Measuring Stigma
An example: Developing a stigma index in India

Excerpted from “Understanding and measuring AIDS-related stigma in health care settings : a developing country perspective,” Vaishali Sharma Mahendra et al., 2007, SAHARA: Journal of Social Aspects of HIV/AIDS Research Alliance 4(2): 616–625.

In India, as elsewhere, people living with HIV face stigma and discrimination in a variety of contexts. Research in India has shown that stigma and discrimination against HIV-positive people and those perceived to be infected are common in hospitals and act as barriers to their seeking and receiving critical treatment and care services.

Horizons and its partners carried out an operations research project to develop and test responses to hospital-based stigma and discrimination against people living with HIV in three New Delhi hospitals (PDF of Report or PDF of Summary).

The research team conducted formative research with 59 purposefully selected informants from the three hospitals to design the stigma index. These included doctors, nurses, ward staff, and HIV-positive patients and their caregivers. They also conducted 30 observations and held six focus group discussions.

Findings from the formative research centered on two areas: judgmental and prejudiced attitudes about people living with HIV, and stigmatizing hospital practices. These practices included testing patients for HIV without their consent, disclosing test results to relatives and other health workers without a patient’s consent, labeling patients as HIV-positive, and unwarranted use of precautions to prevent transmission such as burning bed linens and throwing away eating utensils of HIV-positive patients.

Drawing from the formative research and a literature review on stigma reduction interventions and measures, the team identified 50 items for the stigma index that were then reviewed by a team of stigma specialists in India. After much deliberation and discussion about the wording, relevance, and usefulness of the items, ten items were rejected. A revised list of 40 items was pretested with 45 respondents not connected to the intervention study. Based on this pretest, a further 8 items were removed due to lack of clarity.

A 32-item Stigma Index was used during the baseline survey of the intervention study in New Delhi hospitals. Twenty-two items were worded as statements and 11 were worded as questions. These 32 items address the main issues that emerged from the formative research, including attitudes of blame towards some groups of people living with HIV (eight items), attitudes about personal contact with people living with HIV in society (four items) and attitudes related to hospital practices and policies towards people living with HIV (nine items).

Following the baseline survey (which included 884 doctors, nurses and ward staff) and a review of the responses to the stigma items, it was determined that some questions were either unclear to respondents, or always solicited the same response, and they were removed from the index list. After a final consultation, 21 items that clustered around attitudes toward people living with HIV and attitudes toward health care related practices were retained (see 21-item Stigma Index). Items that reflected sources of stigma particular to the health care setting included informing family members of a patient’s HIV status without his or her consent, burning the linens of HIV-positive patients, and charging HIV-positive patients for the cost of infection control supplies.

Response choices to the statements on the index used a 3-point Likert scale (Agree [1], Can’t say [2], Disagree [3]) with a minimum score of 21 and a maximum total score of 63. For scoring, each of the statements was weighted as negative (stigmatizing statement) or positive (non-stigmatizing statement), and scored accordingly. For example, those who responded “agree” to a stigmatizing (or negative) statement were scored as 3 and to a non-stigmatizing (or positive) statement as 1. Similarly, those who disagreed with a positive statement were scored as 3 and to a negative statement as 1. Those respondents who were unable to agree or disagree with the statements were scored as 2. Thus, a higher score on the index denotes a higher level of stigma. Internal consistency reliability from the baseline survey was good at 0.742 (Cronbach’s alpha).

At baseline, the mean score for the entire group of health care workers (n = 884) was 42.79, with individual scores ranging from a minimum of 23 to a maximum of 61. There was a significant difference between the mean scores of the three groups of health care workers surveyed (p < .0001): ward staff had the highest mean score (47.80), followed by nurses (39.99), and then doctors (36.60). This indicates that ward staff had the most stigmatizing attitudes toward HIV-positive people.

When stigma scores were trichotomized, nearly two-thirds of all staff (64 percent) fell into the moderate stigma category and 24 percent were in the high stigma category. Only 12 percent were classified in the low category.

The stigma index scores were compared to health workers’ knowledge of HIV transmission and their practices. For example, staff who scored higher on the stigma index were more likely to report stigmatizing or discriminatory practices, such as avoiding going near HIV-positive patients, sharing the patient’s HIV status with non-treating staff or family members, or inappropriately using gloves during casual contact with HIV-positive patients. (PDF of Report or PDF of Summary)

21-Item Stigma Index
Attitudes toward people living with HIV

1	People living with HIV have a right to decide who should know about it.
2	People with HIV should still be allowed to get married, as long as both partners know about it.
3	HIV-positive women should not get pregnant.
4	HIV spreads due to immoral behavior.
5	Men who get HIV get what they deserve.
6	Sex workers are the only women who have to worry about getting HIV.
7	Men who go to sex workers or use drugs are the only men who have to worry about getting HIV.
8	A woman who gets HIV gets what she deserves.
9	Would you be willing to share a meal with an HIV-positive person?
10	If you knew that a food seller had HIV would you buy food from him?
11	Would you be willing to move into a home if the neighbor was HIV-positive?
12	If you found out that a co-worker had HIV would you be willing to work with him/her?

Attitudes toward health care-related practices

13	Patients’ blood should never be tested for HIV without their consent.
14	Patients who test positive have the right to decide whether or not their relatives should be informed.
15	When a person tests positive, the doctor should inform the patient’s partner.
16	The need for consent is exaggerated. HIV tests should be handled like any other blood test.
17	HIV/AIDS patients should be made to pay for gloves, AIDS kits, and other infection control supplies.
18	All pregnant women should be tested for HIV
19	Patients with HIV should be kept at a distance from other patients.
20	Clothes and linen used by HIV patients should be disposed of or burned.
21	Patients should be tested for HIV before surgery.