Living with HIV in Karnataka

HIV in Karnataka...

  HIV Sentinel Surveillance Karnataka is one of four large southern India states with a relatively advanced HIV epidemic, with the adult HIV prevalence in several districts exceeding one percent for the past nine years. As a part of the National AIDS Control Program (NACP) of the National AIDS Control Organization (NACO), Karnataka has been conducting the HIV Sentinel Surveillance since 1998. Surveillance is carried out annually by testing for HIV at designated sentinel sites. Testing for infection is conducted among populations at higher risk, represented by patients at sexually transmitted disease (STD) clinics, intravenous drug users (IDUs) who often share needles, female sex workers (FSWs), and men who have sex with men (MSM). Populations at low risk are represented by women attending antenatal clinics (ANCs). Pregnant women attending antenatal clinics are assumed to have the same risk of sexual transmission of HIV as any other sexually active general population. The prevalence among ANC attendees in 2007 is 0.9 percent. This is close to the NFHS III estimate of 0.69 percent in 2006.

HIV prevalence among the ANC attendees indicates a decline in adult HIV prevalence in the state, from 1.5 percent in 2003 and 2004 to less than one percent (0.9 percent) in 2007. HIV prevalence among STD clinic attendees, female sex workers and MSM is approximately ten fold higher than in general population indicating HIV transmission in localized sexual networks of highly vulnerable populations including FSWs, their clients, MSM and transgender groups.


The number of districts with one percent or greater proportion of ANC attendees tested positive has declined from 20 in 2003 to 11 in 2007. In the following five districts, there has been a substantial decline in percentage HIV positive among the ANC attendees between 2006 and 2007: Bagalkot (from 2.1 percent to 0.6 percent), Bellary (1.5 percent to 0.4 percent), Uttara Kannada and Raichur (1.4 percent to 0.5 percent) and Hassan (2.4 percent to 1.3 percent). In district Belgaum, HIV prevalence has almost halved during 2003-2007, from a high of 4.5percent to 2 percent. On the other hand, the following 3 districts have shown substantial increase in HIV prevalence between 2006 and 2007: Gulbarga (from 0.9 percent to 2.9 percent), Chikamagalur (1.5 percent to 2.4 percent), and Bidar (0.9 percent to 1.1 percent). While this is the first time that Bidar district has shown greater than 1percent of HIV prevalence, Chikamagalur had recorded one percent prevalence in 2004. Gulbarga had an HIV prevalence of over 2percent in 2004 and 2005. In 2007, of the 27 districts, 11 districts have recorded a prevalence of 1percent or greater among the ANC attendees.


Overall in the state, there has been a sharper decline in HIV prevalence among the younger women under age 25, indicating a possible decline in the incidence of HIV in general population. Districts in Northern Karnataka (Bidar, Gulbarga, Bijapur, Bagalkot, Belgaum, Dharwad, Gadag, Koppal, Raichur, Uttara Kannada, Haveri and Bellary) tended to have a relatively higher HIV prevalence among ANC attendees. The sentinel surveillance data also show that there has been a sharper decline in the HIV prevalence among ANC attendees in First Referral Units (from 1.6 percent in 2003 to 0.6 percent in 2007) compared to those who attend District Hospitals (1.3 percent to 1.2 percent).


In 2005-06, National Family Health Survey (NFHS) III provided the HIV prevalence for India and selected states, separately for men and women. Compared with an HIV prevalence of 0.28 percent among the men and women age 15-49 in the country, Karnataka has an HIV prevalence of 0.69 percent [slightly higher prevalence among men (0.86) than in women (0.54)], second only to Andhra Pradesh among the states for which the estimate is available. The NFHS does not give the breakdown in HIV prevalence by rural/urban residence or by age for the state.


