Unintended pregnancy increases the risk of preterm birth by 26.7%, according to a large-scale analysis of more than 900,000 pregnancies across 37 low- and middle-income countries published in a 2026 study. Researchers used target trial emulation to strengthen causal inference from observational data, revealing that roughly 21.1% of preterm births in the study population could be tied to mistimed or unwanted conceptions.
Demographic and Health Surveys Data Analysis
Researchers examined Demographic and Health Surveys collected between 1997 and 2021. The dataset covered 911,423 pregnancies among 635,139 women aged 15 to 49 years old across 37 low- and middle-income countries. Out of these total cases, 205,655 pregnancies—representing 22.6%—were classified as unintended at the time of conception.
To move beyond standard observational limits, the study deployed target trial emulation. This statistical approach aims to approximate features of a randomized controlled trial using observational data. Investigators applied statistical weighting to balance key characteristics, including maternal age, body mass index, education level, household wealth, marital status, smoking habits, residential setting, reproductive history, and knowledge of family planning methods.
Comparing Preterm Birth Risks and Methodology Discrepancies
In the target trial emulation model, unintended pregnancy showed a clear association with a 26.7% increased risk of preterm delivery compared to intended pregnancies. Preterm birth occurred in 4.38% of unintended pregnancies, contrasted with 3.96% of intended pregnancies. Researchers estimated a risk ratio of 1.267. Conversely, unintended pregnancy did not show an association with low birth weight under this specific trial emulation approach.

This finding contrasted sharply with conventional adjusted regression analysis. Traditional regression models previously identified links between unintended pregnancy and both preterm birth and low birth weight. According to the research team, differences in accounting for confounding variables and missing outcome data explain this methodological discrepancy.
Did You Know?
Target trial emulation helps researchers correct for hidden biases in observational health data by mimicking the design principles of a randomized controlled trial after the data has already been collected.
Potential Mechanisms and Study Limitations
Several underlying mechanisms may link unintended pregnancy to preterm birth. Psychological stress and depression can impact neuroendocrine and inflammatory pathways, which in turn influence uterine activity and cervical changes. Behavioral factors may also play a contributing role.
However, the study carries notable limitations. Pregnancy intention status was available exclusively for live births, meaning pregnancies ending in miscarriage, stillbirth, or abortion were excluded from the cohort. Birth weight data also suffered from frequent missing records, and gestational duration was tracked in months rather than weeks. A sensitivity analysis restricted strictly to each woman’s most recent birth also showed that the association with preterm birth lost statistical significance.
Improving Maternal Healthcare Infrastructure
Despite data limitations, the authors conclude that expanding access to effective contraception, improving reproductive health education, and strengthening maternal healthcare systems can help reduce the broader health burden tied to unintended pregnancies, particularly in resource-limited settings.
Frequently Asked Questions
Does an unintended pregnancy increase the risk of preterm birth?
Yes. A 2026 analysis utilizing target trial emulation across 37 low- and middle-income countries found a 26.7% increased risk of preterm birth associated with unintended pregnancies.

How many pregnancies were included in the 2026 study?
The analysis evaluated 911,423 pregnancies among 635,139 women aged 15 to 49, drawing from Demographic and Health Surveys collected between 1997 and 2021.
Was low birth weight also linked to unintended pregnancy?
Under the target trial emulation approach, unintended pregnancy was not associated with low birth weight, though conventional regression models had previously suggested a link.
Join the Conversation
What are your thoughts on improving maternal healthcare access in developing regions? Share your perspective in the comments below or subscribe to our newsletter for more updates on global health research.
Related reading