We’re building a new UNICEF.org.
As we swap out old for new, pages will be in transition. Thanks for your patience – please keep coming back to see the improvements.

Evaluation database

Evaluation report

2009 Tanzania: Evaluation of PMTCT Program in Refugee Camps in North Western Tanzania, 2003 - 2007

Author: IFAKARA Health Institute (IHI)

Executive summary


“With the aim to continuously improve transparency and use of evaluation, UNICEF Evaluation Office manages the "Global Evaluation Reports Oversight System". Within this system, an external independent company reviews and rates all evaluation reports. Please ensure that you check the quality of this evaluation report, whether it is “Outstanding”, “Good”, “Almost Satisfactory” or “Unsatisfactory” before using it. You will find the link to the quality rating below, labelled as ‘Part 2’ of the report.”


General Objective:
To evaluate the success of PMTCT program in reducing the proportion of HIV infected infants born to HIV infected mothers.

Specific Objectives:
Objective 1: To evaluate the Impact and Effectiveness of PMTCT activities in the refugee camps
The expected program impact included:
i) To reduce MTCT by 50% from baseline levels by the end of 2005
ii) To counsel 100% of antenatal mothers and test 80% of those counseled
by the end of 2005
iii) To counsel and test 50% of male partners by the end 2005

The specific indicators describing level of effectiveness in PMTCT service delivery:
a) Has the confidentiality of HIV positive mothers and children adequately been observed?
b) Are all the HIV positive clients counseled individually on infant-feeding?
c) Are both mothers and babies provided with ARV?
d) Are HIV-exposed babies provided with CTX?
e) Are HIV positive women provided with the modified obstetric care?
f) Have the sufficient measures been taken for prevention and treatment of anemia and STD?
g) Do pregnant mothers manage to start ANC early?

Objective 2: To evaluate the relevance of PMTCT activities in the refugee camps
To find out if the national guidelines on PMTCT was relevant for the PMTCT program in the refugee camps.

Objective 3: To evaluate the efficiency of PMTCT activities in the refugee camps
This part looked into implementation costs, timely availability, and human resources, planned and actual costs incurred in running the program. It is also providing an insight on training, guidelines, supplies, data management, infrastructure, partner coordination, referral services involved.

Objective 4: To evaluate whether PMTCT activities in the refugee camps have been implemented in a sustainable way whereby its impact will continue after repatriation/camp closure
This objective investigated the level of preparedness in making sure that those who repatriate and are in need of the PMTCT services can access them when they arrive in their homeland. Also, the objective explored the level of training to produce experts in different departments of PMTCT who can be pioneers in the implementation of this program in their homeland. Furthermore, the objective investigated the level of information provided to refugee clients with regards to closest PMTCT centers that can be accessed after repatriation.


Study Area and Population
Kigoma region is located on the rift valley fringe of Lake Tanganyika. Historically, Kigoma has been the city of famous explorer and missionary Dr Livingstone. The region is terminus of the railway from Dar es Salaam. Kigoma is connected by ship with Congo and Burundi and has been an entry point for refugees from Congo Rwanda as well as Burundi. The UNICEF supported PMTCT sites form part of PMTCT sites operating in the region. This evaluation was conducted in three North Western refugee camps namely, Mtabila, Nyarugusu and Lugufu.
Study Design
This evaluation was a cross sectional study which employed a triangulation of methods, both quantitative and qualitative. The approach facilitated thorough understanding of quantitative data, and allowed support of qualitative findings through quantitative outcomes.

