University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Home > Global Health Literature Digest > Strong effects of home-based
Strong effects of home-based voluntary HIV counselling and testing on acceptance and equity: a cluster randomised trial in Zambia
Global Health Sciences Literature Digest
Published June 17, 2013
Journal Article

Fylkesnes K, Sandøy IF, Jürgensen M, Chipimo PJ, Mwangala S, Michelo C. Strong effects of home-based voluntary HIV counselling and testing on acceptance and equity: a cluster randomised trial in Zambia. Soc Sci Med. 2013 Jun; 86:9-16.


To compare home-based voluntary counseling and HIV testing (HB-VCT) with standard HIV testing.


Monze District, Southern Province, rural southern Zambia.

Study Design

Cluster-randomized controlled trial (RCT).


Adults (≥16 years old) residing in any of the included villages ("clusters").

Intervention (Predictor Variable)


Main Outcome Measures

Acceptance of HB-VCT, equity in the uptake of HB-VCT, and occurrence of negative life events following HB-VCT.


Thirty-six clusters were selected. Using covariates from baseline survey data (proportion of men tested for HIV, educational attainment, distance to nearest clinic and number of individuals eligible for the survey), investigators matched the clusters into 18 pairs. Within each pair, clusters were randomized to intervention or control status. To minimize cross-contamination, intervention and control clusters were always separated geographically by at least one non-study village.

All clusters received standard clinic-based VCT and routine HIV counseling. Household adults in the intervention clusters received HB-VCT (pre-test counseling, rapid HIV test using finger-prick, and post-test counseling). The intervention was provided over the course of one session in the home. Sessions averaged one hour in duration. Individuals were offered counseling in privacy. Couples were offered counseling (and if appropriate disclosure) together. The eight counselors (four male and four female) in each cluster were selected from among local lay counselors. Counselors received additional workshop training on basic counseling and HB-VCT principles, and the use of these principles when approaching households. HIV testing was done serially per Zambia's national HIV testing algorithm.

Equity was defined as "the absence of systematic disparities in health between social groups who have different levels of underlying social advantage/disadvantage."(1)

Investigators anticipated that after receiving HIV test results, there could be family conflict in some households where someone had tested HIV-positive. Counselors reported any observed family conflict to their supervisors.


The 36 clusters included 1,694 individuals eligible to complete the baseline survey (intervention arm, n=836; control arm, n=858). Among those eligible, 1,501 (88.6%) accepted the offer to be interviewed (intervention arm, n=728, 87.1%; control arm, n=773, 90.1%).

Mean participant age was 35 years (standard deviation [SD] 16.2). Most (65%) were currently married. More than half (57%) had seven or more years of education. The only statistically significant difference between the intervention and control arms overall was that fewer single (never-married) individuals in the intervention arm reported having previously received HIV testing (12.8% vs. 23%, p=0.010). There was a significant overall difference between the sexes with regard to ever having tested, with about 60% of women reporting prior testing, compared to about 39% of men. The gender difference was also significant for having been tested in the previous year, again with a higher proportion of women having been tested (intervention arm, 35.1% vs. 26.6%; control arm, 32.1% vs. 23.4%). About half of women overall reported that their most recent testing had been related to prevention of mother-to-child HIV transmission (PMTCT) efforts.

In the intervention arm, 739 (88.3%) individuals were eligible to receive HB-VCT. Of these, 686 (92.8%) were identified and invited to receive HB-VCT. During the two-month intervention period, 450 (65.6%) of 686 individuals received HB-VCT. One hundred thirty-three (19.3%) were counseled but not tested. Among couples living together, 361 (62%) of those counseled at home reported to have been counseled together with the partner. Of those couples tested at home, 315 (70%) reported to have received the test together with their partner. The remaining 156 (22.7%) of the 686 invited individuals refused counseling, were absent or did not participate for other unknown reasons. HIV prevalence in the intervention arm was 9.7% among women and 8.8% among men. HIV prevalence in the control arm was not reported.

Investigators conducted a follow-up survey eight months later among the original 1,694 survey-eligible individuals, and had 1,220 (72%) respondents (intervention arm, n=624, 74.6%; control arm, n=596, 69.4%). This included 1,120 (66.1%) individuals who had also participated in the baseline survey (intervention arm, n=565, 67.5%; control arm, n=555, 64.6%).

