Co-
other types of health care is underpro-
lays in medical attention, non-use of
lombian women, with 72% of cases de-
vided, non-continuous, deficient, and in-
ART, lack of financial support, poor
tected in women attending vertical
adequate for the uninsured poor popula-
quality of health care, and problems re-
transmission prevention programs who
tion, many people living with HIV/AIDS
lated to the doctor–patient relationship
report stable relationships with hus-
use a legal mechanism called a tutela (a
(26, 27). These problems are intensified
bands or partners (13, 14).
writ for the protection of constitutional
among women living in rural areas (28).
Within this context, the current model
rights) to gain access to treatment (13, 16).
It has been suggested that certain aspects
for health care services in Colombia was
However, activating this mechanism re-
of the larger social context influence
established in 1993 by Law 100 and rati-
quires engaging in multiple bureaucratic
treatment adherence (29), including fac-
fied in 2007 by Law 1122. Law 100 also
procedures, resulting in delayed access to
tors such as poverty; social/gender in-
established a two-tier system of national
services, stress for patients and their fam-
equality; war; violence; cost of medica-
health care coverage—contributive and
ilies, and risk of worsening condition. Be-
tion; CD4 count5 and viral load; cost of
subsidized—provided through the Gen-
cause Colombia’s current health system is
transportation; cost of missing days from
eral System of Social Security in Health
market-based, unemployed women and
work; cost of food and safe water; and
(Sistema General de Seguridad Social en
those working in the informal sector ex-
costs associated with the inevitable
Salud, SGSSS). Those enrolled in the Con-
perience extremely limited access to
changes in lifestyle resulting from HIV
tributive Mandatory Health Plan (Plan
health care services. Because most of this
infection. Along with the direct effects of
Obligatorio de Salud Contributivo, POS-C),
population is uninsured, any health ser-
gender inequality on women’s health
which includes all individuals who are
vices they do receive require extensive,
care, it is important to analyze the indi-
formally employed or earning more than
often unaffordable out-of-pocket pay-
rect effects of gender on treatment ad-
twice the minimum wage, pay monthly
ments. Barriers to health care for these
herence, such as women’s limited access
dues for their health care coverage. Those
women are further exacerbated in the
to the economic resources that determine
enrolled in the Subsidized Mandatory
event of widowhood, partner abandon-
their social position.
Health Plan (Plan Obligatorio de Salud
ment, marital separation, or partner un-
Subsidiado, POS-S), which includes gov-
employment (7).
5
Number of CD4+ T-lymphocyte cells per mm3 of
ernment-defined subsidy-eligible indi-
In the case of HIV/AIDS, access to
blood (measure used to analyze prognosis of HIV-
viduals (those unable to pay the total
health care resources is based on level of
infected individuals).
Rev Panam Salud Publica/Pan Am J Public Health 26(6), 2009
503

