In 2007 Zanzibar included provider-initiated testing and counselling (PITC) in their HIV testing and counselling programme to diagnose new infections earlier and to increase access to persons who would not otherwise get tested. Our objective was to assess the uptake of PITC and the proportion of HIV-positive tests through PITC in Zanzibar.
Our team examined 2010-2011 programmatic data to assess how many people were tested through PITC compared to other testing approaches. We also compared the rates of HIV positivity through PITC and non-PITC programmes to assess the relative yield of case detection.
From 2010 to 2011, 6.1% of tests were recorded as PITC. PITC served relatively more females (7.6%), persons aged under 15 (24.8%) and over 35 (9.1%), and widowed (12.2%) and married (8.4%) persons, and reached relatively more persons on Pemba Island (12.2%) compared to Unguja Island (5.0%). The rate of HIV seropositivity for PITC testers was 4.1%, nearly twice the rate of non-PITC testers (2.1%, p-value<0.05). The percentage of HIV-positive tests was significantly higher for PITC compared to non-PITC for all categories of testers except clients aged 15 to 24 years compared to other age groups (1.5% vs. 1.0%, p=0.224) and clients who were widowed compared to other marital statuses (4.9% vs. 5.8%, p=0.106).
Despite challenges, PITC reaches persons who may not otherwise test for HIV. PITC may remove barriers to HIV testing for some populations and therefore result in earlier diagnosis of HIV.
The HIV epidemic in Zanzibar has often been characterised as a concentrated epidemic, though more recent evidence suggests it may be a mixed epidemic with a high burden in key populations at risk (Zanzibar Integrated HIV, Tuberculosis and Leprosy Programme, 2014). The 2011-2012 Tanzania HIV/AIDS and Malaria Indicator Survey estimated HIV prevalence to be 1.0%in Zanzibar among those aged 15-49 years (Tanzania Commission for AIDS; Zanzibar AIDS Commission; National Bureau of Statistics; Office of the Chief Government Statistician; ICF International, 2013).An integrated bio-behavioural survey (IBBS) among key populations in 2007 found 16.0%, 10.8%, and 12.3%HIV prevalence for people who inject drugs (PWID), commercial sex workers (CSW), and men who have sex with men (MSM), respectively(Zanzibar Integrated HIV, Tuberculosis and Leprosy Programme, 2007).While the overall prevalence of 1.0%is lower than mainland Tanzania (5.3%), pockets of high transmission among groups at risk make continued efforts to identify and initiate infected individuals onto treatment critically important.
Knowledge of HIV serostatus is critical for prevention of HIV transmission. HIV testing and counselling(HTC) is an evidence-based intervention to increase serostatus awareness, decrease high–risk behaviour, and decrease transmission. Most importantly, it serves as the entry point into care and treatment programmes. HTC must be accessible to all people in settings with a generalised HIV epidemic, butin settings with a concentrated epidemic, HTC must be accessibletothose with high-risk behaviours who are more likely to test HIV positive(Bock et al., 2008).
While voluntary counselling and testing (VCT) remains an effective way for people to learn their HIV status, it has the potential to miss people who are either unaware of VCT services or unable or unwilling to seek out an HIV test voluntarily(Koo, Begier, Henn, Sepkowitz, & Kellerman, 2006).With limited resources for HIV prevention globally, recent evidence promotes provider-initiated testing and counselling (PITC) as an additional means of capturing previously undiagnosed HIV infections as well as increasing access to persons who would not test otherwise(Bock et al., 2008; Koo et al., 2006; Steen et al., 2007; Wall et al., 2012).
Zanzibar has been implementing HTC services since 1988. In an effort to find cost efficiencies, maximize the yield of the HTC programme, and diagnose HIV infections early, the Ministry of Health in Zanzibar through the Zanzibar AIDS Control Programme introduced PITC services in 2007 as another HTC approach.This initiative aimed toincrease access to persons who would not otherwise get tested and to improve efficiency of the HTC programme by detecting more individuals who are HIV-infected. As of January 2012,HTCwas offered at56 facilities in Zanzibar; PITC specifically was available at eight government hospitals and 30 primary health care units.
PITC makes HTC a comprehensive part of medical care and is offered to all patients attending outpatient services, including tuberculosisand sexually transmitted infection (STI) clinics, inpatient wards, and other settings within the health facility, regardlessof whether they show signs and symptoms of HIV infection(WHO Department of HIV/AIDS, Joint United Nations Program on HIV/AIDS, 2007). We have examined two years of data since the roll-out of PITC services to determinethe demographics of people reached through PITC as well as whether it is an efficient means of detecting HIV infections in Zanzibar.
To assess the uptake of PITC and the proportion of people who test HIV-positivethrough PITC in Zanzibar from 2010-2011.
