Role for environmental factors on the relationship between seasons of conception and preterm birth: An observational study in China

doi:10.21203/rs.3.rs-3092383/v1

Download PDF

Research Article

Role for environmental factors on the relationship between seasons of conception and preterm birth: An observational study in China

https://doi.org/10.21203/rs.3.rs-3092383/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background：The relationship between season and the risk of preterm birth is not consistent. Importantly, the role of environmental factors in it is not clear.

Methods：Based on the baseline of the Jinan birth cohort, our study focused on seasons of conception to examine the relationship. The mothers from the birth cohort were the subjects, and they were interviewed face-to-face by the questionnaire. The average exposure levels of temperature and humidity during the first 30 days of pregnancy were calculated, and those of air pollutants were assessed by an inverse distance weighting method.

Results: A total of 4.4 % (263/5984) of mothers had premature babies. The rates of preterm birth were higher in winter (6.0 %) and summer (4.5 %) than that in spring (3.1 %) at the 0.05 significant level. In the multivariate logistic regression model, the adjusted odds ratio (OR) and 95 % confidence interval (CI) for the risk of preterm birth at conception in winter and summer were 2.228 (1.512-3.316) and1.541 (1.055-2.277). However, after further adjustment for humidity, only the association between winter conception and preterm birth remained statistically significant (OR: 2.353, 95% CI: 1.594-3.511). Moreover, the humidity partially mediated the relationship (41.2 %) between conception in winter and preterm birth, and humidity in winter was negatively associated with the risk of preterm birth (indirect effect = -0.007, P < 0.001).

Conclusions: For women living in the temperate regions, conception in summer and winter may increase the risk of preterm birth, and this association persisted in winter after adjusting for environmental factors. During the first 30 days of pregnancy, lower levels of humidity exposure in winter contributed to and mediated the relationship. The results may provide epidemiological evidence for pregnancy planning and care of women from the perspective of environmental factors.

Preterm birth

Season

Environment

Mediating effect

Birth Cohort

Preterm birth is a common adverse pregnancy outcome. The World Health Organization defines preterm birth as the birth of fewer than 37 weeks of pregnancy or the birth before the 259th day of pregnancy(1). It was suggested that preterm birth was one of the causes of death for children under the age of 5 (2). In addition, not only does preterm birth affect the health of newborns such as developing respiratory diseases(3), but also has long-term effects on the growth and development of children such as causing the poor development of the nervous system in childhood (4, 5). The global estimated prevalence of preterm birth was approximately 9.6%, with a 95% confidence interval (CI) of 9.1% − 10.1% in 2005 (6) and 11.1% (95% CI: 9.1%-13.4%) in 2010 (7). Therefore, preventing the occurrence of premature birth is of great significance for promoting the health of children and reducing the associated social burden.

The occurrence of preterm birth is affected by many factors such as maternal age, abnormal pregnancy, tobacco smoke exposure during pregnancy, fetal congenital diseases and temperature (8–13), among which season is an important factor related to the environment, social activities and lifestyle habits. Some studies tried to illuminate the relationship between season and preterm birth, and the seasonal pattern of preterm birth has been found around the world. A study conducted in the United States found that the highest prevalence of preterm birth occurred among the mothers who were pregnant in winter (11.3%) and spring (11.0%) (14). Another study from Greece demonstrated that the occurrence of preterm birth increased in spring (6.2%) and summer (6.2%)(15). Epidemiological evidence from South Korea indicated that the prevalence of preterm births peaked in summer (5.4%) and winter (5.1%) (16). Given the inconsistent results reported in these studies, more research evidence is needed to confirm the link between season and preterm birth. There are multiple potential reasons for this. Firstly, the results came from different countries with different geographical location and social customs(17, 18). In addition, seasonal variables were defined differently: some chose the season of conception(14), others the season of birth (15, 16). Finally, there is a lack of evidence on the association between the season of conception according to the 24 solar terms and the risk of preterm birth. Thus, the seasonal patterns of preterm birth still need to be explored.

A number of studies have found association between environmental factors and preterm birth. Some studies have pointed that exposure to PM_2.5(19), O₃ (20)and high temperature(21) could affect the risk of preterm birth. Meanwhile, the exposure levels of some environmental factors are different in different seasons. Specifically, people are more likely to be exposed to high temperatures in summer and low temperatures in winter, and more likely to be exposed to high levels of PM₁₀ and PM_2.5 in winter as well as O₃ in summer(20, 22). Therefore, it is important to consider the effect of the environmental factors on the risk of preterm birth when we explore the relationship between the season and preterm birth.

Therefore, this study was based on the baseline data from the Jinan Birth Cohort to achieve three goals: 1) to examine the association between season of conception and the risk of preterm birth; 2) to explore whether the association between season of conception and preterm birth persisted after adjusting for climatic factors (temperature and humidity) as well as atmospheric pollutants (PM_2.5 and O₃); and 3) to explore whether there were mediating effects of other environmental factors between season of conception and preterm birth. Not only will the findings shed light on the role of environmental factors in seasonal influences on preterm birth, but also provide a new scientific basis for pregnant planning and pregnancy care of women.