---

Original research
Arrivillaga et al. • Social position, gender role, and HIV/AIDS treatment adherence in Colombian women
To complement current biomedical
ple pool, which comprised a hidden and
representing “never” and 4 representing
and psychosocial approaches to the
hard-to-reach segment of the population
“always”). The variable was categorized
study of treatment adherence, and
(female seropositives), along with data
on a scale of 21 to 84, and “low” and
broaden the social perspective toward
inconsistencies and incomplete records
“high” levels of adherence were estab-
the issue, the current study aimed to: 1)
at the participating health centers, a con-
lished respectively as 21–61 points and
assess and analyze the associations be-
venience sample was used. Inclusive
62–84 points. This classification was
tween treatment adherence and social
criteria included being under medical
based on the overall mean of compliance
position, and 2) identify and examine the
control in a health care center, being
in patients classified as “high adherence”
association between treatment adher-
diagnosed more than six months prior to
in past studies, which ranged from 37%
ence and specific social position and
the study, having more than three
to 88% (32, 33). “High adherence” in the
socio-demographic/clinical variables in
months of treatment, being older than 18
current study was defined as complying
Colombian women living with HIV/
years, and not being pregnant. The main
with at least 64% of the treatment re-
AIDS. These objectives emerged after a
exclusion criterion was documentation
quirements (corresponding to a score of
review of available data revealed a lack
of a mental disorder diagnosis in the
62 or higher in the treatment adherence
of consistency regarding the association
clinical history. Out of the 280 women
scale). Use of Cronbach’s alpha test to as-
between socioeconomic status and treat-
initially contacted, 11 (4%) refused to
sess the reliability of the scale resulted in
ment adherence. The need for this type
participate in the study for fear their
a value of 0.812 (an alpha value ? 0.7 in-
of study was further underscored when
names would be revealed (despite guar-
dicates an acceptable level of scale relia-
a more recent, systematic review includ-
antees of confidentiality).
bility) (34). Table 1 presents the treatment
ing 116 studies concluded that there is
Women were recruited over an 11-
adherence variables.
no statistically significant association be-
month period from HIV/AIDS programs
tween socioeconomic status and adher-
at health care centers and nongovern-
Social position survey. The social posi-
ence to HIV/AIDS treatment (30). At the
mental organizations (NGOs) for people
tion survey included 16 items that evalu-
same time, the lack of existing data on
living with HIV/AIDS. Infection with
ated eight characteristics of social posi-
this topic eliminated the possibility of
HIV was confirmed with a Western blot
tion: residence; socioeconomic status;
formulating directional hypotheses.
test. Three questionnaires were applied
education; type of health care plan; job/
For the purposes of this study, social po-
to each study participant by appropri-
occupational profile, income; [women’s]
sition was defined according to the crite-
ately trained personnel (except in the
property holdings; and access to financial
ria of the social determinants of health
case of low educational level of the par-
credit. The survey content was validated
approach. This approach defines social
ticipant, which required administration
with a pilot test and expert judges. The
position in terms of the variables that col-
of the surveys by the researchers, with
variable was categorized on a scale of 7 to
lectively determine a woman’s “place” in
the assistance of a support team). The re-
21, with levels of social position defined
society as well as the place where she ac-
search project was approved by the insti-
as “low” (7–11 points); “medium” (12–16
tually lives, including area of residence;
tutional review boards of the National
points); and “high” (17–21 points). This
socioeconomic status; education; type of
University of Colombia (Universidad Na-
classification was made taking into ac-
health care coverage (including “none”);
cional de Colombia, UN) and Javeriana
count current Colombian criteria defining
job or livelihood; income level; and access
University in Cali (Pontificia Universidad
socioeconomic levels officially estab-
to economic resources such as credit and
Javeriana Cali, PUJC). Written informed
lished in 1994 (35). The survey parame-
property (3, 5, 31). Therefore, this study
consent was obtained from all study
ters for social position were also based on
focused on the above-mentioned vari-
participants.
Breilh’s criteria (36). Table 2 shows the
ables, taking into account specific effects
variables included in the social position
of gender, to assess the treatment adher-
Measures
survey.
ence of Colombian women.
Socio-demographic and clinical data
Data analysis
METHOD
survey. For the survey on socio-demo-
graphic and clinical characteristics, study
The statistical analyses included de-
This study is part of the Social Perspec-
participants provided data on area of res-
scriptive and analytic procedures. For de-
tive of Adherence to Treatment in Colom-
idence, age, marital status, “family mem-
scriptive purposes, the frequencies of re-
bian Women with HIV/AIDS macro proj-
ber(s) infected with HIV/AIDS,” time of
sponses to the items on the three survey
ect, which consists of two sequential
diagnosis, mechanism of HIV transmis-
instruments were examined first. For an-
studies (quantitative and qualitative)
sion, and most recent viral load and CD4
alytical purposes, univariate and multi-
conducted between 2006 and 2008. This
count.
ple logistic regression models were esti-
report analyzes the quantitative data,
mated and odds ratios (OR) were used to
using a cross-sectional, descriptive, cor-
Treatment adherence scale. The treat-
characterize associations between the
relational design.
ment adherence questionnaire included
three groups of variables (social position,
21 items evaluating practices, motiva-
socio-demographic and clinical charac-
Participants and procedures
tions, beliefs, affect, barriers, and living
teristics, and treatment adherence). The
conditions. Before its application, the
statistical programs SPSS version 14.0
Participants included 269 women di-
content was validated with a pilot test
(SPSS Inc., Chicago, Illinois, USA) and
agnosed with HIV/AIDS in five Colom-
and expert judges. The answer format
SAS version 9.0 (SAS Institute Inc., Cary,
bian cities. Due to the nature of the sam-
followed a four-point Likert scale (with 1
North Carolina, USA) were used.
504
Rev Panam Salud Publica/Pan Am J Public Health 26(6), 2009