Service providers collected routine HTC programme data in paper registers at the point of service and data were aggregated and sent to the Zanzibar AIDS Control ProgrammeTesting and Counselling Unit in 2012. We examined the two most recent years (2010 and 2011) of HTC programmatic data, including the records of all HIV tests performed in Zanzibar across public, private, and non-governmental programmes to assess who was tested through PITC compared to other testing approaches. In addition, we compared the proportion of peoplewho tested positive through PITC and non-PITC programmes to assess the relative yield of case detection.
We analysed records of all HIV tests performed in public, private, and non-governmental programmes in Zanzibar from January 2010 through December 2011.
The present study is limited to the data collected from the client at the time of testing, recorded on a registry form. The registry form records basic demographic information (age, sex, marital status, residence), client category (e.g., MSM, CSW, PWID), reason for testing, whether PITC or not, and test site. History of prior testing was not documented. For each test performed, the result is recorded.
We entered data into an MS Excel database and used Stata 12.0 (College Station, Texas) to clean and analyse the data. In the analysis we described frequencies of demographic characteristics by seropositivityand used the chi-square test to assess if differences were statistically significant. Our analysis comprised two stages to answer two questions. In the first stage, we compared the characteristics of clients of PITC to clients of other HTCapproaches to assess which populationswere more likely to be reached through PITC than HIV-specific testing programmes. The chi-square test was used to assess whether a particular group of clients wasmore likely to test via PITC than other HTC modalities. In the second stage, we compared the proportion of those who tested positive among persons testing through PITC compared to other approaches. Chi-square was used to assess whether differences were statistically significant, which indicateswhether the relative rate of case detection is higher or lower in PITC compared to other approaches.
From 2010 to 2011, a total of 152,711 records of HTC sessions were registered, with a 40%increase in 2011 (63,396 in 2010 vs. 89,315 in 2011) (Table 1). Over the two years, 6.1%of the tests performed were recorded as PITC and PITC accounted for about 1 in 13 (7.8%) of the HIV tests performed in Zanzibar in 2011. PITC served more females than males (7.6%vs. 4.7%, respectively, p-value <0.05), and a greater proportion ofpersons under 15 and over 35 years compared to 15-24 and 25-35 years (24.8% and 9.1%vs. 4.7%and 5.1%, respectively, p-value <0.05).PITC served a greater proportion of widowed and married persons than single, divorced, or cohabitating (12.2%and 8.4%vs. 5.0%, 4.1%, and 0.0%, respectively, p-value <0.05) and reached more persons on Pemba Island compared to UngujaIsland (12.2%vs. 5.0%, respectively, p-value <0.05).
The overall proportion of HIV seropositivity for PITC testers (including tuberculosis and STI clinic patients) was 4.1%, nearly twice the seropositivity among non-PITC testers (2.1%, p-value <0.05) (Table 2). The relative yield of HIV-positive tests was significantly higher for PITC compared to non-PITC for most categories of testers. The exceptions, in which PITC did not detect a higher rate of HIV-positive clients than non-PITC approaches, were for those under 15 years (3.1% vs. 5.8%, p-value <0.05) and widowed clients (4.9%vs. 5.8%, p-value <0.05). The HIV seropositivity rates for PWID, MSM, and CSWweresignificantly higher among the few persons in these categories tested through PITC compared to other testing approaches.
Our study results show that PITC is being widely used in Zanzibar and is reaching people who may not be diagnosed otherwise. We determined that several groups were more likely to be tested through PITC compared to other testing approaches. For example, women and children under age 15 were more likely than men and other age groups to be tested through PITC. This finding is consistent with the concept that women are more likely to access primary health care services than men.Additionally, women and children in Zanzibar are more likely to attend health care services together when a child falls ill and is accompanied by his/her mother, or vice versa. We theorise that this interaction results in more PITC exposure for both women and children.
We also found that PITC achieved a relatively higher uptake on the island of Pemba where it reached one in eight HIV tests performed, compared to the island of Unguja,where it was one in twenty. First, this may be because Unguja is larger in population and offers more alternatives for HIV testing,while Pemba is smaller with fewer options. Second, stigmatization from seeking an HIV test may play a greater role in a smaller community such as Pemba. PITC may provide a “cover” for getting an HIV test since it is done in the context of other health care services. To the extent that this is true, the routine provision of HIV testing through PITC has the advantage of reducing such a barrier in places where seeking an HIV test carries high stigma. Botswana, the first African country to roll out routine HIV testing, had similar successes in scale-up of testing as a result (Steen et al., 2007).
On the other hand, PITC does not appear to be a strategy to increase the testing of key populations such as MSM, PWID, and CSW. PITC is primarily conducted atgovernment inpatient and outpatient hospital settings and clinics where persons in these hidden populations may not disclose their risk behaviours. Therefore, the number of key population members tested through PITC may be underestimated. In addition, the modality of PITC does not allow time for extensive counselling or the opportunity for the client to disclose and discuss their HIV-related risk behaviours. Rather, only brief, general information is provided. However, there is a dearth of evidence showing the efficacy of pre-test counselling (Koo et al., 2006), so this challenge should not be viewed as an obstacle to the expansion of PITC without further evidence. Key population testing is usually conducted through mobile testing, VCT sites, and outreach efforts in communitiestoallow for more extensive, individualized counselling that reveals and addresses specific risk behaviours.