2.1. Study population

The present study was a cross-sectional study. The research population was from the baseline survey of the birth cohort study of which the purpose was to examine the effect of exposure to air pollutants in early life on childhood asthma. Fifteen out of 116 community vaccination clinics in Jinan City, Shandong province, China, were selected as the study sites of the birth cohort. From January 2018 to December 2019, when infants younger than one month visited outpatient clinics for vaccinations, infants and their mothers were recruited as study subjects for the birth cohort. The inclusion criteria(23, 24) were: i) the infants had no congenital disease; ii) the mothers had been living in Jinan City during pregnancy; iii) the mothers did not have serious mental illness or language communication difficulties; iiii) the mothers giving birth to a single child; iiiii) the mothers who gave birth between November 1, 2018 and October 31, 2019. Each mother was given a face-to-face questionnaire to obtain the relevant variables for the study.

2.2. Dependent variable and independent variable

Preterm birth, the dependent variable of this research, is defined as the birth less than 37 weeks of gestation [1]. The gestational week for each mother was calculated by the date of birth minus the date of the first day for the last menstruation. And the gestational week was double determined by B ultrasound diagnose and the date of the first day of last menstruation and the date of delivery.

The season of conception was the independent variable. The date of conception was calculated based on the date of birth and gestational age. The season in which this date occurred was defined as the season of conception. The seasons are divided according to the Lichun, Lixia, Liqiu and Lidong of the 24 solar terms in the traditional Chinese calendar. Specifically, two solar terms in one month, and six solar terms are in each season, twenty-four solar terms are in one year. For spring season, it includes Beginning of Spring, Rain Water, Waking of Insects, Vernal Equinox, Pure Brightness and Grain Rain, which are Lichun, Yushui, Jingzhe, Chunfen, Qingming, and Guyu in Chinese. For summer season, it includes Beginning of Summer, Grain Budding, Grain in Ear, Summer Solstice, Slight Heat and Great Heat, which are Lixia, Xiaoman, Mangzhong, Xiazhi, Xiaoshu, and Dashu in Chinese. For autumn season, it includes Beginning of Autumn, Limit of Heat, White Dew, Autumnal Equinox, Cold Dew and Frost's Descent, which are Liqiu, Chushu, Bailu, Qiufen, Hanlu, and Shuangjiang in Chinese. As the same concern, for winter, it includes Beginning of Winter, Slight Snow, Great Snow, Winter Solstice, Slight Cold and Great Cold, which are Lidong, Xiaoxue, Daxue, Dongzhi, Xiaohan, and Dahan in Chinese.

2.3. Assessment of exposure to environmental factors

Monitoring data of ambient temperature, relative humidity (humidity) and air pollutants of subjects during the first 30 days of pregnancy were obtained from 17 Jinan Environmental Monitoring Station, which covered the study area. The first 30 days of pregnancy was from the first day of conception to the following 29 days. The daily mean of ambient temperature and humidity for each subject was obtained by calculating the mean of the average daily temperature and humidity of environmental monitoring stations in Jinan City. From the same environmental monitoring stations, we obtained average daily monitoring values of six air pollutants including SO₂ (sulfur dioxide), NO₂ (nitrogen dioxide), CO (carbon monoxide), O₃ (ozone), PM₁₀ (particulate matter with an aerodynamic diameter of less than 10 mm) and PM_2.5 (particulate matter with an aerodynamic diameter of less than 2.5 mm). Based on the beginning date of pregnancy and the family address for each subject, the individual daily mean of air pollutant concentration was estimated by an inverse distance weighting (IDW) method [21]. Afterwards, we calculated the arithmetic mean of the daily exposure levels during the first 30 days of pregnancy for each individual as the exposure level to environmental factors.

2.4. Covariates

Covariate information was obtained through the questionnaire, which included general characteristics of the mother, maternal pregnancy information, infant information and information of the family. The general characteristics of the mother included parity, delivery method, age, education level, housewife and regular menstrual cycle. According to the parity, we divided the subjects into primipara and multipara. Age was divided into four groups: ≤25 years, 25–30 years, 30–35 years, and > 35 years. Maternal pregnancy information mainly included pregnancy complications, pregnancy-induced hypertension, abnormal pregnancy, passive smoking, vitamin D supplement, folic acid supplement, calcium supplement, work, and exercise. Pregnancy complications included abnormal umbilical cord, amniotic fluid embolism and uterine rupture. Abnormal pregnancy included placenta previa, threatened preterm labor, gestational diabetes mellitus and prenatal bleeding. Passive smoking was defined as exposure to tobacco smoke for more than 30 minutes per week. Exercise was defined as any planned, organized, and repetitive physical activity(25), including walking, brisk walking, swimming, yoga/Pilates/ maternity gymnastics, squats and others. Infant information just included the gender of infant. Information of the family included number of family members, total monthly income (RMB) and address. Address was defined as the geographical location of the residence where the mother lived for the longest period of her pregnancy. All variable information, including dependent variables, independent variables, and covariates except for environmental monitoring data, was obtained through face-to-face interviews using questionnaires at the baseline survey.

