Ekstrand ML et al.

Journal of the International AIDS Society 2013, 16(Suppl 2):18717

http://www.jiasociety.org/index.php/jias/article/view/18717 | http://dx.doi.org/10.7448/IAS.16.3.18717

Research article

Prevalence and drivers of HIV stigma among health providers in

urban India: implications for interventions
Maria L Ekstrand,1,2,3, Jayashree Ramakrishna4, Shalini Bharat5 and Elsa Heylen1

Corresponding author: Maria L Ekstrand, 50 Beale Street, 13th Floor, San Francisco, CA 94105, USA. Tel: 1-415-597-9160. ([email protected])

Abstract
Introduction: HIV stigma inflicts hardship and suffering on people living with HIV (PLHIV) and interferes with both prevention
and treatment efforts. Health professionals are often named by PLHIV as an important source of stigma. This study was designed
to examine rates and drivers of stigma and discrimination among doctors, nurses and ward staff in different urban healthcare
settings in high HIV prevalence states in India.
Methods: This cross-sectional study enrolled 305 doctors, 369 nurses and 346 ward staff in both governmental and non-
governmental healthcare settings in Mumbai and Bengaluru, India. The approximately one-hour long interviews focused on
knowledge related to HIV transmission, personal and professional experiences with PLHIV, instrumental and symbolic stigma,
endorsement of coercive policies, and intent to discriminate in professional and personal situations that involve high and low
risk of fluid exposure.
Results: High levels of stigma were reported by all groups. This included a willingness to prohibit female PLHIV from having
children (55 to 80%), endorsement of mandatory testing for female sex workers (94 to 97%) and surgery patients (90 to 99%),
and stating that people who acquired HIV through sex or drugs got what they deserved (50 to 83%). In addition, 89% of
doctors, 88% of nurses and 73% of ward staff stated that they would discriminate against PLHIV in professional situations that
involved high likelihood of fluid exposure, and 57% doctors, 40% nurses and 71% ward staff stated that they would do so in low-
risk situations as well. Significant and modifiable drivers of stigma and discrimination included having less frequent contact with
PLHIV, and a greater number of transmission misconceptions, blame, instrumental and symbolic stigma. Participants in all three
groups reported high rates of endorsement of coercive measures and intent to discriminate against PLHIV. Stigma and
discrimination were associated with multiple modifiable drivers, which are consistent with previous research, and which need to
be targeted in future interventions.
Conclusions: Stigma reduction intervention programmes targeting healthcare providers in urban India need to address fear of
transmission, improve universal precaution skills, and involve PLHIV at all stages of the intervention to reduce symbolic stigma
and ensure that relevant patient interaction skills are taught.
Keywords: HIV stigma; stigma drivers; healthcare workers; India.
Received 15 April 2013; Revised 23 August 2013; Accepted 29 August 2013; Published 13 November 2013
Copyright: 2013 Ekstrand ML et al; licensee International AIDS Society. This is an Open Access article distributed under the terms of the Creative Commons
Attribution 3.0 Unported (CC BY 3.0) License (http://creativecommons.org/licenses/by/3.0/), which permits unrestricted use, distribution, and reproduction in any
medium, provided the original work is properly cited.

Introduction avoiding or delaying treatment seeking for STI/HIV infections,

Across cultures, HIV stigma has repeatedly been shown out of fear of public humiliation and fear of discrimina-
to inflict hardship and suffering on people living with HIV tion by healthcare workers [22,23]. Similarly, HIV stigma
(PLHIV) [1], as well as to interfere with their decisions to seek in Botswana, South Africa, Jamaica and India has been asso-
HIV counselling and testing [2,3], prevention of mother- ciated with delays in testing and treatment services, some-
to-child transmission (PMTCT) [48], and their willingness to times resulting in presentation beyond the point of optimal
disclose their infection to their children [9] or partners [10 drug intervention [24,25].
13], which can in turn increase the likelihood of sexual risk Unfortunately, health professionals are often named
taking. HIV stigma has also been found to be a barrier to as one of the most important sources of stigma for PLHIV.
participation in vaccine research [14] and to deter infected In sub-Saharan Africa, studies have documented discrimina-
individuals from seeking timely medical treatment [1517]. tory practices, including patient neglect, provision of differ-
These findings have been reported in both resource-rich ential treatment based on HIV status, denial of care, breach
and resource-constrained settings. Even when treatment is of confidentiality, isolation and verbal abuse by healthcare
sought, stigma fears can prevent individuals from following staff [2628]. High rates of refusal of care have also been
their medical regimen, which can lead to virologic failure and reported among nurses in Jordan [29] and stigma and
the development and transmission of a drug-resistant virus discrimination have been documented in some healthcare
[1821]. PLHIV in Senegal and Indonesia have reported settings in India also [15,3037].

1
Ekstrand ML et al. Journal of the International AIDS Society 2013, 16(Suppl 2):18717
http://www.jiasociety.org/index.php/jias/article/view/18717 | http://dx.doi.org/10.7448/IAS.16.3.18717

In the general population of healthcare seeking indi- Procedures

viduals, behavioural manifestations of HIV stigma appear For each institution, we initially approached the Hospital
to be driven by both fear of casual transmission (instrumental Superintendents or Medical Directors for permission and
stigma) and pre-existing prejudice towards vulnerable groups subsequently contacted the Department heads for assistance
(symbolic stigma) [38]. No such data are available for in recruiting nurses and ward staff. Doctors preferred that we
healthcare providers. Studies of caregivers in other resource- contact them directly to set up individual appointments.
constrained settings suggest that unwillingness to care for Following recruitment, potential participants were adminis-
PLHIV is associated with negative views, high caregiver burden, tered informed consent by study staff and following consent,
low knowledge levels and education, and having a higher an interview was scheduled. Interviews were conducted face
income [39,40]. Understanding the specific drivers of stigma to face in the participants preferred language (Marathi, Hindi
and its effects on behaviour in each setting is vital to the or English in Mumbai; Kannada, Tamil or English in Bengaluru)
by trained study staff in a private space at the work site, and
development of effective stigma reduction interventions for
lasted approximately one hour. Participants received an in-
a given population [41]. The parent study was designed to
kind gift worth 250 rupees (about 5 USD) following their
meet this need by examining levels of stigma and discrimina-
interview, consisting of packets of fancy nuts and dried fruits
tion as well as their potential drivers among healthcare pro-
in Bengaluru and a shopping bag in Mumbai.
viders, patients and the general outpatient population in
Study procedures were approved by the Institutional
two large urban settings in India. Our previous papers report Review Boards of the University of California in San Francisco,
on stigma among PLHIV [13,42,43] and the general patient the National Institute of Mental Health and Neurosciences in
population [38]. This article analyzes data from the healthcare Bengaluru, the Tata Institute of Social Sciences in Mumbai
providers and is the only one to date in India, which directly and received clearance from the Indian government.
examines stigma domains and their drivers in these three
healthcare provider groups, allowing us to conduct inter-staff Measures
comparisons. Another unique strength of this study is that it is The survey measures used in this study were based on research
the only one that includes all types of hospitals (including conducted by Herek [4549] as well as on the theoretical
charity, trust, non-profit, for profit, and public) available model subsequently developed by our team [13]. Stigma scales
in India. It is also the only study that includes data from two and drivers found to be significantly associated with mental
high prevalence areas in India, which allows for some general- health outcomes and delay of care seeking in our previous
ization of results. research with PLHIV [13,42] and with the general outpatient
population [38] were included in these analyses.

