PhD defence Amani Kikula

Supervisors

• Prof. dr. Lenka Beňová (ITM)

• Prof. dr. José L. Peñalvo (University of Antwerp) 

• Prof. dr. Josefien van Olmen (University of Antwerp)

• Prof. dr. Andrea B. Pembe (Muhimibili University of Health and Allied Sciences, Tanzania) 

• Prof. dr. Kaushik Ramaiya (Shree Hindu Mandal Hospital, Tanzania)

• Prof. dr. Nathanael Sirili (Muhimibili University of Health and Allied Sciences, Tanzania)

Abstract

Introduction
Tanzania, like the rest of the global community, is facing an increasing prevalence of obesity. Excess weight before pregnancy is the predominant risk for the development of gestational diabetes mellitus (GDM), which is associated with adverse pregnancy outcomes. Screening for GDM during antenatal care (ANC) is an opportunity for primary prevention of complications during pregnancy and beyond. Tanzania had two national guidelines that recommend different screening algorithms for GDM. There is limited knowledge on the body mass index (BMI) of women of reproductive age and GDM care provided in health care facilities in Tanzania. This thesis aimed to explore the trend of BMI among women of reproductive age, GDM care practices during routine ANC, estimate the magnitude of GDM among ANC users, and compare the performance of two nationally recommended GDM screening algorithms. We further explored experiences of pregnant women, health care workers (HCW), and health administrators on GDM care during ANC and how the service can be improved.

Methods
This study had an observational cross-sectional study design with a sequential explanatory mixed-methods approach. This PhD work involved the use of the readily available Tanzania Demographic and Health Survey (DHS) data from 2004 to 2015/16. Primary data collection was conducted from January 2024 to August 2025 in ANC clinics of two primary-level hospitals in Tanzania: Kisarawe District Hospital in Coast region and Mbagala Rangi Tatu Hospital in Dar es Salaam region. Quantitative data were collected to determine the structural capacity of the hospitals to provide GDM services, the magnitude of GDM among 970 pregnant women, and the sensitivity and specificity of the two nationally recommended screening algorithms. Sensitivity analysis using receiver operating characteristics was used to determine the performance of screening tests. Qualitative data were collected through in-depth interviews with 20 HCWs, 13 health administrators and 22 pregnant women, as well as 192 hours of structured observation. Qualitative content analysis was used for qualitative data sets.

Results
The percentage of women of reproductive age intending to conceive who had a high BMI (overweight and obesity) increased from 14.2% in 2004 to 23.8% in 2015. Within the two hospitals, GDM care was not part of routine ANC provided. Health care workers were not supported to provide GDM care as part of routine ANC. When GDM screening was introduced at the two study hospitals, the prevalence of hyperglycemia in pregnancy was 10% (7.9% GDM and 2.1% diabetes in pregnancy). For the nationally recommended screening tests, glycosuria had a sensitivity of 4.1% and a specificity of 97.1%, while the checklist screening test had a sensitivity of 72.2% and a specificity of 32.4% (compared to 75g OGTT). When asked about their experience with the introduced GDM care services, women wanted GDM care services within ANC and wanted to be informed about it during the group health education sessions, male engagement in GDM care, waiving user fees for GDM care, and increasing time spent for counselling/health education. The HCWs recommended: ANC services that include GDM care should be provided under one roof, training of HCWs on GDM, HCW to be supported with GDM guiding information materials, including GDM service monitoring within routine ANC monitoring tools, and develop SOP for GDM care.

Conclusions
More women of reproductive age are at increased risk for GDM in Tanzania. In the two hospitals studied, GDM care was not part of the routinely provided ANC, largely attributed to low priority for GDM, a lack of knowledge and skills among HCWs regarding GDM care. One in every 10 women screened had hyperglycemia in pregnancy. The nationally recommended screening tests performed poorly as GDM screening tests. Despite these challenges in GDM care provision in ANC, there is an existing platform for health education for the women where GDM can be integrated. There are also continuous medical education forums for HCW, which could be utilized for improving skills, attitude, and knowledge on GDM care. Prioritizing GDM care services within health management team agendas will improve GDM service smooth integration within existing ANC structures.