2.5. Statistical analysis

For continuous variables, when they were normally distributed, the mean and standard deviation (SD) were used to describe them, and Student's t test and Analysis of Variance were used to test for differences between groups; when they are not normally distributed, the median and interquartile range (IQR) are used to describe them, and rank sum test was used to examine the differences between groups. For categorical variables, rates and constituent ratios were used to describe them, and comparisons between groups were performed by Chi-square test. We constructed the binary logistic regression models of univariate analysis and multivariate analysis. In the multifactor model, we adjusted for covariables that might be associated with preterm birth, covariates were grouped into three groups according to basic maternal information, adverse pregnancy symptoms and behavioral factors, progressively adjusted. Environmental factors were also separately included in the multifactor logistic regression model for adjustment. In addition, we calculated the variance inflation factors (VIF) of the covariables in the model for collinearity diagnosis and excluded the covariables of which VIF over 10. The odds ratio (OR) and corresponding 95% confidence interval (CI) were used to described the effect of variates on preterm birth. We adopted Pearson correlation coefficient to conduct the correlation analysis of continuous variables. The mediating effect of environmental factors on the association between season and preterm birth was analyzed by sequential test [22, 23].

Statistical analyses were performed with R version 3.6.1 (http://www.r-project.org/) using gmodels, EnvExpInd, mediation, and car packages. P values of < 0.05 were considered statistically significant.

2.6. Sensitive analysis

Furthermore, because the study subjects were from 15 community vaccination clinics which were selected in the birth cohort study, in order to prevent selection bias, infants from each clinic were excluded successively, and a multivariate analysis was conducted again to prove the stability of the results.

Also, in order to control for the effect of moving during pregnancy, the variable of moving during pregnancy was firstly adjusted in the model. Secondly, after excluding study participants who moved during pregnancy, the analysis was conducted again.

2.7. Quality control

In every procedure of the study, rigorous quality control was carried out. The questionnaire was designed by experts in reproductive epidemiology. After designing, the questionnaire was improved through pre-survey. The investigators at each site had a 3-day training course on maternal and infant knowledge and investigation skills. During the investigation and data input and sorting, there were quality control personnel on the site to supervise. In addition, Epidata 3.1 was used for data entry and logical proofreading.

3.1. The characteristics of the subjects

6,640 pairs of infants and mothers were included in the baseline of the birth cohort. After excluding 139 multiple pregnancies, we selected the mothers who gave birth between November 1, 2018 and October 31, 2019 as the subjects. In the end, 5,984 mothers were included in the study (Figure S1). We presented the distribution of outpatient clinics in Jinan, the location of air monitoring stations and the home addresses of the study participants in Fig. 1. The characteristics of subjects was shown in Table 1.

Table 1

Characteristics of subjects and the distribution of preterm birth in different subgroups.
Classification		Total		Preterm birth		P
Classification		n	proportion (%)	n	rate (%)	P
Maternal general information
age (year)	≤ 25	722	12.1	24	3.3	< 0.001
	25 ~ 30	2422	40.5	86	3.6
	30 ~ 35	1981	33.1	96	4.8
	>35	859	14.4	56	6.6
education level	below college degree	1558	26.0	81	5.2	0.084
	college degree or above	4426	74.0	182	4.1
housewife	no	4731	79.1	193	4.1	0.025
	yes	1253	20.9	70	5.6
regular menstrual cycle	no	701	11.7	34	4.9	0.598
	yes	5283	88.3	229	4.3
Infant
gender	male	3102	51.8	150	4.8	0.097
	female	2882	48.2	113	3.9
Family
number of family members	≤ 3	1576	26.3	57	3.6	0.092
	> 4	4408	73.7	206	4.7
total monthly income (RMB)	< 6000	744	12.4	42	5.6	0.366
	6000 ~ 12000	2819	47.1	119	4.2
	12000 ~ 18000	1627	27.2	69	4.2
	> 18000	794	13.3	33	4.2
Maternal pregnancy information
primipara	no	3035	50.7	150	4.9	0.042
	yes	2949	49.3	113	3.8
delivery method	normal delivery	3275	54.7	103	3.1	< 0.001
	cesarean section	2709	45.3	160	5.9
pregnancy complications	no	5881	98.3	254	4.3	0.046
	yes	103	1.7	9	8.7
pregnancy-induced hypertension	no	5800	96.9	228	3.9	< 0.001
	yes	184	3.1	35	19.0
abnormal pregnancy	no	5684	95.0	206	3.6	< 0.001
	yes	300	5.0	57	19.0
passive smoking	no	5114	85.5	227	4.4	0.756
	yes	870	14.5	36	4.1
vitamin D supplement	no	3589	60.0	140	3.9	0.027
	yes	2395	40.0	123	5.1
folic acid supplement	no	388	6.5	22	5.7	0.254
	yes	5596	93.5	241	4.3
calcium supplement	no	1024	17.11	49	4.8	0.558
	yes	4960	82.89	214	4.3
work	no	2298	38.4	102	4.4	0.948
	yes	3686	61.6	161	4.4
exercise	no	1784	29.8	99	5.5	0.006
	yes	4200	70.2	164	3.9

3.2. The distributions of preterm birth in the population

The range of gestational age for the subjects was 188 days to 309 days, and the median and IQR of gestational age were 275 days and 11 days. Among 5984 subjects, 263 mothers gave birth to premature babies, and the rate of premature babies was 4.4%.