Methods Demographic information

Participants All participants were asked about their gender, age, marital
The participants for this cross-sectional study were re- status, religion, education and HIV training.
cruited in 2009 from different types of healthcare settings
in Bengaluru and Mumbai, two large Indian cities located in Potential drivers of stigma
the HIV high prevalence states, Maharashtra and Karnataka
[44]. At the time of the study, this label was given by the Contact with PLHIV. Participants indicated the fre-
Indian National AIDS Control Organization to any state quency of professional contact with PLHIV (0 never to
reporting 5% HIV prevalence in at least one key population 4 daily), and whether they personally had ever known any
or 1% in antenatal clinic settings. Field sites included PLHIV (0 no; 1 yes).
medical colleges, government hospitals or private facilities,
both for-profit and not-for-profit. Participants professional Transmission misconceptions. Four items described
forms of casual social contact through which HIV cannot be
experience with HIV patients ranged from none to extensive.
transmitted. For each item, participants indicated whether
To meet the inclusion criteria, potential participants had to
they thought HIV could be transmitted this way. The number
have worked as a doctor, nurse or ward staff in the selected
of incorrect responses was summed. A higher score reflects
hospitals/clinics for at least six months; have direct patient
more misconceptions.
contact; be at least 18 years of age; able to speak one of
the study languages; and able to give informed consent. Transmission knowledge. Participants were also asked
The term ward staff included anyone who worked on if they thought HIV could be transmitted by direct exposure
the ward at a lower level than a nurse and who had substantial to several kinds of bodily fluids, or by activities such as
patient contact (including washing, transporting, changing unprotected sex with PLHIV. The number of correct answers
linens). Most ward staff have minimal education or train- to 15 such items was calculated, with higher scores reflecting
ing and typically assist nurses or doctors with medical better knowledge.
interventions. They are also a source of information and
serve as confidants to patients. Final numbers recruited in Instrumental stigma. Two individual items measured
Bengaluru were 149 doctors, 195 nurses and 176 ward how worried participants were (0 not at all to 3 very
staff; for Mumbai the numbers were 156, 174 and 170, worried) about getting HIV-infected (i) at work and (ii) outside
respectively. of work.

2
Ekstrand ML et al. Journal of the International AIDS Society 2013, 16(Suppl 2):18717
http://www.jiasociety.org/index.php/jias/article/view/18717 | http://dx.doi.org/10.7448/IAS.16.3.18717

Negative feelings towards PLHIV. Participants reported Data analysis

their feelings towards PLHIV on a scale from 0 (very negative) Frequencies and summary statistics were used to des-
to 100 (very positive). To control for individual tendencies cribe participants responses in the three groups. Differences
to assign high or low scores in general, we subtracted the between the three healthcare worker types in categorical
rating for PLHIV from a similar rating of feelings towards men outcomes were assessed via Chi-square tests, and in con-
or women in general, so that higher anchored scores reflect tinuous outcomes via analysis of variance, with Bonferroni
more negative feelings towards PLHIV. post-hoc pairwise comparisons in case of a significant F-value.
Separate multiple regressions were performed for each
Blame. Participants indicated their agreement with the type of healthcare worker, using endorsement of coercive
statement People who got HIV/AIDS through sex or drug use policies, and intent to discriminate in personal and profes-
have gotten what they deserve, on a scale from 0 (strongly sional contexts as separate outcomes. Site (Bengaluru vs.
disagree) to 4 (strongly agree). Mumbai) was controlled for in all models. All predictors
Symbolic stigma. Six items assessed how much partici- that were associated bivariately with an outcome at pB0.25
pants moral/religious beliefs and feelings towards key popula- [50] were initially included in the model. In subsequent
tions influenced their opinions about HIV (0 not at all to models, the variable with the largest p-value was removed
4 a great deal). An overall score was computed as the mean until all remaining variables were significant at pB0.10.
of all items. Higher scores express greater stigma. This scale For endorsement of coercive policies and intent to discrimi-
had excellent reliability, with a Cronbachs a of 0.93 for doctors nate in personal context, linear regressions were performed.
and 0.81 for both nurses and ward staff. The two items for intent to discriminate at work were
modelled via separate logistic regressions. Model assump-
Perceived community stigma norms. Ten items tions regarding homoscedasticity, multicollinearity and influ-
assessed participants perceptions of the prevalence of HIV- ential outliers were adequately met. The logistic regressions
stigmatizing attitudes in their community on a five-point were performed using SAS 9.2., and all other analyses were
scale [13]. Answers were averaged into one score, with performed using SPSS 18.0.2.
higher numbers indicating more perceived community stigma.
Cronbachs a ranged from 0.85 for doctors to 0.82 for
nurses. Results
Demographic characteristics
Stigma manifestations As can be seen in Table 1, approximately half of the doctors
Intent to discriminate against PLHIV in professional situations (46%) and ward staff (51%), and almost all of the nurses
Participants were presented with two hypothetical work (98%) were female and most were married. The vast majority
situations involving care for an HIV-positive patient. One of doctors (86%) and ward staff (78%) were identified
situation posed virtually no risk of contact with bodily fluids. as Hindu, while 59% of the nurses reported being Hindu
The second situation posed a greater risk of such contact. and 36% identified as Christians, which is common in Indian
Response options were dichotomized as stigmatizing (refusing hospitals. The mean age was slightly higher among ward
or performing the task only with unnecessary precautions) staff: 39, compared to 35 for nurses and 34 for doctors.
versus non-stigmatizing (performing the task as they would Education level among ward staff varied from less than four
with any other patient). years (32%) to more than 10 years (8%) of schooling, with
45% having at least some secondary education. By definition,
Intent to discriminate against PLHIV in non-professional education was more uniform among doctors and nurses.
contexts Median household income was Rs. 40,000 (about $735) per
This was assessed by two hypothetical situations: (1) having month for doctors, Rs. 15,000 ($276) for nurses and Rs. 6000
a child who attends a school with an HIV-positive student ($110) for ward staff.
and (2) getting medical care at a clinic that treated PLHIV.
Leaving the school/clinic or avoiding contact/demanding HIV-related knowledge and experience
special precautions was scored as stigmatizing. In addition, As can be seen in Table 2, approximately 70% of doctors
participants expressed their agreement (0 strongly disagree and nurses indicated that they had received some form of HIV
to 4 strongly agree) with seven statements about avoiding training, compared to 44% of ward staff (pB0.001). Despite
social or personal contact with PLHIV. Stigmatizing responses their higher levels of HIV education, doctors and nurses did
were summed over the nine items, with higher scores indi- not score significantly higher on transmission knowledge than
cating greater intent to discriminate. ward staff (p 0.18). The mean scores on the transmission
knowledge index ranged from 11.4 out of 15 (ward staff) to
Endorsement of coercive policies 11.7 (doctors). However, the groups did differ in their mean
Participants indicated their agreement (0 strongly disagree number of casual transmission misconceptions, with the
to 4 strongly agree) with 11 statements related to the rights highest number occurring among ward staff (mean 0.8 out
of PLHIV to have a family, education, employment, and health- of 4), lower among nurses (mean0.6), and lowest among
care; the right to choose to disclose HIV status; and manda- doctors (mean0.4). Female ward staff reported less profes-
tory HIV testing. Items were dichotomized, and stigmatizing sional contact with PLHIV than their male colleagues
responses (strongly/somewhat agree) were summed. Higher (mean 2.1 vs. 2.5, respectively, p B0.05) and were, on
scores reflect greater endorsement of coercive policies. average, less knowledgeable about HIV transmission (mean