---

Arrivillaga et al. • Social position, gender role, and HIV/AIDS treatment adherence in Colombian women
Original research
TABLE 1. Survey items for a questionnaire designed to determine treatment adherence in a cross-
To complement the analysis, multiple
sectional, descriptive, correlational study conducted among HIV-positive women in five Colom-
logistic regression was conducted to
bian cities,a 2006–2008
identify and examine which indepen-
dent variables related to social position,
Compliance with medical appointments
socio-demographic, and clinical data
Take medication as prescribed
were associated with treatment adher-
Active research of information and education on HIV/AIDS and treatment
ence. Through multiple correspondent
Direct communication for matters related to treatment with health care personnel
analyses, the problem of multicollinear-
Handle side effects of antiretroviral therapy (ART)
ity was avoided, and the variable set was
Seek family support to comply with treatment
reduced. A general model made up of
Seek social support to comply with treatment
variables including type of health care
Balance adherence with daily life routine and obligations
plan, having HIV-positive children, and
Resolution of health care system barriers to maintain treatment
level of viral load was found to be statis-
Understand and accept diagnosis
tically reliable to predict low or high
Disruption of treatment due to emotional disorders attributed to HIV/AIDS
adherence for the study participants
Role as family caregiver (“The care of my health can wait. Most important is the family care.”)
(–2 * log – likelihood = 316.620; chi-
Doubts about treatment benefits
square (3) = 32.3803; P = < 0.0001). How-
Determination to continue treatment
ever, as shown in Table 5, the effect of
Personal motivations contributing to treatment compliance
the individual variables differed. For ex-
Perception that stigmatization by family and community interferes with treatment
ample, as noted in the table, enrollment
Perception that stigmatization by health care workers interferes with treatment
in the subsidized health care plan or
Geographic barriers to accessing health care center
being uninsured had a greater effect on
Economic problems interfering with treatment
the likelihood of low adherence than en-
Sufficient quality of nutrition and availability of proper diet to ensure adequate health
rollment in the contributive plan (relia-
Difficulties in paying out-of pocket health care expenses for continuous treatment
bility 99%).
a Cali and surrounding towns, Bogotá, Villavicencio, Pasto and surrounding towns, and Medellín.
To verify the effects of the non-signifi-
cant variables of the model obtained
through multiple regression analysis, uni-
variate regression analyses were con-
RESULTS
The univariate logistic regression
ducted. Results of these tests indicated
analysis also indicated (with reliability
that the variables “having HIV-positive
Socio-demographic and clinical
or confidence intervals (CIs) of 95%),
children” (b = 0.8733; P = 0.0274; OR =
characteristics
that the majority of the variables in the
2.395, CI = 1.102–5.203) and “having a
social position survey had a significant
viral load ? 400 copies/ml” (b = 0.7786;
The sample study was made up of 269
effect on low adherence, including low
P = 0.0050; OR = 2.178, CI = 1.266–3.750)
women diagnosed with HIV/AIDS and
socioeconomic status (regression coeffi-
had a significant effect on low adherence.
living in one of five cities in Colombia:
cient (b) = 0.8093; P = 0.0070; OR = 2.246,
Social position, socio-demographic, and
Cali (including surrounding towns), Bo-
CI = 1.248–4.044); educational level clinical variables that did not have a sig-
gotá, Villavicencio, Pasto (including sur-
of high school or less (b = 1.0474; nificant effect on adherence included
rounding towns), and Medellín. Table 3
P = 0.0003; OR = 2.850, CI = 1.625–5.000);
“being a full-time housewife”; “being
shows the socio-demographic and clini-
enrollment in the SGSSS subsidized
solely responsible for family income”;
cal characteristics of the study sample.
health care plan, or categorization as
age; marital status; area of residence; hav-
vinculada (uninsured) (b = 1.4212; P =
ing children; “living with children and
Treatment adherence and social
< 0.0001; OR = 4.142, CI = 2.383–7.201);
partner,” “living with children without a
position
job/occupational profile A (e.g., factory
partner”; “having a partner with HIV/
worker, farm worker, or informal sec-
AIDS”; “living with other family mem-
Of the total sample, 43% of the women
tor worker, such as a day laborer, street
bers or friends”; and time of diagnosis.
presented low adherence to treatment
vender, short-order cook, maid, park-
(vs. 57% who obtained high levels of ad-
ing lot attendant, etc.) (b = 1.0182; P =
DISCUSSION
herence), and 80% were classified as
0.0030; OR = 2.768, CI = 1.413–5.422);
members of a low social position (vs.
monthly income level of less than
Adherence to treatment in HIV/AIDS
20% classified as medium or high social
US $200 (b = 0.7276; P = 0.0133; OR =
cases is a key public health issue, espe-
position) (data not shown).
2.070, CI = 1.164–3.682); and lack of ac-
cially with regard to women, who repre-
To assess the association between so-
cess to financial credit (b = 0.8885; P =
sent a vulnerable population. The main
cial position and treatment adherence, a
0.0194; OR = 2.431, CI = 1.155–5.120). It
findings of the current study indicate
univariate logistic regression model was
should be noted that for the property
1) an association between treatment ad-
constructed. As shown in Table 4, the es-
holdings variable, the “no property” re-
herence and overall social position, and
timated risk for low adherence was five
sponse did not show a significant associ-
between treatment adherence and spe-
times higher for the women in a low so-
ation with treatment adherence (b =
cific social position variables; 2) Colom-
cial position compared to those in a
0.6590; P = 0.0782; OR = 1.933, CI =
bian women with HIV/AIDS, especially
medium or high social position.
0.928–4.025).
those in a low social position, face seri-
Rev Panam Salud Publica/Pan Am J Public Health 26(6), 2009
505