The distributions of preterm birth in different subgroups were showed in Table 1. Compared with the mothers without preterm birth, mothers with preterm birth were more likely to be housewives, older mothers, primiparas, and the mothers with cesarean section, pregnancy complications, pregnancy-induced hypertension, abnormal pregnancy, exercised, and the mothers who took vitamin D supplements during pregnancy (P < 0.05).

3.3. Preterm birth in different seasons of conception

The distribution of solar terms in the time range of maternal conception was described in Table S1, and the distribution of seasons was showed in Figure S2. The proportion of mothers who became pregnant, the proportion and the incidence of preterm birth in different months (according to the solar terms) were presented in Fig. 2. Preterm births occurred in 3.1% (45/1429) of the mothers who conceived in spring, 4.5% (79/1745) in summer, 4.2% (66/1584) of in autumn, and 6.0% (73/1226) in winter. The preterm birth rate was higher among the subjects who conceived in summer (4.5% vs 3.1%, OR = 1.458, 95% CI: 1.009–2.132) and winter (6.0% vs 3.1%, OR = 1.947, 95% CI:1.338–2.864), when comparing with that in spring. More information was showed in Table 2.

Table 2

The rate of preterm birth in different seasons of conception.
Season of conception	Simple	Preterm birth (%)	P	OR (95% CI)
Spring	1429	45 (3.1)	-	1
Summer	1745	79 (4.5)	0.047	1.458 (1.009–2.132)
Autumn	1584	66 (4.2)	0.140	1.337 (0.912–1.978)
Winter	1226	73 (6.0)	0.001	1.947 (1.338–2.864)

3.4. The exposure levels of the environmental factors during the first 30 days of gestation

The median values of concentrations of temperature, humidity, SO₂, NO₂, CO, O₃, PM₁₀, and PM_2.5 exposure during the first 30 days of gestation were 18.257 ℃, 52.908%, 12.935 µg/m³, 39.598 µg/m³, 818.214 µg/m³, 115.721 µg/m³, 106.496 µg/m³, and 42.892 µg/m³, respectively (Table 3). There was a statistically significant difference in maternal exposure to humidity between mothers who delivered preterm and those who did not (P = 0.015). The median and IQR were 53.805% (10.473%) in the preterm birth group, and 52.849% (10.656%) in the non-preterm birth group (Table 3). Except for humidity, there were no significant differences in temperature and six air pollutants during the first 30 days of pregnancy between the two groups (P > 0.05, Table 3).

Table 3

The exposure levels of environmental factors during the first 30 days of gestation between groups.
Environmental factor	Total		Mothers with preterm birth		Mothers without preterm birth		P
Environmental factor	Mean (SD)	Median (IQR)	Mean (SD)	Median (IQR)	Mean (SD)	Median (IQR)	P
Temperature (℃)	16.874 (9.623)	18.257 (17.145)	16.112 (10.031)	16.777 (20.543)	16.909 (9.603)	18.310 (16.920)	0.284
Humidity (%)	54.514 (7.482)	52.908 (10.634)	55.433 (6.965)	53.805 (10.473)	54.471 (7.503)	52.849 (10.656)	0.015
SO₂ (µg/m³)	15.100 (6.272)	12.935 (8.547)	15.115 (6.288)	12.938 (8.597)	14.788 (5.913)	12.732 (7.392)	0.750
NO₂ (µg/m³)	43.355 (12.376)	39.598 (21.154)	44.568 (12.601)	41.532 (20.725)	43.299 (12.364)	39.544 (21.127)	0.109
CO (µg/m³)	874.578 (209.310)	818.214 (314.384)	882.851 (209.097)	823.697 (311.200)	874.198 (214.138)	818.001 (314.984)	0.486
O₃ (µg/m³)	111.256 (45.139)	115.721 (78.314)	113.753 (44.78)	121.676 (75.320)	111.141 (45.156)	115.580 (78.538)	0.340
PM₁₀ (µg/m³)	109.892 (36.022)	106.496 (51.761)	111.933 (37.422)	107.657 (56.931)	109.798 (35.95)	106.421 (51.543)	0.484
PM_2.5 (µg/m³)	51.536 (21.056)	42.892 (32.265)	53.654 (22.153)	44.340 (42.434)	51.439 (21.001)	42.828 (31.744)	0.212

3.5. The relationship between seasons of conception and the risk of preterm birth

The results of multivariate analysis were shown in Table 4. The adjusted OR (95% CI) for the risk of preterm birth associated with conception in winter was 2.003 (1.374–2.951) in Model 1, 2.264 (1.539–3.368) in Model 2, and 2.228 (1.512–3.316) in Model 3, respectively, compared with conception in spring. The OR (95% CI) of conception in summer was 1.482 (1.024–2.169) in Model 1, 1.571 (1.077–2.320) in Model 2, and 1.541 (1.055–2.277) in model 3, respectively. And conception in autumn was not significantly associated with the risk of preterm birth (P > 0.05).