3
Ekstrand ML et al. Journal of the International AIDS Society 2013, 16(Suppl 2):18717
http://www.jiasociety.org/index.php/jias/article/view/18717 | http://dx.doi.org/10.7448/IAS.16.3.18717

Table 1. Demographic characteristics

Doctors (n305) Nurses (n369) Ward staff (n 346)

% n % n % n

Site
Bengaluru 49 149 53 195 51 176
Mumbai 51 156 47 174 49 170
Male 54 165 2 6 49 170
Religion
Hindu 86 262 59 219 78 269
Muslim 5 16 1 5 1 5
Christian 2 6 36 132 6 21
Buddhist 3 10 3 12 15 51
Others 4 11 0 1 0 0
Marital status
Currently married 58 177 68 251 77 267
Never married 41 125 29 105 8 29
Previously married 1 3 3 12 15 50
Education
54 years 32 110
57 years 23 80
810 0 1 37 129
10 years 100 305 100 368 8 27
Age: mean (SD) 33.7 (9.9) 34.9 (10.3) 39.4 (9.6)
Monthly household income in rupees: median (range) 40,000 (4000900,000) 15,000 (2700100,000) 6000 (40050,000)

knowledge score11.1 vs. 11.8; mean misconceptions 1.0 nurses and 83% of ward attendants (pB0.001). The mean
vs. 0.6 for females and males, respectively, p B0.001). scores on the symbolic stigma scale were significantly lower
Similarly, female doctors had slightly more misconceptions for doctors (mean 1.7/4.0) than for nurses (2.3) and ward
than male doctors (females: mean 0.5 vs. males: staff (2.2). Similarly, HIV-stigmatizing community norms
mean 0.3, pB0.05). There were no other gender-related were perceived to be higher by ward staff (mean 2.5/4.0)
differences reported. than by nurses (mean2.3) and doctors (mean 2.2,
More than 90% of all participants reported that they p B0.001). There were significant gender differences with
had experience caring for PLHIV, with about half in each respect to feelings towards PLHIV among both doctors
group stating that they had at least weekly contact with and ward staff. Female doctors reported significantly more
HIV-positive patients. Just over a quarter of the participants negative feelings towards PLHIV (females: mean 8 vs. males:
in each group indicated that they had known a PLHIV mean 0, p B0.01). Similarly, female ward staff held signifi-
personally. Nurses showed similar levels of worry about HIV cantly more negative feelings towards PLHIV than their male
infection at work as doctors, with about three quarters in colleagues (mean19 vs. 8, respectively, p B0.05).
both groups expressing such worries, compared to 51% of In addition, female ward staff scored higher on perceived
ward staff (pB0.001). Outside of work, more nurses (26%) stigmatizing community norms (mean 2.6 vs. 2.4, respec-
and ward attendants (27%) reported worrying about HIV tively, p B0.001) and symbolic stigma (mean2.3 vs. 2.1,
infection than doctors (17%, pB0.01). respectively, pB0.05) than male ward staff. There were no
other significant gender differences in any healthcare provider
Attitudes towards PLHIV group with respect to attitudes towards PLHIV.
Participants attitudes towards PLHIV, compared to their
feelings towards men and women in general, differed Endorsement of coercive policies regarding PLHIV
significantly between the healthcare worker groups. Overall, Ward staff endorsed a mean of 6.7 out of 11 coercive
doctors held the least negative feelings and ward staff the policies, nurses endorsed on average 6.1 and doctors 4.8
most negative, with the mean level of negative feelings (pB0.001). Mandatory testing for different groups was
towards PLHIV being 4 out of 100 (SD 26) for doctors, endorsed by large majorities of each group (See Table 3).
11 (SD31) for nurses and 13 (SD 39) for ward staff Nearly all (94% of doctors and 97% of nurses and ward staff)
(pB0.001). A high proportion of participants in all three supported mandatory testing for female sex workers (FSW),
healthcare worker types agreed with the statement that as well as for surgery patients (90% of doctors to 99% of
people who acquired HIV through sex or drugs, got what nurses, p B0.001). Mandatory testing for surgery personnel
they deserved  ranging from 50% of doctors, to 70% of was also endorsed by a majority, ranging from 73% of

4
Ekstrand ML et al. Journal of the International AIDS Society 2013, 16(Suppl 2):18717
http://www.jiasociety.org/index.php/jias/article/view/18717 | http://dx.doi.org/10.7448/IAS.16.3.18717

Table 2. Frequencies of reported stigma and other key model variables

Doctors (n305) Nurses (n369) Ward staff (n346)

% n % n % n x2

Received HIV training 73 223 71 263 44 152 77.73***

Professional contact w/PLHIV
Never 2 6 4 13 9 32 23.02***
Less than weekly 50 152 51 187 49 170
Weekly 16 48 13 49 11 37
Daily 32 95 32 119 31 107
Know PLHIV personally 26 78 27 99 28 98 0.63
Instrumental stigmaa
Worried about infection at work 78 237 72 264 51 175 60.55***
Worried about infection outside of work 17 52 26 96 27 94 10.09**
Blameb 50 153 70 259 83 284 79.72***

Mean SD Mean SD Mean SD Fd

Negative feelings towards PLHIV, anchored ( 100 to 100)c 4A 26 11B 31 13B 39 7.22***
Perceived community stigma norms (03) 2.2A 0.6 2.3A 0.5 2.5B 0.5 20.54***
Symbolic stigma score (04) 1.7A 1.4 2.3B 1.1 2.2B 1.1 20.27***
Transmission misconceptions (04) 0.35A 0.76 0.56B 0.80 0.78C 1.08 18.10***
Transmission knowledge: items correct (015) 11.7A 1.6 11.5A 1.5 11.4A 1.8 1.73
a
Proportion of participants answering a little bit to very worried.
b
Proportion of participants who (strongly) agreed with the statement People who got HIV from sex/drugs got what they deserved.
c
Anchored: PLHIV rating subtracted from own-gender rating, so scores B0 correspond to positive feelings, and scores0 to negative feelings
towards PLHIV.
d
Means with different subscripts differ significantly (p B0.05) from each other (Bonferroni post-hoc pairwise comparisons).
$
p B0.10; *p B0.05; **p B0.01; ***p B0.001.

