Climate Change and Preterm Birth

Trends in the temperate zone

With a cumulative incidence above 8%, preterm births (PTB) constitute a substantial burden for the public healthcare system. They are associated with increased neonatal mortality and morbidity as well as long-term health sequelae for those affected. The etiology is multifactorial and includes—among others—preterm labor and preterm pre-labor rupture of membranes (PPROM) (1).

A multitude of studies have also focused on climate change as a potential risk factor (2). Pregnant women are considered particularly susceptible to heat stress, whose frequency and intensity have notably increased in Europe in recent years. Heat stress can have acute as well as delayed and potentially cumulative effects on the outcome of a pregnancy, including prematurity and fetal growth restriction.

International data have confirmed the association between heat stress and preterm birth, but robust data for Germany are scarce to date (2). Our study adds to the current data the results for a large rural state in central Germany.

Study design

The study population is based on perinatal data collections in Thuringia 2014–2018 and includes a total of 84,568 births. We obtained daily minimum and maximum temperatures (T_min and T_max) from 134 measuring stations in Thuringia and via the portal of the German Weather Service (DWD). We linked pregnant women’s places of residence (postcode) with the nearest weather stations with a spatial resolution of 0.5° longitude and latitude. We calculated annual temperature percentiles per measuring station for the summer months May to September. We were thus able to define a heat event as the minimal or maximal temperature above the respective 90th percentile.

We used a case-crossover design for our analysis, with each birth serving as its own control. This method allowed us to control for time-invariant, unobserved confounders by exact matching (3). We calculated odds ratios (OR) for preterm births before the 37^th, 32^nd, and 28^th week of gestation. This model considers the frequency during the course of the pregnancy as well as the occurrence of a heat event within the vulnerable time window in the 3 days preceding the preterm birth (2, 4). We used the software package R (version 4.5.0) for our statistical analysis.

Induced births and elective cesareans were considered intended preterm births and were excluded from our analysis; the same applied for births without any assignable temperature data.

Results

During the observation period, the summers of 2015 and 2018 were characterized by particular heat stress (Figure 1), and the calculated 90th percentile of the daily maximum temperature consequently varied between 33.1°C (2015) and 29.5°C (2014). The same applied for the 90th percentile of minimum temperatures, which was 17.5°C in 2015 and 16.3°C in 2017.

Figure 1

Weekly rates of preterm births and maximum temperatures in Thuringia (2014–2018)

Altogether 48 608 births were linked with meteorological data. The mothers’ mean age was 29±5 years, and 46.3% were primiparous. Overall the mean rate of preterm births was 6.96% (<37^th week of gestation), of which 0.4% were extremely preterm (<28^th week of gestation) and 1.1% early preterm births (<32^nd week of gestation).

We were able to show a significant association of the risk of preterm birth with heat events in the vulnerable phase before a preterm birth (Figure 2). The strongest effect was observed for extremely preterm births before the 28^th week of gestation and for high maximum temperatures (OR 2.34; 95% confidence interval: [1.43; 3.83]) as well as for high minimum temperatures (OR 2.06; [1.22; 3.46]). The observed effects were less pronounced for the general risk of preterm birth before the 37^th week of gestation (T_max: OR 1.69; [1.48; 1.94]; T_min: OR 1.72; [1.51; 1.97]).

Figure 2

Risk of preterm birth (odds ratios) as a result of heat stress

Conclusions

Our results confirm international analyses from different climate zones that primarily described risks to reproductive health from the effects of heat events (2). Our study showed the same for the risk of preterm births in a rural area in central Europe. The results are consistent with an observed absolute risk increase for extremely preterm birth before the 28^th week of gestation of 0.37% to 0.73% if heat events occurred in the last 5 days before the birth (5).

We collected the temperature data with their high temporal and spatial resolution. This allowed us to consider the risk on an individual basis and avoids leveling out as a result of averaged population effects. Correspondingly, the obstetric data from the perinatal data collection in Thuringia allowed for a blanket and almost complete collection of all births in the observation period and for high-quality data. A limitation lies in the fact, however, that the dataset collected for the purposes of quality assurance is limited in its scope, and living conditions (for example, air conditioned living spaces) and risk factors (for example, maternal comorbidities) were therefore not reflected in their totality. We are planning an analysis of the nationwide perinatal data collection on the basis of the data from one German state presented in this article.

Heat stress in moderate climate zones is increasing as a result of global climate change, which presents new challenges for perinatal care and healthcare provision in Germany too. It will become necessary to equip hospitals and healthcare institutions generally with air conditioning technology. Municipal heat plans aim to protect vulnerable groups—such as children, older persons, and, as shown, pregnant women—from heat stress as in the summer of 2025. These aspects should also be considered in the necessary future reorganization of obstetric and neonatal care in rural areas, such as Thuringia.

Acknowledgment

We thank Prof. André Scherag and Prof. Peter Schlattmann for their much appreciated contribution to this project.

Yvonne Heimann, Ekkehard Schleußner

Conflict of interest statement
The authors declare that no conflict of interest exists.

Manuscript received on 11 April 2025, revised version accepted on 8 July 2025.

Translated from the original German by Birte Twisselmann, PhD.

Cite this as:
Heimann Y, Schleußner E: Climate change and preterm birth: Trends in the temperate zone. Dtsch Arztebl Int 2025; 122: 508–9. DOI: 10.3238/arztebl.m2025.0128

References

1. Berger R, Abele H, Bahlmann F, et al.: Prevention and therapy of preterm birth. guideline of the DGGG, OEGGG and SGGG (S2k level, AWMF ­registry number 015/025, september 2022)—part 2 with ­recommendations on the tertiary prevention of preterm birth and on the management of preterm premature rupture of membranes. Geburtshilfe Frauenheilkd 2023; 83: 569–601

2. Lakhoo DP, Brink N, Radebe L, et al.: A systematic review and meta-­analysis of heat exposure impacts on maternal, fetal and neonatal health. Nat Med 2025; 31: 684–94

3. Suissa S: The case-time-control design: Further assumptions and ­conditions. Epidemiology 1998; 9: 441–5

4. Zhang Z: Case-crossover design and its implementation in R. Ann Transl Med 2016; 4: 341

5. Schleußner E, Heimann Y: Auswirkungen der Klimakrise auf Schwangerschaft und kindliche Gesundheit. Gynäkologie 2025; 58: 360–4