Table 4

Multifactor logistic regression models for the association between seasons of conception and the risk of preterm birth.
Season of conception	Model 1 ^a
Season of conception	P	OR (95% CI)	P	OR (95% CI)	P	OR (95% CI)
Spring		1.000 (reference)		1.000 (reference)		1.000 (reference)
Summer	0.039	1.482 (1.024–2.169)	0.021	1.571 (1.077–2.320)	0.027	1.541 (1.055–2.277)
Autumn	0.118	1.362 (0.927–2.018)	0.065	1.453 (0.980–2.172)	0.073	1.438 (0.970–2.151)
Winter	< 0.001	2.003 (1.374–2.951)	< 0.001	2.264 (1.539–3.368)	< 0.001	2.228 (1.512–3.316)

^a Adjusted for covariates related to basic maternal information: primipara, delivery method, age, housewife.

^b Adjusted for the covariates in Model 1, continued adjustment for covariates related to abnormal symptoms in pregnancy: pregnancy complications, pregnancy-induced hypertension, and abnormal pregnancy during pregnancy.

^c Adjusted for the covariates in Model 2, continued adjustment for covariates related to behavioural factors: passive smoking, vitamin D supplement, and exercise during pregnancy.

3.6. The roles of environmental factors on the association between seasons of conception and the risk of preterm birth

Based on the covariates in Model 3, we adjusted for the environmental factors to examine the relationship between conception in summer or winter and the risk of preterm birth, and the results were showed in Fig. 3. For conception in summer, the ORs and 95% CIs were 1.200 (0.782–1.850) adjusted for humidity, and 1.429 (0.836–2.483) adjusted for temperature, and 1.505 (0.988–2.317) adjusted for PM_2.5, and 1.536 (1.051–2.270) adjusted for O₃, and 1.408 (0.813–2.480) adjusted for humidity and temperature, and 1.263 (0.807–1.987) adjusted for humidity and PM_2.5, and 1.193 (0.778–1.841) adjusted for humidity and O₃, respectively. For conception in winter, the ORs and 95% CIs of the risk of preterm birth were 2.353 (1.594–3.511) adjusted for humidity, and 2.428 (1.356–4.438) adjusted for temperature, and 2.332 (1.379–3.995) adjusted for PM_2.5, and 2.222 (1.508–3.308) adjusted for O₃, and 1.887 (1.032–3.523) adjusted for humidity and temperature, and 2.025 (1.177–3.529) adjusted for humidity and PM_2.5, and 2.348 (1.590–3.503) adjusted for humidity and O₃, respectively. In addition, the OR (95% CI) of the risk of preterm birth associated with each 10% increase in humidity was 1.020 (1.003–1.038).

Humidity was positively correlated with temperature and negatively correlated with PM_2.5, but had no correlation with O₃. The correlation coefficient was 0.602 between humidity and temperature (P < 0.001), and − 0.576 between humidity and PM_2.5 (P < 0.001). And the correlation coefficient between humidity and O₃ was 0.012 (P = 0.400).

3.7. The result for the mediating effect of humidity

The relationship between the two among humidity, seasons of conception and the risk of preterm birth was showed in Fig. 4 and table S3. For the independent variable in each model, conception in spring and in autumn was the control group. There was no statistical significance in mediating effect of the humidity in the relationship between conception in summer and preterm birth (P > 0.05). The humidity partially mediated the relationship between conception in winter and preterm birth (Indirect effect = −0.007, P < 0.001), which was explained 41.2% of the relationship.

3.8. The result of the sensitivity analysis

To test the robustness of the results, we performed some sensitivity analysis, and the results were showed in Figure S3. After excluding clinics one by one, the effect of maternal conceiving in winter on preterm birth still existed (OR > 2). And conceiving in other seasons was still not associated with preterm birth.

To control for the effect of moving during pregnancy on the study results, the variable moving during pregnancy was adjusted for in the model, and the risk of winter pregnancy on preterm birth remained (aOR = 2.009, 95%CI 1.378–2.960). Secondly, after excluding study participants who moved during pregnancy, the analysis was repeated and winter pregnancy remained a risk factor for preterm birth (aOR = 2.094, 95%CI 1.422–3.123) (Table S4).