---

Original research
Arrivillaga et al. • Social position, gender role, and HIV/AIDS treatment adherence in Colombian women
TABLE 2. Survey items for a questionnaire designed to determine level of social position in a
Although living in poverty and being fe-
cross-sectional, descriptive, correlational study conducted among HIV-positive women in five
male have been shown to be fundamen-
Colombian cities,a 2006–2008
tal determinants of the dynamics of the
HIV epidemic (17–20), research on ad-
Area of residence
Type of national (SGSSSb) health plan
herence to treatment in this specific pop-
Urban
Subsidizedc
Rural
Contributived
ulation is scarce. The results of this study
Contributive plus PMPe
concur with previous data on the rela-
Socioeconomic status (level)
1–2 (low)
Nonef
tionship between poor economic situa-
3–4 (medium)
Type of health plan membership
tion and HIV/AIDS treatment adher-
5–6 (high)
Subscriber (enrollee)
ence in male and female samples (29,
Education (highest level reached)
Beneficiary (dependent)
37–39). However, more research is re-
Grammar school
quired to confirm associations between
High school
treatment adherence and variables re-
Trade school/college
lated to social position and gender role.
Postgraduate
Job/occupational profile
Profile A: Factory worker; farm worker; informal sector worker (e.g., day laborer, street vender, maid,
parking lot attendant, short-order cook, etc.)
Profile B: Small business owner; low-level salesperson; service industry worker; low-level employee of
TABLE 3. Socio-demographic and clinical char-
public or private company (e.g., secretary or clerk); self-employed professional with undergraduate degree.
acteristics of HIV-infected women in a cross-
sectional, descriptive, correlational study con-
Profile C: Female business owner or manager of large company; upper-level employee of public or private
ducted in five Colombian cities,a 2006–2008
company; self-employed professional with postgraduate degree
Full-time housewife
No. HIV-infected
Unemployed
Characteristic
women (n = 269)
%
Monthly income (level)
Property holdings
Area of residence
< US $200
None
Urban
243
90
US $200–US $1 000
1 property
Rural
26
10
> US $1 000
2 or more properties
Age (years)
Solely responsible for family income
Access to financial credit
18–25
30
11
Yes
None
26–40
140
52
No
? US $17 000
41–59
89
33
> US $17 000
? 60
10
4
Marital status
a Cali and surrounding towns, Bogotá, Villavicencio, Pasto and surrounding towns, and Medellín.
Married/in common
b General System of Social Security in Health (Sistema General de Seguridad Social en Salud, SGSSS).
law relationship
141
52
c Subsidized Mandatory Health Plan (Plan Obligatorio de Salud Subsidiado, POS-S) for government-defined subsidy-eligible
Single
59
22
individuals (those unable to pay the total amount of the insurance premium) who are registered with the System for Identifi-
Separated or divorced
40
15
cation of Government Subsidies Recipients (SISBEN)) and receive restricted benefits (about half of those provided under the
Widowed
29
11
POS-C) subsidized by the government.
d Contributive Mandatory Health Plan (Plan Obligatorio de Salud Contributivo, POS-C) for individuals who are formally em-
Has children
ployed or earning more than twice the minimum wage and pay monthly dues for health care coverage.
Yes
161
60
e Prepaid Medical Plans (Planes de Medicina Prepagada, PMP) (supplemental private health insurance).
No
108
40
f In Colombia, those who are unable to pay any health insurance premiums, known as vinculados, remain uninsured and thus
Family member living with
face barriers to medical care.
HIV/AIDS
Partner
136
51
Child
51
19
Living with:
ous barriers to adherence imposed by
cator of social position. In contrast, this
Children (without a partner)
94
35
the structure of the health system; and 3)
study considered social position as a
Children and partner
99
37
Other family members
women’s role as caregivers for HIV-pos-
group of characteristics acting together to
or friends
28
28
itive children is a gender condition that
define, in this case, seropositive wom-
Source of HIV transmission
affects adherence to their own treatment
en’s place in Colombian society.
Sexual
264
98
requirements. The statistically signifi-
According to the current study results
Blood transfusion
3
2
Time of diagnosis
cant associations found in the study un-
on specific barriers to treatment adher-
> 2 years ago
210
78
derscore the importance of social deter-
ence, women living in poverty condi-
1–2 years ago
42
16
minants versus practices, motivations,
tions have a high probability of health
<1 year ago
17
6
beliefs, affect, and living conditions with
care and socioeconomic deficits such as
Viral load
regard to treatment adherence.
limited access to ART, low income, inad-
< 400 copies/ml
169
63
? 400 copies/ml
99
37
Other studies that have explored asso-
equate diet and housing, and low levels
CD4 countb
ciations between socioeconomic status
of education, as well as extra out-of-
< 200 cells/ml
51
19
and HIV/AIDS treatment adherence
pocket health expenses. Some authors
200–300 cells/ml
112
42
have had inconsistent results. In general,
(37) found that poor women with
> 300 cells/ml
105
39
socioeconomic aspects in these studies
HIV/AIDS experience delayed diagno-
a Cali and surrounding towns, Bogotá, Villavicencio, Pasto
are measured and analyzed as indepen-
sis and had less access to health care
and surrounding towns, and Medellín.
b
dent factors. A majority of authors used
compared to their male counterparts,
Number of CD4+ T-lymphocyte cells per mm3 of blood
(measure used to analyze prognosis of HIV-infected
the variable of income level as the indi-
which contributed to a bleak prognosis.
individuals).
506
Rev Panam Salud Publica/Pan Am J Public Health 26(6), 2009