doctors, to 83% of nurses and to 88% of ward staff refuse or take additional precautions before performing a
(pB0.001). Significantly more doctors (13%) than nurses routine physical examination. Similar responses were given
and ward staff (both 5%, pB0.001) were in agreement with by 40% of nurses before dispensing medication and 71%
the statement that health care workers should have the of ward staff before bathing a PLHIV. More than half of
right to refuse to treat PLHIV. The groups differed more the doctors and ward staff and 39% of the nurses reported
widely in their endorsement of restricting PLHIVs rights to discriminatory intent in both situations. Only 10 to 19% of
marry and have children. Slightly over half of the doctors did participants reported no intent to discriminate in any
not think that HIV-positive women should be allowed to have professional situation.
children, compared to more than three quarters of both At least half of the participants in all subsamples said they
nurses and ward staff (pB0.001). Forty-one percent of would stop attending, or demand extra precautions if they
doctors agreed that HIV-positive men should be denied the were patients in clinics that served PLHIV. This proportion
right to marry, as did 77% of nurses and 88% of ward staff was higher among doctors (59%) and nurses (61%) than
(pB0.001). Similar proportions held the same opinion among ward staff (50%, p B0.01). But more ward staff (61%)
regarding HIV-positive women and marriage (37, 73 and and nurses (56%) than doctors (34%) agreed with the
86%, respectively, p B0.001). There were no gender differ- statement that PLHIV should be treated in separate clinics
ences with respect to endorsement of coercive policies (pB0.001), and stated that they would not seek services
among any healthcare provider group. from an HIV-positive healthcare provider (36, 31 and 23%,
respectively; pB0.01). Gender differences were found only
Intent to discriminate among ward staff participants, with 77% of male ward staff
A large majority of participants responded that they would expressing intent to discriminate if they had to bathe an HIV-
either refuse to perform, avoid physical contact or use more positive patient, vs. 65% of female ward staff (p 0.01).
than standard precautions if they were asked to treat an HIV- When asked what they would do if an HIV-infected child
positive patient (see Table 3). This included examining an attended their childs school, somewhat fewer participants 
open wound (89% of doctors), drawing blood (88% of nurses) 15% of doctors, 22% of nurses, and 32% of ward staff
or changing blood-stained linens (73% of ward staff). (pB0.001)  showed discriminatory intent. In line with
When asked about professional behaviours with a low risk results regarding misconceptions, about half of the partici-
of fluid contact, 57% of doctors stated that they would either pants stated that they would not eat from a plate used by

5
Ekstrand ML et al. Journal of the International AIDS Society 2013, 16(Suppl 2):18717
http://www.jiasociety.org/index.php/jias/article/view/18717 | http://dx.doi.org/10.7448/IAS.16.3.18717

Table 3. Endorsement of coercive policies and avoidance intentions towards PLHIV

Doctors Nurses Ward staff

(n 305) (n 369) (n 346)

Individual items % n % n % n x2

Coercive policies
Mandatory testing for FSW 94 287 97 358 97 337 5.38$
Mandatory testing for surgery patients 90 274 99 366 96 332 29.85***
Mandatory testing for surgery staff 73 223 83 305 88 302 22.27***
HIV-positive women banned from having children 55 168 76 279 80 275 53.98***
HIV-positive men should not be allowed to marry 41 124 77 283 88 306 186.16***
HIV-positive women not be allowed to marry 37 112 73 269 86 296 182.98***
HCW should not have to treat PLHIV 13 39 5 19 5 17 18.87***
Intent to discriminate: professional
High likelihood of contact w/bodily fluidsa 89 272 88 324 73 252
Low likelihood of contact w/bodily fluidsb 57 174 40 146 71 243
Intent to discriminate: personal
Change clinic or demand extra precautions if PLHIV were treated where you get care: 59 179 61 224 50 173 9.75**
Change school or avoid HIV-positive child if HIV-infected child in your childs school: 15 46 22 82 32 112 26.58***
Would not eat from plate used by PLHIV 42 128 53 193 56 195 13.62**
PLHIV should be treated in separate clinics 34 103 56 205 61 212 54.10***
Not comfortable feeding PLHIV by hand 33 98 27 101 21 72 11.04**
Not seek services from HIV-positive HCW 23 71 31 114 36 124 12.21**
a
High likelihood of contact w/bodily fluids: doctors: examine open wound; nurses: draw blood; ward staff: change blood-stained linens of PLHIV;
no between-group comparisons done due to different items.
b
Low likelihood of contact w/bodily fluids: doctors: routine physical exam; nurses: dispense medication; ward staff: bathe PLHIV; no between-
group comparisons done due to different items.
$
p B0.10; *p B0.05; **p B0.01; ***p B0.001.

a PLHIV and about a quarter would not feel comfortable of a PLHIV (AOR1.94; 95% CI 1.322.98). More trans-
feeding a PLHIV by hand. The former was most common mission misconceptions was also associated with higher odds
among ward staff, the latter among doctors, with the pro- of discrimination during a routine examination (AOR 1.51;
portion of nurses in between for both items (both p B0.01). 95% CI1.032.32). Gender was not associated with drivers
On average, doctors endorsed fewer personal discrimination of stigma or intent to discriminate among doctors.
intentions (mean 2.1) than nurses (mean2.7) and ward The bivariate correlations and multivariate regression
staff (mean 2.9, p B0.001). models in Table 5 show that nurses with lower levels of HIV
transmission knowledge had significantly higher mean levels
Drivers of stigma of endorsement of coercive policies (b 0.13, p0.022),
Results from the bivariate and final multivariate regression younger age (b 0.10, p0.051), and higher mean levels of
models are presented in Tables 46. Table 4 shows that negative feelings towards PLHIV (b 0.10, p0.050). Intent
doctors with greater instrumental stigma at work (b0.24, to discriminate in personal life was significantly related
p B0.001) and who did not know a PLHIV personally (b0.13, to nurses being non-Hindu (b 0.10, p 0.036), having
p 0.033) reported higher endorsement of coercive policies higher levels of negative feelings towards PLHIV (b0.12,
than did doctors with lower instrumental stigma and those p 0.013), of work and non-work instrumental stigma
with a personal acquaintance with PLHIV. Instrumental stigma (work: b 0.23, p B0.001; non-work: b0.11, p0.029),
at work was also significantly related to higher intent to and of perceived stigmatizing community norms (b0.12,
discriminate in personal situations (b0.19, p0.001), p 0.010). Finally, nurses with more misconceptions
as were higher levels of negative feelings towards PLHIV (b0.20, p B0.001) and less transmission knowledge
(b0.13, p 0.019), blame (b 0.11, p 0.044), trans- (b0.19, pB0.001) also had significantly higher levels of
mission misconceptions (b 0.36, pB0.001), perceived discriminatory intent in personal situations. In both profes-
stigmatizing community norms (b0.11, p0.038), and sional situations, nurses intent to discriminate was signifi-
less frequent professional contact with PLHIV (b 0.11, cantly related to higher levels of instrumental stigma at work
p 0.047). Those with less frequent professional contact (medication: AOR1.37; 95% CI 1.081.73; draw blood:
also had higher odds of showing discriminatory behaviour AOR1.56; 95% CI1.092.30), but the two outcomes
while performing a routine medical examination of a PLHIV varied in their relation to other correlates. Unmarried nurses
(AOR1.35; 95% CI 1.081.70) or dressing an open wound (AOR1.76; 95% CI1.082.88) and those with lower

6
Ekstrand ML et al. Journal of the International AIDS Society 2013, 16(Suppl 2):18717
http://www.jiasociety.org/index.php/jias/article/view/18717 | http://dx.doi.org/10.7448/IAS.16.3.18717