Although some studies have explored the association between season and preterm birth, most of them have used the season of delivery as the independent variable. As far as we know, only two studies have revealed the relationship between season of conception and preterm birth. A study from Pennsylvania, the United States, noted that mothers who conceived in autumn and summer were less likely to have a premature birth than those who conceived in spring and winter (autumn: OR = 0.91, 95% CI: 0.85–0.96; summer: OR = 0.92, 95% CI: 0.86–0.98) (14). Another Danish national cohort study found that a slight increase in risk was observed in autumn (AHR, 1.05; 95% CI, 1.02–1.09) compared with winter(24). Our finding was not quite the same as the result of the previous research. In the present study, mothers who conceived in winter were more likely to give birth prematurely after adjusting for environmental factors such as temperature, humidity and air pollutants (OR = 2.353, 95% CI: 1.594–3.511). Demonstrating that the effect of seasonal patterns on the risk of preterm birth cannot be attributed entirely to meteorological factors and air quality. Other factors such as race/ethnicity(26), lifestyle(27), and economic level(28) may also play an important role in the association between season and preterm birth. Therefore, there may be different seasonal patterns in the risk of preterm birth in different regions, and the association between seasonal and preterm birth risk should be further studied in more regions.

In this study, IDW method was used to assess the pollutant exposure levels of the subjects. It is widely used and has been employed in a number of studies to estimate the concentrations of major air pollutants to which subjects are exposed(29–31), with relatively smooth interpolation results for small measurement areas with relatively dense sampling points and small area variations(32). But IDW modelling method may cause an exposure bias as the number of monitoring stations was sparse with an ignorance in the land used conditions(33). In contrast, our research is a small area study and is located in an urban area with relatively consistent land use, making it more accurate and reasonable to use IDW to assess pollutant exposure concentrations.

Our results showed that the association between summer conception and preterm birth disappeared after adjusting for exposure levels of environmental factors other than O₃ in the first 30 days of pregnancy, but the association between winter conception and preterm birth persisted. In other words, the increased risk of preterm birth for mothers conceived during the summer can be attributed to environmental factors, of which humidity may have played a key role, as the association between summer conception and risk of preterm birth disappeared after adjusting for humidity as well as humidity-related temperature and PM_2.5. Meanwhile, O₃ was not correlated with humidity, adjusting for O₃ did not affect the association. Our results adjusting for humidity and one of the other environmental factors also supported this view. Some studies pointed that eclampsia and preeclampsia were more prevalent in hot, cold and humid seasons or months (34, 35). Based on the evidence to date, we conjecture that the humidity may affect the cardiovascular system of pregnant women, which further causes premature birth. Given that most studies on the relationship between environmental factors and preterm birth have focused on temperature, air pollutants, and greenery(13), the relationship between humidity and preterm birth and its mechanistic will require more evidence in the future. More importantly, after adjusting for environmental factors, the association between winter conception and risk of preterm birth persisted. To further explored the mechanisms, we analyzed the mediating effect of humidity on the relationship between seasons of conception and the risk of preterm birth. Interestingly, we found that humidity partially mediated the relationship between conception in winter and the risk of preterm birth, and more specifically, that exposure to low humidity in winter may increase the risk of preterm birth (proportion of mediation = 41.2%, indirect effect = −0.007, P < 0.001). Except for the humidity, there may be other reasons for the relationship between conception in winter and preterm birth. On the one hand, some researches showed that the virus was more likely to be transmitted in winter(36–38). Infections such as influenza infection and malaria in pregnancy might affect intrauterine growth and lead to preterm birth(39, 40). Therefore, we reckoned that a high risk of infection contributes to more preterm births in mothers who fertilize in winter than in other seasons. On the other hand, nutrition may be another reason for preterm birth among the women who conceive in winter pregnancies. When winter comes, changes in the diet of mothers, such as having a high-fat diet and difference in vegetable and fruit intake, may lead to a rapid weight gain of the mothers, affect the blood perfusion of the placenta in the future, and lead to premature birth(41, 42). Furthermore, lack of prenatal health care may lead to adverse birth outcomes. In winter, the end of the year, woman will be busier in that they need to sum up their work and prepare for the New Year and the Spring Festival. A study found that mothers who became pregnant in October or November were less likely to get prenatal care in the first trimester (43). It is necessary to explore the mechanism for the effect of conception in winter on preterm birth in further studies.

In this study, we chose the seasons based on the solar terms in the traditional Chinese calendar as the independent variables. In the traditional Chinese calendar, the solar year is divided equally into 24 parts, which was called the 24 solar terms. The solar terms are determined by the position of the sun on the returning ecliptic, the apparent path of the sun on the celestial sphere during the year. The position of the sun's annual trajectory is therefore relatively fixed for each of the solar terms. Lichun, Lixia, Liqiu and Lidong are when the sun reaches 315°, 45°, 135° and 225° of the ecliptic respectively, representing the beginning of spring, summer, autumn and winter. It scientifically reveals the laws of astronomical and meteorological changes in China, combines astronomy, agriculture, phenology and folklore skillfully(44–46), and is used to indicate climate change(46), to make decisions about the timing of crop sowing and harvesting(47), and derives a large number of seasonal cultures related to it. In contrast, the seasons cannot be accurately delineated according to the months, as the same date each year corresponds to a different annual trajectory of the sun. Some studies have examined the relationship between seasons or climate change based on the traditional Chinese calendar and human health (48–50). Therefore, we speculate that to explore the relationship between season based on solar terms and preterm birth could discover a more appropriate link of them in Chinese people.