Knowledge of AIDS among ever-married women in Karnataka increased from 58percent in NFHS-2 to 85 percent in NFHS III. Only 35 percent of the women and 74 percent of men reported condom use as one of the methods of preventing HIV. Only 35 percent of the women and 46 percent of men REJECT all the three misconceptions about modes of transmission of HIV – that HIV is transmitted through mosquito bites, by sharing food with person who has AIDS, and by hugging someone with AIDS. NFHS III classified persons as having comprehensive knowledge if they know:

(1) That using a condom and having just one uninfected partner limits the risk of getting AIDS AND (2) That a healthy looking person can have AIDS, and

(3) Reject the two most common misconceptions about AIDS transmission (transmission by mosquito bites and by sharing food). Only 12 percent of the women and 29 percent of men in the state have a comprehensive knowledge about HIV/AIDS. The knowledge about mother to child transmission of HIV is also low in the state.


The NFHS III also revealed that only nine percent of women and five percent of men in Karnataka have ever been tested for HIV and the groups that are most likely to have been tested include Professionals (23percent of women and 14percent of men), Education 12+ (21 percent of women and 12 percent of men), Women not living with their husbands (18percent), and those in the highest wealth quintile (17percent of women and 10percent of men). Almost all those tested for HIV received the results of the test.

In NFHS III, overall, 0.3 percent of the men age 15-49 reported to have ever paid for sexual intercourse in last 12 months, and only f such men reported to have used a condom at last paid sex. Among never married men age 15-49 that had sexual intercourse in the past 12 months, 14 percent had more than one partner. However, only 35percent of never married men who had higher risk sex in the past 12 months used a condom the last time they had sex

As a part of monitoring and evaluation of the AVAHAN supported Sankalp project in the state, KHPT has conducted Integrated Biological and Behavioural Assessments (IBBA) among FSWs, clients of FSWs, MSM-T and ANC attendees in 5 districts, (Belgaum, Bellary, Bangalore Urban, Shimoga and Mysore) selected to represent different socio-cultural regions of the state.

Overall, of the 2,312 FSWs tested under IBBA from the five districts, 17 percent were found HIV positive, and this proportion in the HIV sentinel surveillance for 2007 was 6 percent. HIV prevalence among FSWs ranges from 10 percent in Shimoga district to 34 percent in Belgaum. The FSWs in Mysore have an HIV prevalence of 25 percent. Syphilis, both active and high-titre is relatively higher in urban centres such as Bangalore Urban and Mysore.

As per the IBBA among clients of FSWs in 4 districts, HIV prevalence among clients of FSWs is about one-fourth of the prevalence among the FSWs (4 percent compared with 17 percent). Clients of FSWs have a prevalence of 6 percent each in Belgaum and Bellary districts. An analysis of determinants of HIV prevalence among FSWs suggests the following:

• Younger the age of the FSW, greater is the chance that she is HIV positive, and hence it is important to cover all the young FSWs in the TIs.

• Newer the FSW to sex work, greater is the chance of her being infected with HIV. Hence it is important to reach out to the FSWs early in their sex work.

• FSWs who are non-localites (who come from outside for sex work) are more likely to be HIV positive.

• FSWs in brothels/lodges/dabhas are more likely to be HIV positive than the FSWs in public places and homes

• FSWs who have done sex work outside their usual place of residence are more likely to be HIV positive than the others

• FSWs having any STIs (syphilis, gonorrhoea, chlamydia and HSV2) are more likely to be HIV positive than the others.

The IBBA in Bangalore Urban district estimated a prevalence of 19 percent among the MSM and transgender. It was found 34 percent of the 321 respondents identified themselves as Hijras, 28 percent as Kothis, 21 percent as Double Deckers and the rest as “others” including Panthis. Overall 19 percent of the sample (and 54 percent of the self-identified Hijras) are castrated. On an average, the MSM-T surveyed had 3.4 anal sex partners in the week prior to survey, with 41 percent having three or more anal sex partners. A greater proportion of Hijras (60percent) and Kothis (34percent) had three or more anal sex partners in the past one week, compared with the DDs and others (26 and 29 percent, respectively). The use of lubricants is much uncommon compared to the use of condoms in anal sex. Overall, 87 percent of the MSM-T had ever used a condom in anal sex, compared with 36 percent who reported having ever used lubricants in anal sex. Among MSM-Ts, 85 percent reported that they used condom the last time they had anal sex. Condom use during last sexual intercourse with the main female partner was 26 percent, with “unrecognized” men it was 64 percent, and with known male regular partner it was 50 percent. The MSM-T population is as much at risk, or sometimes is at a higher risk for HIV compared to the FSWs. This risk varies considerably by sexual identities, age, current marital status, and sex work.

