Optimizing data collection for obstetrical ultrasound research at the primary health care level in rural Uganda
Background Neonatal and maternal mortality remains high in low- and middle-income countries (LMIC), especially in sub-Saharan Africa. Quality data collection is crucial to understand the magnitude of these problems and to measure the impact of interventions aimed at improving neonatal and maternal m...
Saved in:
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2024-12-01
|
| Series: | Global Health Action |
| Subjects: | |
| Online Access: | http://dx.doi.org/10.1080/16549716.2024.2436715 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Background Neonatal and maternal mortality remains high in low- and middle-income countries (LMIC), especially in sub-Saharan Africa. Quality data collection is crucial to understand the magnitude of these problems and to measure the impact of interventions aimed at improving neonatal and maternal mortality. However, data collection in the low-income country setting, especially in rural areas, has been a challenge for researchers, policy makers, and public health officials. Here, we describe the methodology, experience and lessons learned while collecting data at lower-level primary health care facilities in rural Uganda. Methods Data collection was performed at Health Center III sites in rural Uganda, in partnership with Imaging the World and its affiliate Imaging the World Africa. The primary purpose of the data collection was to study the efficacy and clinical effect of introducing prenatal ultrasound services at these sites. Local data clerks were hired to perform the data collection through a combination of intensive training and on-the-ground support. Frequent oversight was used to support data collection. Results Of 2,397 enrolled pregnant women, 1,977 (82.5%) had complete outcome data. Upon independent expert audit, the data were >80% accurate for 10/11 variables and >90% accurate for 6/11 variables. Overall, the data collected at the rural HCs were 90% accurate. Discussion Accurate and complete data collection is possible in an LMIC setting if appropriate training and oversight are employed. |
|---|---|
| ISSN: | 1654-9880 |