There were mainly three advantages in the present study. First, we looked at the relationship between season and preterm birth in more detail and found for the first time that humidity partially mediated the relationship. It provides a reference for the following mechanism research. Second, compared with the birth season in other studies, the conception season based on traditional Chinese calendar solar terms was selected as the research variable in our study, which was more in line with the seasonal pattern of exposure of the study subjects, and was of more preventive significance and value in the exploration of etiology. Third, we considered more season-related covariates in this study. We assessed individual exposure levels for environmental factors, performed a differential analysis to select the factors that might affect preterm birth, and included them in the multivariate model for adjustment.

However, our study still had some limitations. The subjects were all mothers who had live babies, and infants who died in the perinatal period due to preterm delivery were not included in the study. Besides, some covariates such as maternal pregnancy information was reported only by relying on maternal recollections, which may have a certain degree of recall bias. Thirdly, due to the uneven distribution of monitoring stations, the assessment of exposure to environmental factors for study subjects located at a distance from the monitoring stations may not be sufficiently precise. In addition, as climatic factors, air quality and other factors vary from region to region, the conclusions of this study need to be extrapolated with caution.

For women living in temperate regions, conception in summer and winter may increase the risk of preterm birth. The increased risk of preterm birth during summer conception was strongly associated with environmental factors. However, in addition to the mediating effect of humidity, other unknown factors played a role in the increased risk of preterm birth during winter conception, and the possible causes of preterm birth in women during winter conception need to be further investigated.

OR Odds ratio

CI Confidence interval

SO₂ Sulfur dioxide

NO₂ Nitrogen dioxide

CO Carbon monoxide

O₃ Ozone

PM₁₀ Particulate matter with an aerodynamic diameter of less than 10 mm

PM_2.5 Particulate matter with an aerodynamic diameter of less than 2.5 mm

IDW Inverse distance weighting

SD Standard deviation

IQR Interquartile range

VIF Variance inflation factors

Ethics approval and consent to participate

The study was approved by the Ethics Committee of Preventive Medicine of Shandong University (Approved number: 20170315). Informed consent was obtained from all subjects and/or their legal guardian(s). All methods were carried out in accordance with relevant guidelines and regulations.

Consent for publication

Not applicable.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests

The authors declare that they have no competing interests.

Funding

This study was funded by the National Natural Science Foundation of China (81773386).

Author Contributions

Yuxiu Liang and Shujing Zou: Investigation, Software, Writing-Original draft preparation. Zhaojun Wang: Resources. Shuoxin Bai: Investigation, Visualization, Data Curation. Shuang Du: Investigation, Software. Shaoqian Lin and Xiaodong Zhao: Supervision, Investigation. Zhiping Wang: Project administration, Methodology, Conceptualization.

Acknowledgments

Thanks to the Jinan Center for Disease Control and Prevention and the staff of the community vaccination clinic for the investigation as well as the Jinan Ecological Environment Bureau for reviewing environmental data. We also thank all mothers and infants for their cooperation.

