Assessment

 Health needs assessment – Malaria – Kenya

• Purpose: Defining the proportion of children treated appropriately with anti-malarial treatment for fever, and advisory practice of storekeepers, before introduction of a new intervention.
• Approach: Household survey in Kenyan coastal community with a population of 70,000 - 5,000 households visited and 1,800 interviewed; grocery store survey (100) and simulated client survey (300). Data analysed in EPI-INFO. 

Kenya Medical Research Institute (KEMRI) – Wellcome Trust Collaborative Research Programme and DFID, Kenya. 

Health needs assessment and situation analysis – Reproductive Health – Tanzania

• Purpose: To determine the reproductive health needs of men and women of reproductive age in rural, roadside and urban settings and understand the community perspectives and behaviours around sexual and reproductive health and access to car. 
• Approach: Qualitative techniques (exploratory, descriptive and contextual) including focus-group discussions (54) and in depth interviews (400) in 9 districts of Tanzania. The results were analysed using Nudist software and the results were triangulated with different data sources assessing the same topic in order to enhance the validity and reliability of the situation analysis.

AMREF and London School of Hygiene and Tropical Medicine and DFID, Tanzania

 Health needs assessment – Nutrition – Burma

• Purpose: To describe the local feeding practices and perceptions of nutrition in order to improve the district-led health promotion strategy in a displaced population in Rangoon, Burma.
• Approach: Household survey conducted with 938 parents who were systematically identified during a nutrition survey. Total population of 132 thousand. Data was analysed with EPI-INFO.

Medicins Sans Frontiers – Burma

Health needs assessment – Nutrition – Kenya

• Purpose: To gather, analyse, interpret and disseminate data of child feeding practices, local perceptions of malnutrition and parental interactions with malnourished children to inform district led health education strategies.
• Approach: Developed data collection methods with research team.  Focus group discussions (12), in-depth interviews (23) were analysed using the Framework Approach; while Observations (21) were scored to identify common behaviours.

Kenya Medical Research Institute (KEMRI)- Wellcome Trust Collaborative Research Programme and Kilifi District Hospital, Kenya

Prevalence and Behavioural survey – STI and HIV – South Africa

• Purpose: A baseline behavioural and prevalence survey was conducted in order to inform the development of a comprehensive public health intervention and in turn increase the uptake of HIV/AIDS prevention and care services.
• Approach: Target population of 40 000 residents of inner city migrant hostels and women living in informal settlements (slums) in inner city Johannesburg. An adapted household survey was conducted, random sampling of shacks and hostel units using randomising tables – 2000 men living in single-sex hostels and 1000 women living in informal settlements & shack farms (emerging slums within abandoned buildings) were sampled. Consenting adults underwent a comprehensive behavioural questionnaire and urine and saliva sampling for HIV & STI prevalence purposes was carried out. Data was analysed using STATA.

Reproductive Health and HIV Research Unit and European Union, South Africa

Prevalence and Behavioral surveys – STI and HIV – South Africa

• Purpose: A baseline behavioural and prevalence survey of long distance trucking industry to determine actual HIV prevalence rates and sexual behaviour as well as occupational hazards. In order to inform industry on ways to address health of their workforce.
• Approach: All consenting long distance truckers & assistants (2000) from 2000 randomly sampled depots (10% random sample) affiliated with the Road Freight Association were interviewed and HIV/STI tested (urine & saliva). Data was analysed using STATA.

Reproductive Health & HIV Research Unit  and National Bargaining Council for the trucking industry and USAID, South Africa

Baseline Qualitative surveys – STI/HIV – Uganda and Zambia

• Purpose: To assess community and provider perceptions of the quality of existing STI/HIV/AIDS care in both formal and traditional sectors, and attitudes towards potential collaboration between sectors.
• Approach: Study was carried out in 2 districts of Zambia and 2 districts of Uganda (districts were stratified then randomly sampled).  Using systematic sampling, in-depth interviews were conducted with sampled traditional healers (302) while all biomedical service providers at the government facilities were interviewed (299), community members (733) residing in the sampled communities were interviewed (systematic sampling) as well as focus group discussions (42) with residents (snowball sampling). Quantitative data was analysed using STATA, qualitative data was analysed thematically.