Table 4. Bivariate and multivariate associations with out- Table 4 (Continued )

comes, for doctors
Multivariatea
a
Multivariate Bivariate
Bivariate Pearson r b sig.
Pearson r b sig.
$
Higher income 0.11
Outcome: endorsement of (n 271, R2 0.11) (log-transformed)
coercive policies More negative feelings 0.17**
Younger age 0.09% towards PLHIV
Higher income 0.12$ More work-related 0.16** 1.28$ (0.961.71)
(log-transformed) instrumental stigma
More negative feelings 0.12* More non-work instrumental 0.08%
towards PLHIV stigma
More blame 0.17** More transmission 0.12* 1.51* (1.032.32)
More work-related 0.26*** 0.241 .000 misconceptions (4 items)
instrumental stigma Lower transmission knowledge 0.10$
%
More non-work instrumental 0.07 (15 items)
stigma Less frequent professional 0.11$ 1.35* (1.081.70)
More transmission 0.13* 0.108 .071 contact w/PLHIV
misconceptions (4 items) More symbolic stigma 0.14*
Lower transmission knowledge 0.13 More stigmatizing perceived 0.12* 0.64$ (0.371.07)
(15 items) community norms
Less frequent professional 0.14*
Outcome: intent to discriminate, professional: (n 270)
contact w/PLHIV
open wound
Not knowing any PLHIV 0.13* 0.125 .033
Non-Hindu religion 0.08%
personally
Higher income 0.07%
More symbolic stigma 0.18** 0.120 .077
(log-transformed)
2
Outcome: intent to discriminate, (n 265; R 0.32) More negative feelings 0.12*
personal life towards PLHIV
Younger age 0.08% More blame 0.15*
Higher income 0.12$ More work-related 0.18**
(log-transformed) instrumental stigma
More negative feelings 0.22*** 0.133 0.019 Lower transmission knowledge 0.14*
towards PLHIV (15 items)
More blame 0.24*** 0.110 0.044 Less frequent professional 0.17** 1.94** (1.322.98)
More work-related 0.27*** 0.188 0.001 contact w/PLHIV
instrumental stigma Not knowing any PLHIV 0.14* 2.27$ (0.935.38)
More transmission 0.40*** 0.364 0.000 personally
misconceptions (4 items) More stigmatizing perceived 0.08%
Lower transmission knowledge 0.16** community norms
(15 items)
Less frequent professional 0.17** 0.110 0.047 Note: all models adjusted for site.
b, standardized regression coefficient; AOR, adjusted odds ratio;
contact w/PLHIV
CI, confidence interval.
Not knowing any PLHIV 0.13* a
Multivariate regression: final model, obtained via backward elimi-
personally nation starting from all variables bivariately associated at p B0.25,
More symbolic stigma 0.25*** 0.121 0.056 until all p B0.10.
%
More stigmatizing perceived 0.11* 0.113 0.038 p B0.25; $p B0.10; *p B0.05; **pB0.01; ***p B0.001.
community norms
Pearson r AOR 95% CI household income (AOR 0.32; 95% CI0.130.79) showed
higher odds of discrimination than married nurses and nurses
Outcome: intent to discriminate, professional: (n 268) with higher income, respectively, when dispensing medication
routine exam to PLHIV, while for the draw blood item, it was nurses with
Non-Hindu religion 0.12* higher household income (AOR4.33; 95% CI1.2116.12)
Unmarried 0.09% and younger nurses (AOR 1.04; 95% CI 1.011.07) who
Younger age 0.17** were more likely to express discriminatory intent. Finally, more

7
Ekstrand ML et al. Journal of the International AIDS Society 2013, 16(Suppl 2):18717
http://www.jiasociety.org/index.php/jias/article/view/18717 | http://dx.doi.org/10.7448/IAS.16.3.18717

Table 5. Bivariate and multivariate associations with outcomes, for nurses

Multivariatea
Bivariate
Pearson r b sig.

Outcome: endorsement of coercive policies (n 367; R2 0.05)

%
Younger age 0.08 0.101 .051
More negative feelings towards PLHIV 10$ 0.104 .050
More work-related instrumental stigma 0.10$
More transmission misconceptions (4 items) 0.12* 0.097 .074
Lower transmission knowledge (15 items) 0.15** 0.127 .022
2
Outcome: intent to discriminate, personal life (n 362; R  0.26)
Non-Hindu religion 0.14** 0.101 .036
More negative feelings towards PLHIV 0.17*** 0.119 .013
More work-related instrumental stigma 0.30*** 0.234 .000
More non-work instrumental stigma 0.23*** 0.112 .029
More transmission misconceptions (4 items) 0.32*** 0.195 .000
Lower transmission knowledge (15 items) 0.29*** 0.193 .000
Less frequent professional contact PLHIV 0.08%
More stigmatizing perceived community norms 0.11* 0.124 .010

Pearson r AOR 95% CI

Outcome: intent to discriminate, professional: dispense medication (n 344)

Non-Hindu religion 0.08%
Unmarried 0.16** 1.76* (1.082.88)
Younger age 0.09$
Higher income (log-transformed) 0.14** 0.32* (0.130.79)
More negative feelings towards PLHIV 0.09$
More work-related instrumental stigma 0.15** 1.37** (1.081.73)
More non-work instrumental stigma 0.10$
More transmission misconceptions (4 items) 0.22*** 1.69*** (1.272.26)
Lower transmission knowledge (15 items) 0.13*
Less frequent professional contact PLHIV 0.11*
More symbolic stigma 0.09$ 0.77* (0.600.97)
Outcome: intent to discriminate, professional: draw blood (n 359)
Younger age 0.16** 1.04* (1.011.07)
Higher income (log-transformed) 0.11* 4.33* (1.2116.12)
More negative feelings towards PLHIV 0.13* 1.01$ (1.001.03)
More work-related instrumental stigma 0.19*** 1.56* (1.092.30)
More transmission misconceptions (4 items) 0.07$
Not knowing any PLHIV personally 0.10$ 0.47$ (0.181.07)

Note: all models adjusted for site.

b, standardized regression coefficient; AOR, adjusted odds ratio; CI, confidence interval.
a
Multivariate regression: final model, obtained via backward elimination starting from all variables bivariately associated at p B0.25, until all
p B0.10.
%
p B0.25; $p B0.10; *p B0.05; **pB0.01; ***p B0.001.

transmission misconceptions (AOR 1.69; 95% CI 1.27 feelings towards PLHIV (b 0.15, p0.006; b 0.15,
2.26) and greater symbolic stigma (AOR 0.77; 95% p 0.003, respectively), more blame (b 0.13, p 0.020;
CI0.600.97) were associated with treating PLHIV differ- b 0.12, p 0.013), more misconceptions (b 0.13,
ently when dispensing medication. p 0.014; b 0.34, p B0.001) and more symbolic stigma
The bivariate correlations and multivariate regression (b0.14, p 0.014; b 0.14, p 0.006). In addition, intent
models for ward staff are shown in Table 6. Endorsement to discriminate in personal situations also increased with
of coercive policies and intent to discriminate in personal younger age (b0.17, p B0.001), decreasing frequency
situations were both significantly related to more negative of professional contact with PLHIV (b0.11, p0.025),

8
Ekstrand ML et al. Journal of the International AIDS Society 2013, 16(Suppl 2):18717
http://www.jiasociety.org/index.php/jias/article/view/18717 | http://dx.doi.org/10.7448/IAS.16.3.18717

Table 6. Bivariate and multivariate associations with outcomes, for ward staff

Multivariatea
Bivariate
Pearson r b sig.