---

Arrivillaga et al. • Social position, gender role, and HIV/AIDS treatment adherence in Colombian women
Original research
TABLE 4. Univariate logistic regression analysis of the association between low treatment ad-
herence and low social position (versus medium and high social position) in a cross-sectional,
descriptive, correlational study conducted among 269 HIV-positive women in five Colombian
cities,a 2006–2008
Regression
Wald test
Odds ratio
coefficient (b)
statistic
DFb
P-value
(95% CIc)
Constant (b )
–1.7427
18.0960
1
<0.0001
NAd
0
Low social position (b )
1.7319
15.8323
1
<0.0001
5.651 (2.408–13.262)
1
a Cali and surrounding towns, Bogotá, Villavicencio, Pasto and surrounding towns, and Medellín.
b DF = degrees of freedom.
c CI = confidence interval.
d NA = not applicable.
TABLE 5. Multivariate logistic regression analysis of social position and socio-demographic/clin-
ical characteristics associated with treatment adherence among HIV-positive women (n = 269) in
a cross-sectional, descriptive, correlational study conducted in five Colombian cities,a 2006–2008
Regression
Wald test
Odds
95% CIc
coefficient
statistic
DFb
ratio
(range)
P-value
Constant
–1.1936
27.6470
1
NAd
NA
< 0.0001
Member of subsidized
national health care plan,
or vinculada (uninsured)
1.2465
18.0501
1
3.478
1.957–6.181
< 0.0001
Has children with HIV/AIDS
0.7317
2.9675
1
2.079
0.904–4.779
0.0850
Viral load ? 400 copies/ml
0.5128
2.8939
1
1.670
0.925–3.015
0.0889
a Cali and surrounding towns, Bogotá, Villavicencio, Pasto and surrounding towns, and Medellín.
b DF = degrees of freedom.
c CI = confidence interval.
d NA = not applicable.
Another significant finding of the
the capacity to afford it or the ability to
In other words, the Colombian national
study was that one of the main barriers
meet the eligibility criteria for SISBEN
health care system does not provide
to treatment adherence for poor women
subsidies (42, 43). In terms of insurance
gender-equitable access to its services (ac-
in Colombia with HIV/AIDS was being
deficits, the continued prevalence of eco-
cess based on need, regardless of gender
uninsured, or enrollment in the subsi-
nomic and gender discrimination leaves
or social position).
dized national health plan. Women en-
women more negatively affected than the
In addition to the above, a main finding
rolled in the subsidized plan generally
majority of male enrollees (43, 44). Specif-
of this study involves the effect of hav-
receive care through POS-S Administra-
ically, women in a lower social position
ing HIV-positive children on adherence
tor Agencies (Empresas Promotoras de
receive insufficient health services, have
among maternal caregivers. Some au-
Salud del Régimen Subsidiado), which pro-
higher out-of-pocket health care costs,
thors have shown that women with HIV/
vide limited benefits. This finding may
and, as this study shows, a higher proba-
AIDS face different, specific stresses and
help explain the low health care quality
bility of less adherence to HIV/AIDS
are often overburdened with family care
and patient satisfaction reported by
treatment, resulting in a much greater
responsibilities at the expense of their
other researchers (28, 40). According to
risk of mortality from AIDS. Most ad-
own health (45–47). Women facing the
the authors of the current study, this
herence problems among seropositive
double challenge of being patients and
treatment adherence barrier is an indi-
women enrolled in the subsidized plan
homemakers frequently must handle
rect consequence of less-than-equal ac-
stem from the restricted access to health
family-related difficulties that interfere
cess to timely and continuous care by
services that characterizes this plan.
with their treatment adherence. Some au-
women of low social position in Colom-
Many vinculados remain on long waiting
thors have observed that traditional mod-
bia, stemming from the flawed structure
lists for access to treatment in public hos-
els of medical care do not sufficiently ad-
of the current national health system.
pitals and thus experience serious prob-
dress the specific needs of these women.
This study confirms that in the case of
lems related to access and quality of care.
For example, according to previous stud-
poor women with HIV/AIDS there is a
In addition, these uninsured individuals
ies, when women receive help from oth-
systematic non-fulfillment of the commit-
have no chance of participating in the few
ers in the area of childcare their adher-
ments made by the Colombian govern-
existing HIV/AIDS integral treatment
ence to medical appointments improves
ment related to the right to universal and
programs. These barriers to access often
significantly (26). According to predomi-
equitable access to health care (41). Access
force women to take legal action against
nant cultural norms in Colombia, women
to the national health system for the
the national health system in order to
assume the main responsibility of taking
Colombian population depends on either
maintain access to HIV/AIDS treatment.
care of their families. A diagnosis of
Rev Panam Salud Publica/Pan Am J Public Health 26(6), 2009
507