Institute of Public Health, Makerere University, Karolinska Institutet, London School of Hygiene and Tropical Medicine; Institute for Economics and Social Research (Zambia) and Traditional Healer Association of Zambia

Situation Analysis – TB and Malaria – Malawi

• Purpose: To define the health seeking behaviour amongst adults with a chronic cough, febrile children and the potential of storekeeper’s health advisory role.
• Approach: Household survey of 3,500 households visited systematically, 2,400 interviews, store survey (50), simulated client survey (250) in 3 squatter settlements in urban Lilongwe, population approximately 50,000. Data analysed in STATA.

REACH Trust, Lilongwe Malawi and DFID TB Knowledge Programme, Liverpool School of Tropical Medicine, UK

Rapid ethnographic research – Public Health – South Africa

• Purpose: To collect ethnographic insights into the target community as part of a formative phase of a larger prevalence survey and subsequent intervention to be conducted in 6 hostels and 5 informal settlements of inner city Johannesburg.
• Approach: Two residential stays of 2 weeks each were carried out in the hostels and shacks of the proposed research and intervention site – using anthropological participant observation and a modified snowball sampling method, informal ‘conversations’ were conducted with informants and residents. Conversations were recalled, transcribed and analysed using modified anthropological content analysis.

Reproductive Health and HIV Research Unit and European Union, South Africa

Situation Analysis – TB – Malawi

• Purpose: To explore and describe the potential role of retail shops in a squatter settlement to improve community access to TB diagnosis.
• Approach: Grocery shop survey of all 30 selling drugs. Storekeepers, community leaders, village market managers and lay groups selected by gender and education/role in community health. Informal discussions (11), in-depth interviews (14), focus group discussions (7), store observations (2) conducted. Data analysed with EPI-INFO and the Framework Approach.

TB Equity Project, Malawi, Liverpool School of Tropical Medicine and DFID, Malawi

Rapid appraisal of alternative sources of care for STI/HIV – South Africa

• Purpose: To assess attendance, training, regulation and supervision of all government facilities and ‘community’ recognised and practicing traditional healers and registered pharmacies in order to identify opportunities for strengthening and collaboration between these services in an effort to better serve the community residing in the immediate vicinity.  
• Approach: Repeated visits for interviews at all government clinics visits, recognised traditional healers and registered pharmacies in the intervention community. Data was analysed with emphasis on identifying cross cutting themes raised by the majority of health providers.

AMREF and LSHTM and Reproductive Health and HIV Research Unit, South Africa

External evaluation – Nutrition – Burma

• Purpose: To define the proportion of children with severe malnutrition in order to determine the impact of a nutrition feeding centre for a displaced population in Rangoon, Burma.
• Approach: A household survey to ‘height and weight’ a sample of 938 children systematically selected in a population of 132 thousand. 24 field assistants were trained and supervised. Data was analysed in EPI-NUT and EPI-INFO.

Medicins Sans Frontiers, Holland

Evaluation – Malaria – Kenya

• Purpose: To describe the impact of a rural storekeeper training intervention on home use of shop-bought anti-malarial drugs for uncomplicated childhood fevers at divisional level in coastal Kenya.
• Approach: 40 in-depth interviews, 15 case narrative, 12 focus group discussions, cost-effectiveness analysis, household survey (5,000 houses visited & 1,800 interviewed systematically), grocery store survey (50) and simulated client survey (100)  methods. Population of 70,000. Data analysed in EPI-INFO and STATA.

Kenya Medical Research Institute (KEMRI) – Wellcome Trust Collaborative Research Programme and DFID, Kenya

Situation Analysis – Nutrition – Kenya

• Purpose: To define the nutritional status of all children 0 to 5 years living in a remote island community and to develop an intervention to address their nutritional needs. 
• Approach: A household survey to ‘height and weight’ a sample of all children (100) between the ages of 0 & 5 years. All homes in the village with children between 0 to 5 years were visited and all parents of children 0 to 5 were invited to participate in the study, 97% participated. Data was analysed using the WHO’s anthropometric data analysis software.  

University of Washington and District Health Department, Kenya

Stakeholder analysis – National level Reproductive Health, STI and HIV – Tanzania

• Purpose: to engage central level players from the start of a study addressing barriers to integration of HIV and STI care into primary health services allowing them to comment on the research focus and on the research sites selected. 
• Approach: All relevant and prominent stakeholders in reproductive and HIV related services at the national level were contacted and those who agreed to participate were interviewed (52 interviews). A detailed policy review, determining perceived levels of influence on policy development and influence was conducted. All interviews were transcribed and analysed thematically.

