Community midwives reduced preterm birth Community-based midwife continuity of care was associated with fewer preterm births in disadvantaged, ethnically diverse London communities.
*Prospective matched cohort study; Level 3 (OCEBM).

Citation

Fernandez Turienzo C, Burton S, Khan Z, et al. The impact of community-based midwife continuity of care models for women living in areas of social disadvantage and ethnic diversity in the United Kingdom. BJOG. 2026;133:958-973. doi:10.1111/1471-0528.70101

Background

Maternal and newborn outcomes are worse among women living in deprivation and among many ethnic minority groups in the United Kingdom. This study tested whether community-based care from the same midwife or small midwife team could reduce these gaps.

Patients

13,609 singleton pregnancies in South London, 2018-2020. Multiple pregnancies and records missing key first-visit data were excluded.

Intervention

Community-based midwife continuity of care from pregnancy through the postnatal period.

Control

Standard maternity care, including usual midwife, general practitioner, and obstetric care without planned continuity.

Outcome

Primary outcome: preterm birth before 37 weeks. Secondary outcomes included birth mode, newborn size, visits, referrals, and hospital use.

Follow-up Period

First antenatal visit through discharge from maternity services.

Results

After propensity matching, 1,338 community-model pregnancies were compared with 5,352 standard-care pregnancies.

OutcomeCommunity modelStandard careAdjusted effect
Preterm birth (primary)4.6%8.4%RR 0.54 (95% CI 0.40-0.70); NNT 27
Preterm birth, ethnic minority subgroup6.4%9.5%RR 0.66 (95% CI 0.44-0.95); NNT 33
Preterm birth, most deprived areas5.1%8.2%RR 0.60 (95% CI 0.43-0.82); NNT 33
Caesarean birth25.0%35.7%RR 0.53 (95% CI 0.46-0.61); NNT 10
Low birthweight5.7%9.4%RR 0.60 (95% CI 0.46-0.75); NNT 28
Missed appointments5.4%15.7%RR 0.66 (95% CI 0.55-0.78); NNT 10
Mental health referral5.2%3.5%RR 1.80 (95% CI 1.26-2.37)

RR: risk ratio. CI: confidence interval. NNT: number needed to treat.

Results were analyzed by assigned care model. Standard care was the current available treatment.

Limitations

This was not randomized, so unmeasured differences may remain. Data depended on routine records, and ethnicity, mental health, and social risks may be misclassified or underreported. Several care models and broad ethnic groups were combined. Preterm births were not separated into spontaneous versus medically indicated births.

Funding

NIHR and MRC; no company funding reported.

Clinical Application

Consider community midwife continuity models in high-inequality maternity settings; evidence supports policy direction but randomized and cost studies are still needed.