---

Original research
Arrivillaga et al. • Social position, gender role, and HIV/AIDS treatment adherence in Colombian women
HIV/AIDS in one or more of their chil-
mological, and praxiological framework
impact of social position and gender on
dren constitutes an additional burden for
for analyzing social processes related to
adherence among people living with
mothers who are themselves infected.
individual health. Combined with the
HIV/AIDS is also needed.
This situation is especially onerous for
application of a social stratification
Finally, through future research on
women living without a partner. In the
model (4), the use of this framework can
treatment adherence and its association
current study, it was observed that
help generate viable mechanisms for car-
with social position, socio-demographic,
women with HIV/AIDS were more wor-
rying out political actions promoting the
and clinical variables, the authors recom-
ried about and devoted more attention to
health, quality of life, and treatment ad-
mend further validation of the hypothe-
their children’s treatment than their own.
herence of people living with HIV/AIDS.
sis presented in this report. Comparative
They kept medical appointments, com-
Carrying out structural interventions in
studies should be carried out between
plied with prevention practices, ensured
Colombia to promote gender equality
male and female samples to identify
their children received the prescribed
and related improvements in health out-
links between these variables and thus
schedule of ART, and, with the same level
comes for women requires the design and
improve the characterization of gender
of diligence, pursued legal actions to try
implementation of more and better public
differences. Other studies on structural
to guarantee continued access to it.
policies to promote equitable income for
barriers of health systems and their asso-
woman, facilitate their access to credit,
ciation with gender-role variables would
Conclusions
and provide them with the right to work
also contribute to a more integrated ap-
in order to increase their well-being and
proach to analysis of treatment adher-
The main findings of this study pro-
economic independence. This, in turn, re-
ence. Lastly, to address the limitations in
vide evidence of statistical associations
quires consideration of women’s role as
the scope of the current study’s sample
between social position and HIV/AIDS
maternal caretakers as well as other gen-
population in terms of social position,
treatment adherence. They also demon-
der-related issues directly or indirectly af-
explained above, the authors recom-
strate associations between adherence
fecting women’s health. Another prereq-
mend the inclusion of a wider propor-
and the socio-demographic variables
uisite for achieving progress in this area is
tion of women with medium and high
“type of health care plan” and “having
a rights-based health system based on the
social position. While the current study
HIV-positive children.” However, the
principles of equity and universality. It is
experienced difficulty in locating sero-
authors conclude that the concepts of
inconceivable that the risk of low adher-
positives with a high social position (be-
gender role and social position were not
ence for a poor, uninsured woman (vincu-
cause they tended to hide their condition
wholly integrated in their analyses of
lada) with HIV/AIDS is five times that of
and did not appear on health institu-
treatment adherence among seropositive
a woman with a medium or high social
tions’ registers or attend NGO support
women. Therefore, additional studies
position, and three times that of a woman
groups), this population could be stud-
based on social determinants of health
enrolled in the national subsidized health
ied further through the use of a different
are needed to confirm these associations.
care plan. This level of disparity calls for
research approach, such as the use of
Advocating a perspective beyond that of
structural change in the health system as
special quantitative research sampling
traditional biomedical and psychosocial
soon as possible. In the meantime, the
methods, or qualitative data collection.
approaches, in future research, the au-
Colombian government must at least
thors of this study recommend HIV/
guarantee access to timely and continu-
AIDS treatment adherence be defined as
ous ART. In terms of care, the authors rec-
Acknowledgments. This study was
“complex behavior promoting adapta-
ommend comprehensive gender-focused
conducted with funding from Javeriana
tion, psychological adjustment, appropri-
programs, under state control and super-
University (Pontificia Universidad Javeri-
ate health care, and quality of life during
vision, that include childcare; psychologi-
ana) in Cali. The authors also acknowl-
the HIV/AIDS infection process, deter-
cal and family social services interven-
edge the contributions to the study’s
mined by way of life, social position, and
tion; gynecological/specialized care; and
data collection and literature review by
the health care system.” Relevant vari-
support in HIV/AIDS education, and nu-
Paula Andrea Hoyos, research assistant
ables related to “way of life” include pa-
trition. It is reasonable to argue that atten-
for the Health and Quality of Life Re-
tient practices, motivations, beliefs, and
tion to basic needs can be more beneficial
search Group, a category A research
affect as well as any living conditions af-
than direct attention to adherence, partic-
group recognized by the Administrative
fecting adherence. This definition is
ularly in the case of women with a low so-
Department for Science, Technology and
based on a conceptual model devised by
cial position. Better education and train-
Innovation (COLCIENCIAS), a Colom-
Breilh (5) that offers a theoretical, episte-
ing among health care providers on the
bian government agency.
REFERENCES
1. Rosen G. From medical police to social
3. Lynch J, Kaplan G. Socioeconomic position.
Bhuiya A, Wirth E, editors. Challenging
medicine: essays on the history of health care.
In: Berkman LF, Kawachi I, editors. Social
inequities in health: from ethics to action.
New York: Science History Publications;
epidemiology. 1st ed. Oxford: Oxford Univer-
New York: Oxford University Press; 2001.
1974.
sity Press; 2000. p. 13–35.
p. 13–23.
2. Krieger N. Theories for social epidemiology
4. Diderichsen F, Evans T, Whitehead M. The
5. Breilh J. Epidemiología crítica. Ciencia eman-
in the 21st century: an ecosocial perspective.
social basis of disparities in health. In:
cipadora e interculturalidad. Buenos Aires:
Int J Epidemiol. 2001;30(4):668–77.
Evans T, Whitehead M, Diderichsen F,
Lugar Editorial; 2003.
508
Rev Panam Salud Publica/Pan Am J Public Health 26(6), 2009