In terms of sample size, most of the data were qualitative and desk review. As such, the number of respondents was selected to represent subgroups of interest to gain an understanding of the objectives. For client exit interviews, one of the key indicators in the line of inquiry in the evaluation was client pre-test counseling. Therefore we used this indicator for sample size calculation. From desk review findings, “the overall pre-test counseling was 92.3%”. Based on this, we determined the sample size which was needed. Actual computations were done by using WHO sample size software. In order to allow for drop-outs or refusals, we added 10% in the sample and this gave a total of 129 as the number of clients required for interview.
Data Collection
The tools included Questionnaires, Focused Group Discussion (FGD) guides, and guides for In-Depth Interviews and Physical Site Assessment. Translations of the tools from English to Kiswahili were done before pre-testing and were revised accordingly. The following tools were produced:
1. Desk Review (For PMTCT indicators)
2. In-depth / key informants tools (For PMTCT implementers and PMTCT
3. Physical site assessment tools
4. Client exit questionnaires
5. Providers questionnaires (For ANC, HCT, L & D and PNC health workers)
6. Focus Group Discussion (For Current and potential service users, Adult
female from community members and Adult male from community
Training and Pre-testing of the Survey Tools and Field Work
A total of seven research assistants were trained at IHI Dar office for two days in January 2009. Pre-testing of the instruments was done at Amtulabhai Clinic in Dar Es Salaam. The pre-testing was very useful in testing flow and relevance of the questions, wording and terminologies (whether they were understood by the community), validity and estimating the duration for each questionnaire. Significant changes were made on the questionnaires after the pre-test. These included changes on the sequence of some questions for better flow and rewording for better understanding. Some responses were restructured into “mentioned” and “not mentioned” and a few questions were added. For each district a survey team comprised of three supervisors and research assistants.
Desk review
The desk work involved review of all the PMTCT-related implementation documents available at both UNICEF Country office (Dar es Salaam) and the Zonal offices (North Western Tanzania) using developed review guides. For this review, annual descriptive reports were collected especially on all PMTCT indicators in particular year starting from 2003 to 2007. MTCT reduction rate was calculated based on UNGASS core indicator 6.
Quality Control, Data Processing and Analysis
Measures employed in the field to ensure good quality data included proper training of the field team and daily review of completed forms. As soon as completed questionnaires reached the data unit at IHI, each was assigned a unique serial number. Data was then double entered using MS Access. Analysis was done in STATA 10.0 software. Qualitative data was manually managed whereby, the tapes used to record the interviews in field were transcribed and the notes typed. The data was grouped in respective sub-subjects based on the interview guides, and then analysis was done by themes as guided by the study objectives.
Ethical Consideration
This evaluative study was reviewed and approved by the Ifakara Health Institute – Institutional Review Board (IHI-IRB). Informed Consent of participants was obtained verbally. Major areas of verbal concern included the potential emotional stress that participants may face in discussions or interviews with participants who themselves are HIV cases or have been affected indirectly. To address that challenge when it happens, the evaluation team recruited a qualified counselor on board.

Findings and Conclusions:


PMTCT Impact and Effectiveness
Key Findings:
• The PMTCT program reached it 50% MTCT reduction target in 2004 and 2005-2006 had almost close to 50% reduction rate. The overall MTCT reduction rate, from 2003 to 2007 was 41.2%.
• The camps demonstrated high frequency of health facility delivery at 77.1%, higher than the national average at 47.2%. Prevalence of HIV among pregnant women in the camps was found to be 0.9%, which is comparatively lower than the regional (Kigoma) level at 1.8% as well as the national average of more than 9.6%.
• There is outstanding knowledge of family planning methods, but females are less frequent users of these services and the decision for family planning option depend on male’s recommendations.
• There was consistent decline in prevalence of STIs among pregnant women and partners.
• Lack of alternative infant feeding was a challenge in PMTCT implementation.
• Confidentiality of HIV positive mothers and children has been adequately observed.
• All the HIV positive clients counseled individually on infant-feeding.
• The PMTCT program in the camps has constantly provided Nevirapine to both mothers and babies from 2003 to 2007.
• All HIV positive women provided with the modified obstetric care as prerequisite to all clients irrespective of the HIV sero status.
• All babies born to identifiable HIV positive women were given cotrimoxazole prophylaxis.
• Prophylaxis anaemia drugs were given to pregnant mothers during ANC visits.