The IBBA among ANC attendees was conducted in a set of government and private ANC clinics in selected five districts. There were no significant differences observed in HIV prevalence by age, parity, residence (rural/urban), literacy, site (private/public), duration of sexually active life, age at first sex, ever sold sex or age and education of the spouse. However, the following 4 factors emerged as important determinants of HIV prevalence among the ANC attendees:

• Women reporting multiple partners had a significantly greater chances of being HIV positive

• women reporting an STI ever had a significantly greater chances of being HIV positive

• spouse travels for work outside the district

• spouse has multiple sexual partners

A General Population Survey (GPS) to measure HIV prevalence in general population was carried out by ICHAP in selected villages and urban areas of three talukas of Bagalkot district in 2003. Currently, as a part of the evaluation of AVAHAN programs, general population surveys collecting information on risk behaviours and prevalence of HIV and other STIs were planned in three Karnataka districts – Belgaum, Mysore and Bellary. While the data from Mysore and Belgaum are available, the data collection in Bellary is currently in progress.

Considering that the HIV sentinel surveillance among ANC attendees represents only the sexually active women and considering that the surveillance often overestimates population prevalence in moderate HIV epidemics, these general population surveys are useful at district level for community-based programs and are essential to assess the population impact of targeted interventions. These general populations cover about 6,000 randomly selected men and women age 15-49 (equal number from rural and urban areas) in each district.

Overall, five to eight percent of the unmarried men ever had sex, and six to 19 percent of the currently married men ever had an extramarital sex. The percentage of men who ever had sex with more than one partner ranges from 10 percent to 15 percent and the percentage of men who ever had a paid sex ranges from one percent to two percent. Assuming that the responses to sensitive questions on sexual behaviours will be adversely affected by social desirability biases, the responses to face-to-face interviews were validated by Polling Booth Surveys (PBS) in Mysore and Belgaum districts as a part of the GPS. As per the PBS, around 20 percent each of the unmarried men in Mysore and Belgaum reported sex. Similarly, in the PBS, around 35 percent each of the unmarried men in these two districts reported in the affirmative to ‘ever had’ extra-marital sex. GPS estimated a prevalence of 2.9 percent in Bagalkot (in 2003), 1.4 percent in Belgaum (for 2007) and 0.8 percent in Mysore (2005-06). It is important to observe that while the 2006 sentinel surveillance estimated an HIV prevalence of 3.13 percent in Belgaum, the GPS in 2007 estimated an HIV prevalence of 1.43. This difference could be because the former does not include the sexually inactive population.

The adult HIV prevalence rate in Karnataka is showing a declining trend from 1.52 in 2004 to 1.13 percent in 2006. However, the prevalence rate based on NFHS III for Karnataka in 2006 is 0.69 and is second highest in the Southern part of country following Andhra Pradesh where prevalence is 0.97. Out of 27 districts in Karnataka, 26 are ‘A’ category and one district in category ‘C’. This classification has been done on ANC data for 2004–06. To understand the epicentre of HIV in Karnataka, prevalence data for past three years (2004- 06), PPTCT sero-positivity data and high risk group prevalence data was used. It was seen that out of the 10 districts of North Karnataka, nine are the priority districts except for Bidar. In four districts of Coastal Karnataka, none come under high priority districts and in South Karnataka, out of 13 districts, three are high priority districts. To understand the high priority district, districts that had prevalence of one or more than one in all past three years were chosen on priority. Then districts that had prevalence of one or more than one in at least two years were also considered and then, for those districts, PPTCT sero-positivity was also seen. Finally, IBBA data was also used to see HRG prevalence to finalize high priority districts. Chamarajnagar was dropped from the high priority list as PPTCT sero-positivity was low.