---

Arrivillaga et al. • Social position, gender role, and HIV/AIDS treatment adherence in Colombian women
Original research
6. Breilh J. El género entre fuegos de inequidad
19. International Community of Women Living
33. DiMatteo MR. Variations in patients’ adher-
y esperanza. In: Siu Bermúdez I, Dierckxsens
with HIV/AIDS. La visión de ICW 3: mujeres
ence to medical recommendations: a quanti-
W, Guzmán L, editors. Antología latinoame-
seropositivas, pobreza y desigualdad de
tative review of 50 years of research. Med
ricana y del Caribe: mujer y género. Periodo
género. London: ICW; 2004 [cited 2005 Sep
Care. 2004;42(3):200–9.
80–90. Managua: Universidad Centroameri-
10]. Available from: www.icwlatina.org/
34. Nunnally JC, Bernstein IH. Psychometric the-
cana; 1999.
imagenes/biblioteca/cuad.3.pdf.
ory. 3rd ed. New York: McGraw-Hill; 1994.
7. Gómez Gómez E. Género, equidad y acceso a
20. Stillwaggon E. Reducing environmental risk
35. Departamento Administrativo Nacional de
los servicios de salud: una aproximación em-
to prevent HIV transmission in Sub-Saharan
Estadística (CO). Estratificación socioeconó-
pírica. Rev Panam Salud Publica. 2002;11(5–
Africa. Africa Policy J. 2006;1:36–56.
mica. Versión mayo 2004. Manual metodo-
6):327–34.
21. Chesney MA. Factors affecting adherence to
lógico No. 1. Información para alcaldes y au-
8. Kunst AE, Bos V, Lahelma E, Bartley M, Lis-
antiretroviral therapy. Clin Infect Dis. 2000;30
toridades de los municipios y distritos.
sau I, Regidor E, et al. Trends in socioeco-
Suppl 2:S171–6.
Bogotá: DANE; 2004.
nomic inequalities in self-assessed health in
22. Turner-Cobb J, Gore-Felton C, Maroua F,
36. Breilh J. Breve recopilación sobre opera-
10 European countries. Int J Epidemiol.
Koopman C, Kim P, Israelski D, et al. Coping,
cionalización de la clase social para encuestas
2005;34(2):295–305.
social support, and attachment style as psy-
en la investigación social. Quito: Centro de
9. Economic Commission for Latin America and
chosocial correlates of adjustment in men
Estudios y Asesoría en Salud; 2003.
the Caribbean. Social panorama of Latin
and women with HIV/AIDS. J Behav Med.
37. Daily J, Farmer P, Rhatigan J, Katz J, Furin J.
America 2006. Santiago: United Nations; 2006
2002;25(4):337–53.
Women and HIV infection: a different dis-
[cited 2008 Mar 20]. Available from: http://
23. Burke-Miller JK, Cook JA, Cohen MH, Hessol
ease? In: Farmer P, Connors M, Simmons J,
www.eclac.cl/publicaciones/xml/4/27484/
NA, Wilson TE, Richardson JL, et al. Longitu-
editors. Women, poverty, and AIDS: sex,
PSI2006_FullText.pdf.
dinal relationships between use of highly
drugs and structural violence. Monroe (ME):
10. Departamento Administrativo Nacional de
active antiretroviral therapy and satisfac-
Common Courage Press; 1996. p. 125–44.
Estadística (CO). Mercado laboral por sexo.
tion with care among women living with
38. Carballo E, Cadarso-Suárez C, Carrera I,
Gran Encuesta Integrada de Hogares. Bogotá:
HIV/AIDS. Am J Public Health. 2006;96(6):
Fraga J, de la Fuente J, Ocampo A, et al. As-
DANE; 2008.
1044–51. Epub 2006 May 2.
sessing relationships between health-related
11. Prieto F, Osorio A, De Neira M; Grupo de
24. Arrivillaga M, Correa D, Varela MT, Holguín
quality of life and adherence to antiretroviral
Vigilancia Centinela de VIH. Prevalencia del
LE, Tovar JR. Variables psicológicas en mu-
therapy. Qual Life Res. 2004;13(3):587–99.
VIH en población general de Colombia 2003:
jeres diagnosticadas con VIH/SIDA: un estu-
39. Glass TR, De Geest S, Weber R, Vernazza PL,
VI Estudio Nacional Centinela de VIH 2003–
dio correlacional. Univ Psychol. 2006;5(3):
Rickenbach M, Furrer H, et al. Correlates of
2004. Primera Fase. Inf Quinc Epidemiol Nac
659–67.
self-reported nonadherence to antiretroviral
(Colombia). 2004;9(23–24):362–76.
25. Correa D, Salazar IC, Arrivillaga M. Impacto
therapy in HIV-infected patients: the Swiss
12. Garcia R, Prieto F, Rincon J, Arenas C,
emocional e intervención psicológica con per-
HIV Cohort Study. J Acquir Immune Defic
Caicedo S, Rey G. Reducción de la trans-
sonas diagnosticadas con VIH/SIDA. In:
Syndr. 2006;41(3):385–92.
misión madre hijo del VIH en Colombia: dos
Arrivillaga M, Correa D, Salazar IC, editors.
40. Bell EJ. Advocacy training by the Interna-
años de experiencia nacional, 2003–2005. Bio-
Psicología de la salud: abordaje integral de la
tional Community of Women Living with
médica. 2005;25(4):547–64.
enfermedad crónica. Bogotá: Manual Mo-
HIV/AIDS. Gend Dev. 2005;13(3):70–9.
13. García R, Luque R, McDouall J, Moreno LA.
derno; 2007. p. 27–50.
41. Arbeláez M. Evaluación de la eficacia del
Infección por VIH/SIDA en Colombia: estado
26. Roberts KJ, Mann T. Barriers to antiretrovi-
derecho a la salud en Colombia a partir del
del arte 2000–2005. Bogotá: Programa Con-
ral medication adherence in HIV-infected
proceso de descentralización sanitaria. Dere-
junto de las Naciones Unidas sobre el VIH/
women. AIDS Care. 2000;12(4):377–86.
cho y Salud. 2007;15(1):29–74.
SIDA, Ministerio de la Protección Social de
27. Ickovics JR, Meade CS. Adherence to HAART
42. Corte Constitucional de Colombia, Sala Se-
Colombia; 2006.
among patients with HIV: breakthroughs and
gunda de Revision, Sentencia T-760 [Colom-
14. World Health Organization; Joint United Na-
barriers. AIDS Care. 2002;14(3):309–18.
bian Official Document]. July 31, 2008.
tions Programme on HIV/AIDS; United Na-
28. Jia H, Uphold C, Wu S, Reid K, Findley K,
43. Echeverri E. La salud en Colombia: abriendo
tions Children’s Fund. Epidemiological fact
Duncan P. Health-related quality of life
el siglo . . . y la brecha de las inequidades. Rev
sheet on HIV and AIDS. Core data on epi-
among men with HIV infection: effects of so-
Gerenc Políticas Salud. 2002;1(3):76–94.
demiology and response. Colombia 2008
cial support, coping, and depression. AIDS
44. Franco S. Para que la salud sea pública: algu-
update. Geneva: UNAIDS/WHO Working
Patient Care STDS. 2004;18(10):594–603.
nas lecciones de la reforma de la seguridad
Group on Global HIV/AIDS and STI Surveil-
29. Castro A. Adherence to antiretroviral ther-
social en Colombia. Rev Gerenc Políticas
lance; 2008.
apy: merging the clinical and social course of
Salud. 2003;2(4):58–69.
15. Vega R, Jara MI. Década de los noventa: dis-
AIDS. PLoS Med. 2005;2(12):e338. Epub 2005
45. Gurung R, Taylor S, Kemeny M, Myers H.
curso de la focalización de subsidios de salud.
Oct 4.
HIV is not my biggest problem: the impact of
Rev Gerenc Políticas Salud. 2002;1(2):82–112.
30. Falagas ME, Zarkadoulia EA, Pliatsika PA,
HIV and chronic burden on depression in
16. Abadia CE, Oviedo DG. Bureaucratic itiner-
Panos G. Socioeconomic status (SES) as a de-
women at risk for AIDS. J Soc Clin Psychol.
aries in Colombia. A theoretical and meth-
terminant of adherence to treatment in HIV
2004;23(4):490–511.
odological tool to assess managed-care
infected patients: a systematic review of the
46. Hackl KL, Somlai AM, Kelly JA, Kalichman
health care systems. Soc Sci Med. 2009;68(6):
literature. Retrovirology. 2008;5(1):13. doi:
SC. Women living with HIV/AIDS: the dual
1153–60.
10.1186/1742-4690-5-13.
challenge of being a patient and caregiver.
17. Farmer P. Women, poverty, and AIDS. In:
31. Evans T, Whitehead M, Diderichsen F,
Health Soc Work. 1997;22(1):53–62.
Farmer P, Connors M, Simmons J, editors.
Bhuiya A, Wirth M, editors. Challenging in-
47. Hudson AL, Lee KA, Portillo CJ. Symptom
Women, poverty, and AIDS: sex, drugs, and
equities in health: from ethics to action. New
experience and functional status among
structural violence. Monroe (ME): Common
York: Oxford University Press; 2001.
HIV-infected women. AIDS Care. 2003;15(4):
Courage Press; 1996. p. 3–38.
32. Stein MD, Rich JD, Maksad J, Chen MH, Hu
483–92.
18. Poundstone KE, Strathdee SA, Celentano DD.
P, Sobota M, et al. Adherence to antiretro-
The social epidemiology of human immu-
viral therapy among HIV-infected metha-
nodeficiency virus/acquired immunodefi-
done patients: effect of ongoing illicit drug
ciency syndrome. Epidemiol Rev. 2004;26(1):
use. Am J Drug Alcohol Abuse. 2000;26(2):
Manuscript received on 21 October 2008. Revised version
22–35.
195–205.
accepted for publication on 20 June 2009.
Rev Panam Salud Publica/Pan Am J Public Health 26(6), 2009
509

