Prejudice and misconceptions about tuberculosis and HIV in rural and urban communities in Ethiopia: a challenge for the TB/HIV control program

The source population consisted of adults older than 14 years of age who lived in the 10 Kebeles of the GGFRA. From the source population, a total of 862 adults were proposed to be included for the quantitative survey by considering the following assumptions: prevalence of prejudice towards PLHA of 50%, 95%CI, margin of error of 3.5% and a non-response rate of 10%. The total sample size was proportionally distributed to the ten study Kebeles. In each Kebele, households were selected by simple random sampling using the unique household number. In a household, one adult person was selected randomly for the interview. To get the target study participants, households were visited repeatedly. If the respondent was not found in three visits, the next household was included to replace another respondent. Individuals younger than 15 years old or temporary residents were excluded from the study. To triangulate the findings of the quantitative survey, eight (4 among women and 4 among men) focus group discussions (FGD) were conducted. For each FGD, 8-10 individuals who were supposed to have adequate information were selected in consultation with the chairpersons of the study Kebeles.

Quantitative data were collected by trained personnel who completed grade 12. They used pretested structured questionnaire. The questionnaire included socio-demographic variables, knowledge about HIV/AIDS (15 items), prejudice or stigma towards PLHA (10 items), knowledge concerning TB (5 items), prejudice towards a patient with TB (14 items) and health care seeking behavior of the community for TB. The knowledge and prejudice questions for HIV [38, 39] and TB[20, 38] were adopted from published questionnaires. The internal consistencies of the HIV and TB prejudice scales were 0.71 and 0.83 respectively. Knowledge on the transmission of HIV was measured by 8 questions with yes/no responses: Is HIV transmitted by a mosquito/fly?, by eating raw beef (a common tradition in Ethiopia) prepared by a person who lives with HIV?, by eating together with a patients with HIV?, from a pregnant woman to her child?, by breast feeding?, by eating an uncooked egg laid by a chicken which swallowed a used condom?, by unprotected sexual practice?, and through an injury by unsterile sharp objects? Prejudice about HIV was evaluated by 10 questions: should a person living with HIV/AIDS be isolated?; are you willing to share a meal with a person living with HIV/AIDS?, to buy food from a hotel owned by a person living with HIV/AIDS?, to care for an HIV positive female relative?, to care for an HIV infected male relative?, to disclose the sero-status of HIV infected household member?, to allow an HIV infected family member to take Anti-retroviral Treatment (ART)?; do you allow a family member to marry without an HIV counseling and testing?; do you believe an HIV positive student can continue his education?; do you believe an HIV infected teacher should pursue his/her work? The TB prejudice scale included the beliefs of the respondents on the isolation of TB patients; disclosure of TB to others; perception about the social, sexual and marital problems of TB patients. Based on these questions, the degree of prejudice on HIV and TB was scored. An answer consistent with prejudice towards HIV or TB was scored with one point. An answer not consistent with prejudice towards HIV or TB was scored as zero points. A total prejudice score for HIV or TB was created by summing the scores of all questions. The HIV prejudice score ranged from 0 to 10, with the higher the score, the greater the degree of prejudice towards HIV. Individuals who had a prejudice score of equal to or greater than the mean score of the study population were categorized as having high prejudice towards HIV or TB. On the other hand, individuals who scored a prejudice score below the mean were categorized as having low prejudice towards HIV or TB. Since the prejudice scores of HIV and TB were normally distributed, the means were used to classify the study population as having high or low prejudice.

Similar scoring was done for the knowledge questions concerning HIV and TB. Qualitative data were collected by experienced public health and health education experts using FGD guides which consisted of stigma, knowledge, attitude, and health seeking behaviors of the community for TB.

Data were double entered using Epi-data version 3.1(Epi-data, Norway, 2006). For analysis, the data were exported to SPSS version 16.0 statistical software (SPSS Inc. Chicago, 2007). Descriptive analysis was done to measure knowledge, stigma and health care seeking behavior. Pearson's Chi-square was used to assess the association between socio-demographic variables and knowledge with prejudice. To control for the effect of confounding variables, stepwise logistic regression was done. Variables which had a statistically significant association (P < 0.05) in the Pearson' Chi-square test were included in the final logistic regression model. All FGD interviews were transcribed by public health and health education experts immediately after the interview. The transcribed data were commented by the investigators. After several readings, key categories & themes were identified. The data were interpreted and presented verbatim.

The proposal was approved by the ethical review committees of Jimma University, Armaur Hansen research institute and the Institute of Tropical Medicine, Belgium. Written consent was obtained from the study participants.