AMREF Tanzania and London School of Hygiene and Tropical Medicine and DFID, Tanzania

Stakeholder analysis – National level – STI and HIV – Uganda and Zambia

• Purpose: To ensure implementation of a policy tracking component, in parallel to designing, implementing and evaluating a new HIV intervention.
• Approach: 51 national policy makers in two stages across Uganda and Zambia. Initial exploratory interviews were used to introduce the project; identify and collect reports on relevant policy issues; and to assess potential policy obstacles and stakeholder support for collaboration. Semi-structured follow-up interviews (20) assessed risks, challenges, possible resistance and approaches for managing stakeholders. Interviews were transcribed and analysed thematically.

Institute of Public Health, Makerere University, Karolinska Institutet, London School of Hygiene and Tropical Medicine; Institute for Economics and Social Research (Zambia) and Traditional Healer Association of Zambia

Stakeholder analysis – Regional and District level – Tanzania

• Purpose: Assessing procurement and distribution of supplies and drugs, supervision, decision-making/ team-work, the influence of health sector reform and local government reform on day to day planning activities and the impact of decentralisation on the district and regional levels sense of autonomy.
• Approach: Focus group discussions with the District Health Management Teams (DHMT) and the Regional Health Management Teams (RHMT) in 9 districts and 3 regions where a larger study was underway. All focus group recordings were transcribed and data was analysed using NUDIST.

AMREF Tanzania and London School of Tropical Medicine and DFID, Tanzania

Quality of care assessments - Tanzania

• Purpose: To determine the quality of care being delivered in government facilities (dispensaries, health centres and outpatient hospital clinics) in Tanzania and the barriers to enhancing quality in these facilities and integrating services, particularly reproductive health, including family planning and STI care as well as HIV prevention and treatment.
• Approach: Health facilities across three regions were ordered according to three strata (rural, roadside and urban), since each strata accentuated the importance of HIV/STI epidemiology. A random selection of health units in each stratum within in each region was then taken. Following a systems-based ‘Structure, Process and Outcome’ framework a comprehensive clinic audit was carried out including: Clinical observation (for every patient aged 15 or over who entered the consultation room, the study clinician completed an observation sheet.) The predetermined minimum sample size for observations was established as 100 observations per facility; Client flow analysis (CFA) of patients (minimum of 60 patients per facility); Health facility audit (using a standardised audit checklist) infrastructure, health learning material, drugs, supplies and equipment, health education sessions, staff training/ education, clinic attendance and the health service provider self-assessment questionnaires. The data was triangulated and analysed using STATA.

AMREF and London School of Hygiene and Tropical Medicine and DFID, Tanzania

Simulated Client Methodology (SCM) – STI/HIV – South Africa

• Purpose: To determine why the majority of residents from inner-city hostels and informal settlements do not seek care for STI or HIV related concerns at the free public health clinics available to them.
• Approach: Nine residents were trained as simulated clients and over a 6 month period were sent to 9 inner city public health facilities (44 visits). Simulated clients presented at clinics with either symptoms of a sexually transmitted infection (STI); request for voluntary counselling and testing (VCT), and enquiries about antiretroviral therapy (ARV). Data was entered into EPIDATA and analysed using STATA.

Reproductive Health and HIV Research Unit and European Union, South Africa

Simulated Client Methodology – Traditional and Biomedical – Uganda and Zambia

• Purpose: To determine the quality of care across sectors in 2 countries and to determine the effectiveness of the simulated client methodology in measuring quality of care from a patient’s perspective as compared to direct observation.
• Approach: 534 simulated client visits were conducted among biomedical and traditional health practitioners in urban Zambia and rural Uganda. The visits followed one of four standardized HIV or STI care-seeking scenarios. Data was collected and recorded qualitatively in short descriptive narratives and quantitatively through a structured questionnaire from simulated clients immediately after their provider visit. Data was entered into EPIDATA & analysed using STATA. 

Institute of Public Health, Makerere University, Karolinska Institute, London School of Hygiene and Tropical Medicine; Institute for Economics and Social Research (Zambia) and Traditional Healer Association of Zambia