---

Original research
Arrivillaga et al. • Social position, gender role, and HIV/AIDS treatment adherence in Colombian women
RESUMEN
Objetivo.
Evaluar y analizar las asociaciones existentes entre la adhesión al trata-
miento y la posición social de las mujeres con VIH/sida.
Métodos.
Se realizó un estudio transversal, descriptivo y correlacional con 269 co-
Posición social, papel de
lombianas. Las participantes respondieron tres cuestionarios: uno sobre las caracte-
género y adhesión al
rísticas sociodemográficas y clínicas, uno sobre su posición social y una escala sobre
la adhesión al tratamiento.
tratamiento en mujeres
Resultados.
Las mujeres de baja posición social tenían una mayor probabilidad de
colombianas con VIH/sida:
presentar baja adhesión al tratamiento (OR = 5,651; P < 0,0001); la mayoría de las va-
enfoque de los determinantes
riables de posición social influyeron significativamente sobre la adhesión. Un modelo
sociales de la salud
general con las variables “tipo de plan nacional de salud” (“contributivo”, “subsi-
diado” o, en las vinculadas y las no aseguradas, “ninguno”), “tener hijos positivos al
VIH” y “nivel de carga viral” resultó estadísticamente fiable para predecir la adhesión
al tratamiento. Ser miembro del plan subsidiado o no tener seguro influyeron más en
la probabilidad de baja adhesión al tratamiento que ser miembro del plan contribu-
tivo (OR = 3,478; P < 0,0001). El análisis de regresión monofactorial confirmó que las
mujeres con hijos positivos al VIH o carga viral ? 400 copias/mL tenían mayor
probabilidad de presentar baja adhesión que las mujeres sin esas características
(OR = 2,395; P = 0,0274 y OR = 2,178; P = 0,005, respectivamente).
Conclusiones.
Mejorar la adhesión de las mujeres al tratamiento para el VIH/sida
en Colombia requiere eliminar barreras al sistema nacional de salud, brindar servicios
integrales e implementar programas que tomen en cuenta el papel de las mujeres
como madres cuidadoras de enfermos. Estos resultados subrayan la necesidad de in-
tegrar variables relacionadas con la inequidad de género y la posición social al análi-
sis de la adhesión al tratamiento, como promueve el enfoque de determinantes socia-
les de la salud.
Palabras clave
Cooperación del paciente; VIH; SIDA; identidad de género; Colombia.
510
Rev Panam Salud Publica/Pan Am J Public Health 26(6), 2009

---