We found higher rates of HIV detection within PITC compared to other HIV testing and counselling approaches. This finding may reflect the fact that persons with health problems who are already seeking care may be more likely to have HIV, which is particularly true of persons with suspected tuberculosis and STIs. It is important to point out, however, that while the rate of HIV detection was higher for PITC, the vast majority of HIV diagnoses were still made through other testing approaches. It remains to be seen whether further expansion of PITC will result in a larger number of new diagnoses compared to other testing approaches (Koo et al., 2006)
The studyhas several limitations. First, thedata were programmatic and were not designed to answera specific research question. We therefore have few variables by which to characterizethestudy population. In addition,there are no data on the quality of the services provided or feedback on the satisfaction with the services by the population served. In addition to limitations in the data, the PITC programme itself faces many challenges. Limited resources, both material and human, and high rate of turnover of trained staff restrict the quantity and quality of HTCwithin the PITC modality.
Despite challenges, we found that PITC is being used in Zanzibar and is reaching persons who may not otherwise be tested and diagnosed. Such barriers and delays may result in later diagnosis with accompanying increases in morbidity and mortality. However, our findings indicate that PITC may not be the solution for all groups, and targeted efforts such as peer outreach continue to be needed to reach disenfranchised groups who are often at higher risk for infection and who may attend other health services.
We are extremely grateful to all the health care providers who work to serve the people of Zanzibar, as well as those who performed data collection, without which this study would not have been possible. We also acknowledge UCSF for their support and guidance in the manuscript development process.
Data included in this article were part of routine datacollection activities and did not include any personalidentifying information. In addition, ZIHTLP StrategicInformation manages this data and did not requireexternal approval to analyze it. Therefore, the studyprotocol did not require review by an institutional reviewboard.
T Ali, T Amour and SO conceived and designed the study. T Ali, T Amour and SO carried out the data analysis. T Ali, T Amour and SO drafted the manuscript. All authors read, revised and approved the final manuscript.
Conflicts of interest
The authors declare that they have no competing interests.
Bock, N. N., Nadol, P., Rogers, M., Fenley, M. A., Moore, J., & Miller, B. (2008). Provider-initiated HIV testing and counseling in TB clinical settings: tools for program implementation. International Journal of Tuberculosis and Lung Disease,12, 69-72. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/18302826.
Koo, D. J., Begier, E. M., Henn, M. H., Sepkowitz, K. A., & Kellerman, S. E. (2006). HIV Counseling and Testing: Less Targeting, More Testing. American Journal of Public Health, 96, 962-964. doi:10.2105/AJPH.2006.089235
Steen, T. W., Seipone, K., Gomez, F. L., Anderson, M. G., Kejelepula, M., Keapoletswe, K., et al. (2007). Two and a half years of routine HIV testing in Botswana. Journal of Acquired Immune Deficiency Syndrome,44, 484-488.doi: 10.1097/QAI.0b013e318030ffa9
Tanzania Commission for AIDS (TACAIDS); Zanzibar AIDS Commission (ZAC); National Bureau of Statistics (NBS); Office of the Chief Government Statistician (OCGS); ICF International. (2013). Tanzania HIV/AIDS and Malaria Indicator Survey 2011/12.Retrieved from https://dhsprogram.com/pubs/pdf/AIS11/AIS11.pdf
Wall, K. M., Kilembe, W., Nizam, A., Vwalika, C., Kautzman, M., Chomba, E., Tichacek A,…Allen, S.(2012). Promotion of couples’ voluntary HIV counselling and testing in Lusaka, Zambia by influence network and agents. BMJ Open, 2, 221-227. doi: 10.1136/bmjopen-2012-001171
WHO Department of HIV/AIDS, Joint United Nations Program on HIV/AIDS. (2007). Guidance on Provider Initiated HIV Testing and Counseling in Health Facilities. Geneva, Switzerland: WHO, UNAIDS.
Zanzibar AIDS Control Programme. (2007). Integrated Behavioural and Biological Surveillance Survey among Most at Risk Populations in Zanzibar.(Unpublished Report). Zanzibar AIDS Control Programme, Zanzibar.
Zanzibar Integrated HIV, Tuberculosis and Leprosy Programme. (2014). Epidemiological Profile.(Unpublished Report). Zanzibar Integrated HIV, Tuberculosis and Leprosy Programme, Zanzibar.
All inquiries can be sent via email to: email@example.com
IN COLLABORATION WITH
The Ministry of Health, Community Development, Gender, Elderly and Children.