MARCH Outcome Evaluation
and non-exposed respondents might not be adequately controlled for. Fourthly, due to the nature of the survey design, the sample size for the highly-exposed group in both of the sites was relatively small; and as a result some notable positive changes (e.g. in sexual behavior) across exposure categories did not reach the desired level of statistical significance. Naturally a p-value is a function of the sample size.
With the caveat of the findings are drawn from a study that relied on a non-experimental design as well as that suffered from some of the aforementioned limitations, this evaluation, taken together, demonstrated that the overall goals of the MARCH intervention were largely met. In particular, the study revealed that the MARCH intervention produced a net increase in the knowledge of HIV/AIDS and STIs, improved interpersonal communication regarding HIV/AIDS and sexual matters, created positive attitude towards PLWHAs, increased individual’s self-efficacy in relation to condom use, and increased the perceived benefits and self-efficacy regarding HIV testing among its participants. Consistent with these changes, the intervention was also associated with an increase in the uptake of HIV blood testing in the general target audiences, a modest increase in the level of sexual abstinence as well as an increase in condom use among the never married youth.
Since the MARCH intervention has already been phased-out, it appeared tardily to provide recommendations for its improvement. However, any future effort that aims at implementing similar intervention approach in any population group in the country needs to be cautious about the fact that the impact of such intervention could not be assumed uniform across varying socio-demographics. One plausible approach to address this issue is through a program that takes proper account of the special needs and concerns of the various groups in the population, i.e. “audience segmentation”. In this respect, we suggest that future intervention efforts must consider the need for segmenting the target population at least by gender, age and marital status. It is also important to take note the fact that only moderate exposure to such intervention is not adequate to induce behavioral change in these populations. Rather, one had to regularly pass through the various peer group
Addis Ababa and West Hararghe