PMTCT Relevance to National Guideline
Key Findings:
• PMTCT activities in the camp were implemented according to national guidelines.
• Training of staff was carried out regularly by professionals from the Ministry of Health and Social Welfare.
• Across all three camps, guidelines were displayed and were easily traced particularly in the ANC and HCT sections.
• Update of guidelines was well done whenever there was a need.
• There was frequent generation of new guidelines for the purpose of improving services and survival. However, it was deduced from the evaluation that disruptions in routines as a result of adopting new guidelines were not favorable among staff.

Efficiency of PMTCT activities in the refugee camps
Key Findings:
• Human resources were available, although turnover was very high compelling the conduct of trainings on annual basis.
• Supplies and equipments were available on demand and financial resources to execute program activities were in most cases obtained on time and according to plans.
• The national PMTCT guidelines were applied almost consistently.
• The training is carried out by professionals from the Ministry of Health and Social Welfare (MoHSW).
• There were no formal data management systems at health facilities in the camps.
• Referral systems in the all three camps followed the MoHSW guideline.
• Cost effectiveness analysis and partner coordination information could not be done due to lack of reliable data.

PMTCT Sustainability
Key Findings:
• Implementation of PMTCT activities was performed in the refugee camps by staff from both the refugee communities and Tanzanian health personnel
• Clients were provided with PMTCT repatriation packages that included supply of three months medication and information regarding access of closest PMTCT centre in their homeland.
• Political instability in Democratic Republic of Congo has been a key limitation in establishing collaborative efforts between DRC and the PMTCT refugee program in Tanzania for the purpose of the former to provide PMTCT services to returning refugees.
• Collaborative effects to provide PMTCT were very successful in Burundi. Refugees from Burundi had clear knowledge on where to access PMTCT services in their homeland.
• Training of staff (refugees) was part of the sustainability approach. This created a pool of PMTCT professionals to pioneer the implementation of PMTCT in their homeland.
• Retention of local health care providers particularly counselors has been a challenge. They may be attracted elsewhere in the labour market.


• To increase impact and effectiveness of PMTCT, the recommendations are:
o The ANC and L&D should monitor STD trends among HIV positive pregnant women as a risk factor indicator to MTCT and a proxy indicator for HIV transmission among adults
o Women should know their HIV status and should receive appropriate counseling to help them make and carry out informed infant feeding decisions
o Utilize the PMTCT provision set up in the refugee camps as a future point for care and treatment.
o PMTCT implementers and administrators should work together in developing strategies to further strengthen male partner involvement in maternal care and family support for infant feeding
o Issues around anemia and blood transfusion need to be considered as part of PMTCT monitoring since these are risk factors to MTCT
o Include community sensitization as part and parcel of the PMTCT program. Its contribution greatly affects the PMTCT outcome and overall impact. There is need for development of a sensitization action plan which aims at promoting local community support to community members living with HIV. This may help rectify issues of lack of trust, stigma and misconceptions.
• For PMTCT efficiency, the recommendations are:
o Implementing agencies should improve data management system and financial recording system. This is important in order to monitor the program systematically and effectively, which would also allow further analysis on cost-effectiveness and improvement of the program in the future.
• For PMTCT sustainability and adherence to guidelines, the recommendations are:
o Provision of enough information on the exact geographical availability of PMTCT services in the refugee home country and on how to access them after repatriation is important.
o Strategize on how to help the Congolese establish / identify PMTCT centers in their country of origin despite existing political instability.
o Record keeping is important for successful management of the PMTCT program. The PMTCT program should develop an M&E framework that incorporates all guidelines and indicators that affect the PMTCT program it should also include indicators that can measure the performance of HIT and TBAs
o The records department should develop a user friendly archiving system that makes retrieval of records easier

Full report in PDF

PDF files require Acrobat Reader.



Report information

New enhanced search