Using intention-to-treat (ITT) analysis, 504/836 (60.3%) participants in the intervention arm reported HIV testing in the preceding year, compared to 312/858 (36.4%) in the control arm (relative risk [RR] 1.7, 95% confidence interval [CI] 1.4 to 2.0). Of participants completing both surveys, 461/565 (81.6%) participants in the intervention arm reported having been tested in the preceding year, compared to 287/554 (51.8%) in the control arm (RR 1.57, 95% CI 1.44 to 1.72).*

In both ITT and per-protocol analyses, individuals in the intervention arm with little education were significantly more likely to have been tested in the preceding year than were their control arm counterparts. For example, among participants completing both surveys, 50 (79.4%) of 63 intervention arm participants with no formal education reported testing, compared to 13 (27.7%) of 47 in the control arm (RR 2.9, 95% CI 1.5 to 5.7). In participants with one to four years of education, 80 (76.9%) of 104 intervention arm participants reported having been tested, compared to 27 (36%) of 75 in the control arm (RR 2.1, 95% CI 1.4 to 3.3).*

Effect estimates were comparable in modified ITT analysis (with data limited to those of individuals completing at least one survey), and the differences were significant in the comparisons at all levels of reported educational attainment. These differences remained significant when PMTCT-related testing was excluded from the analysis.

With regard to the proportion of negative life events reported following HIV testing in the preceding six months, e.g. break-up of marriage or sexual relationship, physical violence, being disowned by family etc., there were no significant differences between the arms.


The authors conclude that HB-VCT was feasible in a rural setting, that there was very high acceptance of HB-VCT in individuals and couples, and that this model of service delivery has the potential to diminish social differences in the uptake of HIV testing.

Risk of Bias

The risk of bias in this trial is moderate. The sample-size calculation and cluster-randomization procedures were performed appropriately, but allocation was not concealed and the trial was not blinded. LTFU was high at 34%. Many of the results were stated in terms of percentages instead of absolute numbers.

In Context

HB-VCT for household members of HIV-infected individuals has been shown to be feasible and to have a higher uptake than clinic-based counseling and testing.(2) A previous cluster-RCT of HB-VCT was conducted in an urban setting in Zambia as part of a population-based HIV survey. This trial reported HB-VCT acceptance to be much higher among those randomized to home-based vs. clinic-based VCT (RR 4.7, 95% CI 3.62 to 6.21),(3) though this trial was found by a Cochrane review to have serious methodological flaws and to be at high risk of bias.(4) Another study found HB-VCT to be acceptable in a rural setting.(5) HB-VCT can help to overcome barriers of access to testing services, and provide testing to individuals who might not otherwise seek services.(6)

Programmatic Implications

National and subnational agencies considering to offer HB-VCT should follow national HIV testing and counseling guidelines and standard operating procedures, and take care to ensure that the minimum standards are met of informed consent, confidentiality, counseling, correct HIV test results, and linkage to prevention, care, and treatment services.


  1. Braveman, P., & Gruskin, S. (2003). Defining equity in health. Journal of Epidemiology & Community Health, 57(4), 254e258.
  2. Lugada, E., Levin, J., Abang, B., Mermin, J., Mugalanzi, E., Namara, G., et al. (2010). Comparison of home and clinic-based HIV testing among household members of persons taking antiretroviral therapy in Uganda: results from a randomized trial. Journal of Acquired Immune Deficiency Syndromes, 55(2), 245e252.
  3. Fylkesnes, K., & Siziya, S. (2004). A randomized trial on acceptability of voluntary HIV counselling and testing. Tropical Medicine & International Health, 9(5), 566e572
  4. Bateganya, M., Abdulwadud, O. A., & Kiene, S. M. (2010). Home-based HIV voluntary counselling and testing (VCT) for improving uptake of HIV testing. Cochrane Database of Systematic Reviews, 7. CD006493.
  5. Mutale, W., Michelo, C., Jürgensen, M., & Fylkesnes, K. (2010). Home-based voluntary HIV counselling and testing found highly acceptable and to reduce inequalities. BMC Public Health, 10, 347.
  6. World Health Organization (2012). Planning, implementing and monitoring home-based HIV testing and counselling: A practical handbook for Sub-Saharan Africa. [accessed 11 June 2013]