Outcome: endorsement of coercive policies (n318; R2 0.14)

Non-Hindu religion 0.12*
More negative feelings towards PLHIV 0.21*** 0.151 0.006
More blame 0.20*** 0.126 0.020
More work-related instrumental stigma 0.22*** 0.111 0.073
More non-work instrumental stigma 0.12*
More transmission misconceptions (4 items) 0.20*** 0.134 0.014
Less frequent professional contact w/PLHIV 0.15**
More symbolic stigma 0.12* 0.143 0.014
Outcome: intent to discriminate, personal life (n314; R2 0.37)
Younger age 0.13* 0.169 0.000
More negative feelings towards PLHIV 0.24*** 0.145 0.003
More blame 0.22*** 0.120 0.013
More work-related instrumental stigma 0.30*** 0.216 0.000
More non-work instrumental stigma 0.18***
More transmission misconceptions (4 items) 0.42*** 0.335 0.000
Lower transmission knowledge (15 items) 0.24*** 0.091 0.072
Less frequent professional contact w/PLHIV 0.22*** 0.107 0.025
More symbolic stigma 0.20*** 0.140 0.006
More stigmatizing perceived community norms 0.14**

Pearson r AOR 95% CI

Outcome: intent to discriminate, professional: bathe PLHIV (n 336)

Male gender 0.14* 1.98** (1.213.28)
Unmarried 0.06%
Younger age 0.09$
Not knowing any PLHIV personally 0.08%
More stigmatizing perceived community norms 0.08% 1.59$ (0.972.62)
Outcome: intent to discriminate, professional: change blood-stained linens (n 314)
Male gender 0.06% 1.95* (1.133.45)
Younger age 0.06%
Lower education 0.06%
More negative feelings towards PLHIV 0.14* 1.10** (1.001.02)
More work-related instrumental stigma 0.15** 1.66*** (1.282.19)
More non-work instrumental stigma 0.09$
More transmission misconceptions (4 items) 0.10$
Less frequent professional contact w/PLHIV 0.10$ 1.22* (1.001.49)
Not knowing any PLHIV personally 0.06% 0.57$ (0.291.05)
More symbolic stigma 0.11*
More stigmatizing perceived community norms 0.06%

Note: all models adjusted for site.

b, standardized regression coefficient; AOR, adjusted odds ratio; CI, confidence interval.
a
Multivariate regression: final model, obtained via backward elimination starting from all variables bivariately associated at p B0.25, until all
p B0.10.
%
p B0.25; $p B0.10; *p B0.05; **pB0.01; ***p B0.001.

and increasing levels of instrumental stigma at work (b0.22, females (AOR 1.95; 95% CI 1.133.45) to discriminate
p B0.001). Male ward staff were twice as likely (AOR 1.98; when asked to change a PLHIVs blood-stained linens. Having
95% CI 1.213.28) to discriminate when bathing patients more negative feelings towards PLHIV (AOR 1.10; 95%
than female ward staff. Males were also more likely than CI1.001.02), greater work-related instrumental stigma

9
Ekstrand ML et al. Journal of the International AIDS Society 2013, 16(Suppl 2):18717
http://www.jiasociety.org/index.php/jias/article/view/18717 | http://dx.doi.org/10.7448/IAS.16.3.18717

(AOR1.66; 95% CI1.282.19) and less frequent profes- driving both endorsement of coercive measures and intent
sional contact with PLHIV (AOR1.22; 95% CI1.001.49) to discriminate against PLHIV in personal and professional
were also associated with higher odds of discrimination in this contexts, regardless of whether the latter situations actually
situation. involve risk of fluid exposure. This is consistent with findings
from previous studies [30,31,33,34], and our previous paper
on stigma among outpatients in Mumbai and Bengaluru [38],
Discussion
suggesting that misconceptions are a consistent driver of
The results reveal disturbingly high rates of stigma attitudes
HIV stigma in India. The findings from this study thus indi-
and intent to discriminate among doctors, nurses and ward
staff in these urban healthcare settings. The rates are similar cate that stigma reduction interventions need to target
to those reported by outpatients in these settings as well as common misconceptions, even among highly educated and
to results of studies conducted in other parts of the country already trained healthcare providers. Since younger and
[1113,30,3236,38,42] and may thus represent wider less experienced nurses and ward staff were more likely
community norms. The almost universal endorsement of to discriminate, there may also be a need to ensure that
mandatory testing for FSW and surgery patients may be one they are thoroughly trained in universal precautions until
of the reasons why testing is now routinely performed in they are comfortable and confident in their ability to prevent
Indian healthcare settings for surgery patients and pregnant transmission.
women. Routine periodic testing of key populations is also The fact that more experience with PLHIV was associated
currently done in some areas. Although doctors were less with lower rates of stigma and discrimination in all three
likely (37 to 55%) than nurses (73 to 76%) and ward staff groups suggests that interventions may be more effective if
(80 to 88%) to endorse the different coercive measures in PLHIV are involved at all stages of intervention development
relation to marriage and children, their rates were and implementation to ensure sufficient and meaningful
still surprisingly high. These endorsements are of particular interactions. It might also be helpful to involve experienced
concern since they involve the denial of basic human healthcare providers, who have extensive experience treating
rights of PLHIV to enjoy marital status and parenthood, PLHIV as role models for their junior colleagues to provide
which are crucial aspects of Indian culture. These findings opportunities for observational learning, help change norms
highlight the need for a rights-based approach to address- in the workplace and to increase the likelihood of interven-
ing stigma in future regional and national intervention tion sustainability. Doctors treating PLHIV respectfully
programmes. are also likely to make an impression and set a standard
Participants also reported high rates of intent to treat HIV- for both nurses and ward staff in their institutions, given the
positive patients differently from uninfected patients, both hierarchical nature of relationships in these settings.
in situations that involved a risk of fluid exposure and in Both female doctors and female ward staff reported a
situations that are typically considered low risk. Since female greater number of transmission misconceptions than did
ward staff reported more transmission misconceptions and a their male counterparts, in spite of their very different levels
more negative view of PLHIV, the finding that male ward staff of education. This suggests that there may be differences in
were more likely to report intent to discriminate may reflect HIV-related education received by male and female students
their perception that they have more control over their job in Indian schools. It is thus important for future HIV pre-
duties than their female counterparts. This needs to be vention and stigma interventions to address basic transmis-
explored further to determine how to best address this
sion facts when targeting female participants, regardless of
gender difference in a stigma reduction intervention. It was
their level of education.
encouraging that physicians and nurses were significantly less
Similar to every study, ours has a number of limitations
likely to state that they intended to discriminate in low-
that need to be considered when interpreting its results.
risk situations; however, healthcare professionals who use
Since this study used a cross-sectional design, we are unable
universal precautions do not need to use double gloves or
to draw conclusions about causality and can only state which
avoid HIV-infected patients in order to be safe. In addition to
variables are associated. Future research is needed to
stigma, these high rates might also be indicative of lack of
confidence in standard universal measures to prevent examine these relationships in a longitudinal fashion to clarify
infection. the nature of these associations. In addition, the general-
Intent to discriminate was only slightly less in non- izability of these findings is limited to the types of healthcare
professional situations. The majority of all groups stated settings that collaborated with us in these two large urban
that they did not want to be treated in the same clinics as areas. We made every effort to recruit healthcare providers
PLHIV and more than half of the nurses and ward staff from a wide range of clinics and hospitals, in order to be
reported that they would be unwilling to eat from the same as representative as possible of healthcare settings that
plate as an infected individual. This item was endorsed by are accessible to patients of all socioeconomic backgrounds.
42% of the doctors also. However, our sample did not include healthcare providers
Although there are minor variations, the drivers of stigma in non-allopathic institutions. We are also limited by our
and discrimination appear to be fairly consistent across the reliance on self-reported measures, which may be subject to
different groups. Transmission-related fears and misconcep- social desirability biases. Additional studies using behavioural
tions, as well as limited experience working with PLHIV, observations are needed to provide data on enacted stigma
blame and negative feelings towards PLHIV seem to be in these settings.

