← Research Gaps
Public health

Is the timing of national female education and labor force participation policy interventions associated with an acceleration of the maternal mortality transition, as measured by the annual rate of reduction in the maternal mortality ratio, in sub-Saharan African countries from 2000 to 2020?

The gap

While the retrieved literature establishes links between female education, political power, and health outcomes (e.g., [3], [6], [9]), no study directly addresses how specific labor or education policy interventions accelerate the 'maternal mortality transition' in high-mortality regions, leaving the mechanism and impact of such specific policies unresolved.

Study design

Retrospective country-level panel cohort study with an embedded interrupted time-series component

High-level approach

The primary analysis uses two-way fixed-effects panel regression of log-transformed MMR on policy-adoption indicators and a policy × time interaction to test for slope change in MMR reduction post-adoption. A sensitivity analysis models maternal deaths as a count outcome using Poisson or negative-binomial panel regression with a log-live-births offset.

Methodology

Design justification

The two-way fixed-effects panel leverages staggered policy-adoption timing across countries to test whether the MMR reduction slope accelerates after intervention, while controlling for time-invariant country heterogeneity and common year shocks. The embedded interrupted time-series interaction term directly captures the associational question of whether the rate of MMR decline shifts post-policy. This identification strategy follows [4] and the time-series methodology for policy-mortality associations follows [8].

Population

Sub-Saharan African countries (as defined by the World Bank classification) that had a maternal mortality ratio above 300 per 100,000 live births in the year 2000, yielding an estimated 30–35 countries as the cohort.

Setting

National-level (country-level) analysis using existing international databases (World Bank World Development Indicators, UNESCO Institute for Statistics, ILOSTAT, UNICEF, WHO Maternal Mortality Estimation Inter-Agency Group). No primary data collection.

Sampling

Purposive / purposeful sampling — all sub-Saharan African countries meeting the high-mortality threshold (MMR > 300 in 2000) are included as the complete accessible cohort of high-mortality countries in the region. This is a census of eligible countries rather than a sample, ensuring maximum statistical power and generalizability within the target region.

Sample-size approach

n/a — census of all eligible sub-Saharan African countries with MMR > 300 per 100,000 live births in 2000 (approximately 30–35 countries). No sampling is performed; all eligible units are included. The effective sample size for panel analysis is N countries × T years (approximately 30 × 21 = 630 country-year observations), though the number of independent units (~30–35) limits the number of policy variables that can be modeled simultaneously (rule of thumb: ≤ N/10 covariates in fixed-effects regression).

Variables

Outcome: annual maternal mortality ratio (MMR, per 100,000 live births) and computed annual rate of reduction. Primary exposures (time-varying, binary indicators coded 0 before policy adoption, 1 from year of adoption): (1) adoption of free/compulsory secondary education policy for girls, (2) adoption of national female labor force participation policies (e.g., paid maternity leave expansion, anti-discrimination legislation, female employment quotas). Covariates (time-varying, country-level): GDP per capita, total health expenditure per capita, proportion of births attended by skilled health personnel, female secondary school enrollment rate, contraceptive prevalence rate, urban population share, and total fertility rate. These covariates are informed by the determinants framework in [12] and the mechanisms identified in [4] (skilled birth attendance) and [7] (maternal education).

Data sources

World Bank World Development Indicators (GDP per capita, health expenditure, skilled birth attendance, female secondary enrollment, contraceptive prevalence, urbanization, fertility rate); WHO/UNICEF/UNFPA/World Bank Group Maternal Mortality Estimation Inter-Agency Group (MMR estimates); UNESCO Institute for Statistics (education policy adoption dates and enrollment); ILOSTAT/NATLEX database (labor policy adoption dates); African Union and country-level policy documents (verification of policy enactment dates). This multi-source approach follows the data integration strategy of [[1]] and [[2]], which synthesized country-level data from international databases.

Time

6–9 months: 2–3 months for database compilation and policy adoption date verification, 1–2 months for data cleaning and harmonization, 2–3 months for analysis, and 1 month for write-up.

Cost

Low — uses exclusively secondary, publicly available data from international databases. Costs limited to research software (Stata or R, both potentially available through institutional licenses) and possible travel for archive-based policy verification. Estimated total: USD 500–2,000.

Ethics

IRB exemption expected — analysis uses only aggregated, publicly available country-level data with no individual human subjects. No informed consent required. Ethical considerations include ensuring accurate representation of country-level data, transparent reporting of data quality limitations (especially MMR estimation uncertainty in high-mortality settings), and avoiding stigmatization of countries with slower progress.

Grounding references

  1. Global burden and strength of evidence for 88 risk factors in 204 countries and 811 subnational locations, 1990–2021: a systematic analysis for the Global Burden of Disease Study 2021 — Masayuki Teramoto, Gregory A. Roth, Aleksandr Y. Aravkin, Peng Zheng, Kalkidan Hassen Abate, Yohannes Abate · The Lancet, 2024 · DOI 10.1016/s0140-6736(24)00933-4
  2. Global age-sex-specific mortality, life expectancy, and population estimates in 204 countries and territories and 811 subnational locations, 1950–2021, and the impact of the COVID-19 pandemic: a comprehensive demographic analysis for the Global Burden of Disease Study 2021 — Masayuki Teramoto, Hmwe Hmwe Kyu, Amirali Aali, Cristiana Abbafati, Jaffar Abbas, Rouzbeh Abbasgholizadeh · The Lancet, 2024 · DOI 10.1016/s0140-6736(24)00476-8
  3. Determinants of female labor force participation: implications for policy in Qatar — Noora Lari, Amal Awadalla, Mohammad Al-Ansari, Engi Elmaghraby · Cogent Social Sciences, 2022 · DOI 10.1080/23311886.2022.2130223
  4. Maternal Mortality and Women’s Political Power — Sonia Bhalotra, Damian Clarke, Joseph Flavian Gomes, Atheendar Venkataramani · Journal of the European Economic Association, 2023 · DOI 10.1093/jeea/jvad012
  5. The labor force participation of Indian women before and after widowhood — Megan N. Reed · Demographic Research, 2020 · DOI 10.4054/demres.2020.43.24
  6. The public health effects of interventions similar to basic income: a scoping review — Marcia Gibson, Wendy Hearty, Peter Craig · The Lancet Public Health, 2020 · DOI 10.1016/s2468-2667(20)30005-0
  7. Differential impact of maternal education on under-five mortality in rural and urban India — Moradhvaj, K. C. Samir · Health & Place, 2023 · DOI 10.1016/j.healthplace.2023.102987
  8. Public Health Expenditure and Under-five Mortality in Nigeria: An Overview for Policy Intervention — Dominic E. Azuh, Romanus Osabohien, Mary U Orbih, Abigail Godwin · Open Access Macedonian Journal of Medical Sciences, 2020 · DOI 10.3889/oamjms.2020.4327
  9. Efficacy of companion-integrated childbirth preparation for childbirth fear, self-efficacy, and maternal support in primigravid women in Malawi — Berlington Munkhondya, Tiwonge Ethel Mbeya Munkhondya, Ellen Chirwa, Honghong Wang · BMC Pregnancy and Childbirth, 2020 · DOI 10.1186/s12884-019-2717-5
  10. The contribution of female health to economic development — David E. Bloom, Michael Kühn, Klaus Prettner · The Economic Journal, 2020 · DOI 10.1093/ej/ueaa061

Open this in the editor to start your project

You’ll get the gap and study design seeded into a working draft. The full protocol — worked sample size, variable table, analysis plan, and lit-review starter — unlocks inside the editor.