Of the 862 selected study participants, 750(87%) accepted to be interviewed; 112(13%) declined. Majority (70%) of those who refused to accept the interview were males. The mean age of the respondents was 31.2 (SD ± 11.0). Of the total interviewed individuals, 85.2%, 75% and 58% were Muslims, married and females respectively. Six hundred and forty five (86%) of the study participants were Oromo by ethnicity. The median monthly income of the individuals was 400 Ethiopian Birr (34 USD)/per month (Table 1).

Almost all of the study participants (97.5%) had heard of HIV/AIDS. More than half of the respondents did not know the possibility of transmission of HIV from a mother to a child or by breast milk. A larger proportion of non-literate individuals were more likely to have misconceptions about transmission of HIV than literates (Table 2).

The majority of the FGD participants believed that HIV/AIDS is a punishment from God for unacceptable human sexual behavior.

The majority of the male FGD participants described that unprotected sexual practices with multiple partners such as polygamous marriage in their village could play a major role for the transmission of HIV/AIDS. On the other hand, the majority of the female participants believed that the mosquito that transmits malaria and flies could transmit the virus. The role of flies as a transmission agent for HIV was exemplified by a 35 years old woman:

One hundred and seventy six (24%) participants had high prejudices towards PLHA. A higher proportion of females [OR = 1.5, (95% CI: 1.0, 2.2)], non-literate individuals [OR = 2.3, (95%CI: 1.4, 3.6)], rural residents [OR = 3.8, (95%CI: 2.2, 6.6)], individuals who had poor knowledge on HIV/AIDS [OR = 2.8, (95%CI: 1.8, 2.2)] and individuals with lower income [OR = 1.7,(95%CI:1.1,2.6)] had high prejudice towards PLHA than respectively males, literates, urban residents, individuals who had good knowledge and a higher income (Table 3).

HIV/AIDS was a highly stigmatized disease in the community. Almost all of the FGD discussants described that people in their villages do not want to eat, drink, share utensils or clothes and participate in a social group such as "Iddir" with PLHA. Iddir (funeral insurance) is a community based organization whereby people contribute money or food regularly to support the family of a deceased individual. The major reason of stigma towards PLHA was the fear of transmission of the virus through direct contact. To substantiate this idea, a 42 years old woman said,

The idea of isolation and discrimination of PLHA from social groups was illustrated by a 36 years old woman:

Real example of discrimination against a person who was believed to have HIV/AIDS was discussed by the female FGD participants in one of the Kebeles.

Three hundred and thirty nine (45%) and 512(70.2%) participants had heard of ART and voluntary counseling and testing. Of those who knew HIV counseling and testing, 212 (28%) were tested at least once. Six hundred and thirty four (84.5%) individuals have heard of condoms. In the 12 months prior to the survey, 521(69.5%) of the study participants had sexual intercourse, 27(5.2%) with more than one partner. Only 1.3% of individuals who had one sexual partner and 11% of those who had multiple sexual partners had utilized a condom at least once in the last one year.

A total of 708 (94.4%) participants had ever heard of TB. Literates had a much better knowledge about the cause, symptoms, and methods of transmission and prevention of TB than non- literate individuals (Table 4).

The qualitative findings revealed other causes of TB such as alcohol, Khat (natural stimulant from Catha edulis plant) and exchange of drinking and eating utensils with a TB patient. It was believed that locally made strong alcohol called 'Katikala' and Khat could cause TB through their direct toxic effect on the lung. Participants believed that TB patients should have separate eating plates and drinking cups since the microorganism could not be removed from the plate or dish by washing alone.

Three hundred and ninety nine (56%) individuals had high prejudice towards a patient with TB. A larger proportion of females and non-literate individuals had high prejudice than males and literate individuals (Table 5).

TB was considered a less stigmatized disease as compared to HIV/AIDS. On the other hand, a significant number of the FGD participants believe that TB and HIV have similar symptoms and people afraid of TB patients because of the associated HIV infection.

A total of 49(6.5%) study participants had cough of 2 weeks or more during the survey and 23(46.9%) of them did not seek help in the health institutions or other places. Of those who had cough, 34.7% visited health institutions and 12.2% went to traditional healers. TB suspects in the study area usually did nothing for a long time or visited traditional healers to take herbal medicine like Dama Kese (Qcimum sp).

"We usually drink or smell herbal medicine such as 'Dama Kese' for the treatment of 'Samba' (TB) . If the disease becomes worse or not cured by the Dama Kese, we go to health facility." A 28 years old woman at rural Kebele.

The majority of the FGD participants described that the quality of care in the nearby health posts or health centers is poor and people often do not visit them.

"People would prefer health facilities for the treatment of Samba . However, the poor quality of service in the health posts and health centers forced people to take traditional drugs like Dama Kese.", A 45 years old man in rural Kebele said.