10
Ekstrand ML et al. Journal of the International AIDS Society 2013, 16(Suppl 2):18717
http://www.jiasociety.org/index.php/jias/article/view/18717 | http://dx.doi.org/10.7448/IAS.16.3.18717

prevention of mother-to-child transmission of HIV: issues and findings from

Conclusions community research in Botswana and Zambia. Washington, DC: Interventional
The high rates of stigma and discrimination among health- Center for Research on Women; 2000.
care providers in these urban Indian healthcare settings 9. Madiba S. The impact of fear, secrecy, and stigma on parental disclosure
appear to be driven primarily by negative feelings towards of HIV status to children: a qualitative exploration with HIV positive parents
PLHIV, lack of experience as well as misconceptions and fear attending an ART clinic in South Africa. Glob J Health Sci. 2013;5(2):4961.
10. Calin T, Green J, Hetherton J, Brook G. Disclosure of HIV among black
of casual transmission. Stigma reduction interventions are
African men and women attending a London HIV clinic. AIDS Care. 2007;
thus urgently needed to target transmission misconceptions 19(3):38591.
and to increase interactions with PLHIV. Such programmes 11. Chandra PS, Deepthivarma S, Manjula V. Disclosure of HIV infection in
need to be designed and implemented in collaboration with south India: patterns, reasons and reactions. AIDS Care. 2003;15(2):20715.
PLHIV networks and use a rights-based and gender-sensitive 12. Kumarasamy N, Safren SA, Raminani SR, Pickard R, James R, Krishnan AK,
approach. In order to be both effective and sustainable, et al. Barriers and facilitators to antiretroviral medication adherence among
patients with HIV in Chennai, India: a qualitative study. AIDS Patient Care STDS.
interventions should ideally make use of professional role
2005;19(8):52637.
models and be integrated into existing training structures in 13. Steward WT, Herek GM, Ramakrishna J, Bharat S, Chandy S, Wrubel J, et al.
hospital clinics and the curricula in nursing and medical HIV-related stigma: adapting a theoretical framework for use in India. Soc Sci
schools. Med. 2008;67(8):122535.
14. Nyblade L, Singh S, Ashburn K, Brady L, Olenja J. Once I begin to
Authors affiliations participate, people will run away from me: understanding stigma as a barrier
1
Center for AIDS Prevention Studies, University of California, San Francisco, to HIV vaccine research participation in Kenya. Vaccine. 2011;29(48):89248.
CA, USA; 2St. Johns Research Institute, Bengaluru, Karnataka, India; 3School of 15. Bharat S, Aggleton P, Tyrer P. India: HIV and AIDS-related discrimination,
Public Health, University of California, Berkeley, CA, USA; 4Department of stigma and denial. Geneva: UNAIDS; 2001. ISBN 92-9173-104-8.
Health Education, National Institute of Mental Health and Neurosciences, 16. Kinsler JJ, Wong MD, Sayles JN, Davis C, Cunningham WE. The effect of
Bengaluru, Karnataka, India; 5Centre for Health and Social Sciences, School of perceived stigma from a health care provider on access to care among a low-
Health Systems Studies, Tata Institute of Social Sciences, Mumbai, Maharastra, income HIV-positive population. AIDS Patient Care STDS. 2007;21(8):58492.
India 17. Morrison SD, Banushi VH, Sarnquist C, Gashi VH, Osterberg L, Maldonado
Y, et al. Barriers to care and current medical and social needs of HIV-positive
Competing interests patients in Albania. Cent Eur J Public Health. 2011;19(2):917.
The authors declare that they have no competing interests. 18. Paterson DL, Swindells S, Mohr J, Brester M, Vergis EN, Squier C, et al.
Adherence to protease inhibitor therapy and outcomes in patients with HIV
Authors contributions infection. Ann Intern Med. 2000;133(1):2130.
ME, SB and JR collaborated on the design of the study as well as the collection 19. Bangsberg DR, Hecht FM, Charlebois ED, Zolopa AR, Holodniy M, Sheiner L,
and interpretation of data. EH conducted the data analyses presented in this et al. Adherence to protease inhibitors, HIV-1 viral load, and development of
article and prepared the tables. ME took the lead on writing the first draft of drug resistance in an indigent population. AIDS. 2000;14(4):35766.
this article, with assistance by EH. SB and JR read the article and provided 20. Arnsten JH, Demas PA, Farzadegan H, Grant RW, Gourevitch MN, Chang CJ,
critical feedback. All authors read and approved the final article. et al. Antiretroviral therapy adherence and viral suppression in HIV-infected
drug users: comparison of self-report and electronic monitoring. Clin Infect Dis.
Acknowledgements
2001;33(8):141723.
This work was supported by the John E. Fogarty International Center for
21. Bangsberg DR, Perry S, Charlebois ED, Clark RA, Roberston M, Zolopa AR,
Advanced Study in the Health Sciences at the National Institutes of Health (R01
et al. Non-adherence to highly active antiretroviral therapy predicts progres-
TW006314). They also gratefully acknowledge Ms Emily Shamban for her help
sion to AIDS. AIDS. 2001;15(9):11813.
with manuscript preparation, the project field staff in Mumbai and Bengaluru
22. Ford K, Wirawan DN, Sumantera GM, Sawitri AA, Stahre M. Voluntary HIV
for their thorough and careful work, as well as the many participants who gave
so generously of their time, in spite of busy work schedules to help us better testing, disclosure, and stigma among injection drug users in Bali, Indonesia.
understand the drivers of HIV stigma in Indian healthcare settings. AIDS Educ Prev. 2004;16(6):48798.
23. Niang CI, Tapsoba P, Weiss E, Diagne M, Niang Y, Moreau AM, et al.
Its raining stones: stigma, violence and HIV vulnerability among men who
References
have sex with men in Dakar, Senegal. Cult Health Sex. 2003;5(6):499512.
1. Krishna V, Bhatti R, Chandra P, Juvva S. Unheard voices: experiences 24. Wolfe WR, Weiser SD, Leiter K, Steward WT, Percy-De Korte F, Phaladze N,
of families living with HIV/AIDS in India*. Contemp Fam Ther. 2005;27(4):
et al. The impact of universal access to antiretroviral therapy on HIV stigma in
483506.
Botswana. Am J Public Health. 2008;98(10):186571.
2. Ma W, Detels R, Feng Y, Wu Z, Shen L, Li Y, et al. Acceptance of and barriers
25. White RC, Carr R. Homosexuality and HIV/AIDS stigma in Jamaica.
to voluntary HIV counselling and testing among adults in Guizhou province,
Cult Health Sex. 2005;7(4):34759.
China. AIDS. 2007;21(Suppl 8):S12935.
26. Feyissa GT, Abebe L, Girma E, Woldie M. Stigma and discrimination against
3. Obermeyer CM, Osborn M. The utilization of testing and counseling for
people living with HIV by healthcare providers, Southwest Ethiopia. BMC Public
HIV: a review of the social and behavioral evidence. Am J Public Health.
Health. 2012;12:522.
2007;97(10):176274.
27. Banteyerga H, Kidanu A, Nyblade L, MacQuarrie K, Pande R. Yicha-laliko!
4. Nguyen TA, Oosterhoff P, Pham YN, Hardon A, Wright P. Health workers
Exploring HIV and AIDS stigma and related discrimination in Ethiopia: causes,
views on quality of prevention of mother-to-child transmission and postnatal
care for HIV-infected women and their children. Hum Resour Health. 2009; manifestations, consequences, and coping mechanisms. Addis Ababa: Miz-
7:39. Hasab Research Center; 2004.
5. Varga CA, Sherman GG, Jones SA. HIV-disclosure in the context of vertical 28. Working report measuring HIV stigma: results of a field test in Tanzania.
transmission: HIV-positive mothers in Johannesburg, South Africa. AIDS Care. Tanzania Stigma-Indicators Field Test Group: United States Agency Interna-
2006;18(8):95260. tional Development (USAID); 2005. Available from: http://www.icrw.org/files/
6. Eide M, Myhre M, Lindbaek M, Sundby J, Arimi P, Thior I. Social publications/Working-Report-Measuring-HIV-Stigma-Results-of-a-Field-Test-in-
consequences of HIV-positive womens participation in prevention of Tanzania.pdf
mother-to-child transmission programmes. Patient Educ Couns. 2006;60(2): 29. Hassan ZM, Wahsheh MA. Knowledge and attitudes of Jordanian nurses
14651. towards patients with HIV/AIDS: findings from a nationwide survey. Issues
7. Bond V, Chase E, Aggleton P. Stigma, HIV/AIDS and prevention of mother- Ment Health Nurs. 2011;32(12):77484.
to-child transmission in Zambia. Eval Program Plann. 2002;25(4):34756. 30. Vyas KJ, Patel GR, Shukla D, Mathews WC. A comparison in HIV-associated
8. Nyblade L, Field ML. Community involvement in prevention of mother- stigma among healthcare workers in urban and rural Gujarat. SAHARA J.
to-child transmission (PMTCT) initiative. Women, communities and the 2010;7(2):715.

11
Ekstrand ML et al. Journal of the International AIDS Society 2013, 16(Suppl 2):18717
http://www.jiasociety.org/index.php/jias/article/view/18717 | http://dx.doi.org/10.7448/IAS.16.3.18717

31. Mahendra VS, Gilborn L, Bharat S, Mudoi R, Gupta I, George B, et al. 40. Messer LC, Pence BW, Whetten K, Whetten R, Thielman N, ODonnell K,
Understanding and measuring AIDS-related stigma in health care settings: et al. Prevalence and predictors of HIV-related stigma among institutional- and
a developing country perspective. SAHARA J. 2007;4(2):61625. community-based caregivers of orphans and vulnerable children living in five
32. Bharat S. Facing the HIV/AIDS challenge: a study on household and less-wealthy countries. BMC Public Health. 2010;10:504.
community responses. Health Millions. 1998;24(1):1516, 19. 41. Weiss MG, Ramakrishna J, Somma D. Health-related stigma: rethinking
33. Kurien M, Thomas K, Ahuja RC, Patel A, Shyla PR, Wig N, et al. Screening concepts and interventions. Psychol Health Med. 2006;11(3):27787.
for HIV infection by health professionals in India. Natl Med J India. 2007; 42. Steward WT, Bharat S, Ramakrishna J, Heylen E, Ekstrand ML. Stigma
20(2):5966. is associated with delays in seeking care among HIV-infected people in India.
34. Kermode M, Holmes W, Langkham B, Thomas MS, Gifford S. HIV-related J Int Assoc Provid AIDS Care. 2013;12(2):1039.
knowledge, attitudes and risk perception amongst nurses, doctors and other 43. Steward WT, Chandy S, Singh G, Panicker ST, Osmand TA, Heylen E, et al.
Depression is not an inevitable outcome of disclosure avoidance: HIV stigma
healthcare workers in rural India. Indian J Med Res. 2005;122(3):25864.
and mental health in a cohort of HIV-infected individuals from Southern India.
35. Kumar R, Mohan N, Seenu V, Kumar A, Nandi M, Sarma RK. Knowledge,
Psychol Health Med. 2011;16(1):7485.
attitude and practices towards HIV among nurses in a tertiary care teaching
44. National AIDS Control Organization. Available from: http://naco.
hospital: two decades after the discovery. J Commun Dis. 2002;34(4):
gov.in/upload/Surveillance/Reports%20&%20Publication/HIV%20Sentinel%20
24556.
Surveillance%20India%20Country%20Report,%202008-09.pdf
36. Bharat S. A systematic review of HIV/AIDS-related stigma and discrimina-
45. Herek GM. Thinking about AIDS and stigma: a psychologists perspective.
tion in India: current understanding and future needs. SAHARA J. 2011; J Law Med Ethics. 2002;30(4):594607.
8(3):13849. 46. Herek GM. AIDS and stigma. Am Behav Sci. 1997;42:110212.
37. Mahendra VS, Gilborn L, George B, Samson L, Mudoi R, Jadav S, 47. Herek GM. Confronting sexual stigma and prejudice: theory and practice.
et al. Reducing AIDS-related stigma and discrimination in Indian hospitals. J Soc Issues. 2007;63:90525.
New Delhi: Population Council; 2006. 48. Herek GM, Capitanio JP. Public reactions to AIDS in the United States:
38. Ekstrand ML, Bharat S, Ramakrishna J, Heylen E. Blame, symbolic stigma a second decade of stigma. Am J Public Health. 1993;83(4):5747.
and HIV misconceptions are associated with support for coercive measures in 49. Herek GM, Mitnick L, Burris S, Chesney M, Devine P, Fullilove MT, et al.
urban India. AIDS Behav. 2012;16:70010. Workshop report: AIDS and stigma: a conceptual framework and research
39. Singh D, Chaudoir SR, Escobar MC, Kalichman S. Stigma, burden, social agenda. AIDS Public Policy J. 1998;13(1):3647.
support, and willingness to care among caregivers of PLWHA in home-based 50. Hosmer D, Lemeshow S. Applied logistic regression. New York: Wiley;
care in South Africa. AIDS Care. 2011;23(7):83945. 2000.

12