WHO. WHO. recommended definitions, terminology and format for statistical tables related to the perinatal period and use of a new certificate for cause of perinatal deaths. Modifications recommended by FIGO as amended October 14, 1976. Acta Obstet Gynecol Scand. 1977;56(3):247–53.
Vogel JP, Chawanpaiboon S, Moller AB, Watananirun K, Bonet M, Lumbiganon P. The global epidemiology of preterm birth. Best Pract Res Clin Obstet Gynaecol. 2018;52:3–12.
Kwinta P, Pietrzyk JJ. Preterm birth and respiratory disease in later life. Expert Rev Respir Med. 2010;4(5):593–604.
Johnson S, Marlow N. Preterm birth and childhood psychiatric disorders. Pediatr Res. 2011;69(5 Pt 2):11R–8R.
Span LC, van Dokkum NH, Ravensbergen A-G, Bos AF, Jaschke AC. Combining Kangaroo Care and Live-Performed Music Therapy: Effects on Physiological Stability and Neurological Functioning in Extremely and Very Preterm Infants. Int J Environ Res Public Health. 2021;18(12).
Beck S, Wojdyla D, Say L, Betran AP, Merialdi M, Requejo JH, et al. The worldwide incidence of preterm birth: a systematic review of maternal mortality and morbidity. Bull World Health Organ. 2010;88(1):31–8.
Blencowe H, Cousens S, Oestergaard MZ, Chou D, Moller AB, Narwal R, et al. National, regional, and worldwide estimates of preterm birth rates in the year 2010 with time trends since 1990 for selected countries: a systematic analysis and implications. Lancet. 2012;379(9832):2162–72.
Chang YK, Tseng YT, Chen KT. The epidemiologic characteristics and associated risk factors of preterm birth from 2004 to 2013 in Taiwan. BMC Pregnancy Childbirth. 2020;20(1):201.
Waldenstrom U. Adverse Pregnancy Outcomes Related to Advanced Maternal Age Compared With Smoking and Being Overweight (vol 123, pg 104, 2014). Obstet Gynecol. 2014;123(3):669-.
Carolan M. Maternal age >/=45 years and maternal and perinatal outcomes: a review of the evidence. Midwifery. 2013;29(5):479–89.
Cnattingius S. The epidemiology of smoking during pregnancy: smoking prevalence, maternal characteristics, and pregnancy outcomes. Nicotine Tob Res. 2004;6(Suppl 2):125–40.
Purisch SE, DeFranco EA, Muglia LJ, Odibo AO, Stamilio DM. Preterm birth in pregnancies complicated by major congenital malformations: a population-based study. Am J Obstet Gynecol. 2008;199(3):287e1–8.
Kloog I. Air pollution, ambient temperature, green space and preterm birth. Curr Opin Pediatr. 2019;31(2):237–43.
Bodnar LM, Simhan HN. The prevalence of preterm birth and season of conception. Paediatr Perinat Epidemiol. 2008;22(6):538–45.
Flouris AD, Spiropoulos Y, Sakellariou GJ, Koutedakis Y. Effect of seasonal programming on fetal development and longevity: links with environmental temperature. Am J Hum Biol. 2009;21(2):214–6.
Woo Y, Ouh YT, Ahn KH, Cho GJ, Hong SC, Oh MJ, et al. Seasonal Pattern of Preterm Births in Korea for 2000–2012. J Korean Med Sci. 2016;31(11):1797–801.
Raglan GB, Lannon SM, Jones KM, Schulkin J. Racial and Ethnic Disparities in Preterm Birth Among American Indian and Alaska Native Women. Matern Child Health J. 2016;20(1):16–24.
Uphoff EP, Pickett KE, Wright J. Social gradients in health for Pakistani and White British women and infants in two UK birth cohorts. Ethn Health. 2016;21(5):452–67.
Cai J, Zhao Y, Kan J, Chen R, Martin R, van Donkelaar A, et al. Prenatal Exposure to Specific PM2.5 Chemical Constituents and Preterm Birth in China: A Nationwide Cohort Study. Environ Sci Technol. 2020;54(22):14494–501.
Bai S, Du S, Liu H, Lin S, Zhao X, Wang Z, et al. The causal and independent effect of ozone exposure during pregnancy on the risk of preterm birth: Evidence from northern China. Environ Res. 2022;214(Pt 1):113879.
Bekkar B, Pacheco S, Basu R, DeNicola N. Association of Air Pollution and Heat Exposure With Preterm Birth, Low Birth Weight, and Stillbirth in the US: A Systematic Review. JAMA Netw open. 2020;3(6):e208243.
Cichowicz R, Wielgosinski G, Fetter W. Dispersion of atmospheric air pollution in summer and winter season. Environ Monit Assess. 2017;189(12):605.
Du S, Bai S, Zhao X, Lin S, Zhai Y, Wang Z, et al. The effect and its critical window for ambient temperature and humidity in pregnancy on term low birth weight. Environmental science and pollution research international; 2022.
Hviid A, Laksafoss A, Hedley P, Lausten-Thomsen U, Hjalgrim H, Christiansen M, et al. Assessment of Seasonality and Extremely Preterm Birth in Denmark. JAMA Netw open. 2022;5(2):e2145800.
Davenport MH, Meah VL, Ruchat S-M, Davies GA, Skow RJ, Barrowman N, et al. Impact of prenatal exercise on neonatal and childhood outcomes: a systematic review and meta-analysis. Br J Sports Med. 2018;52(21):1386–96.
Smid MC, Lee JH, Grant JH, Miles G, Stoddard GJ, Chapman DA, et al. Maternal race and intergenerational preterm birth recurrence. Am J Obstet Gynecol. 2017;217(4):480. .e1-.e9.
Zhu Y, Hedderson MM, Brown SD, Badon SE, Feng J, Quesenberry CP, et al. Healthy preconception and early-pregnancy lifestyle and risk of preterm birth: a prospective cohort study. Am J Clin Nutr. 2021;114(2):813–21.
Webster JL, Paul D, Purtle J, Locke R, Goldstein ND. State-Level Social and Economic Policies and Their Association With Perinatal and Infant Outcomes. Milbank Q. 2022;100(1):218–60.
Deng Q, Lu C, Norbäck D, Bornehag C-G, Zhang Y, Liu W, et al. Early life exposure to ambient air pollution and childhood asthma in China. Environ Res. 2015;143(Pt A):83–92.
Lu C, Zhang Y, Li B, Zhao Z, Huang C, Zhang X, et al. Interaction effect of prenatal and postnatal exposure to ambient air pollution and temperature on childhood asthma. Environ Int. 2022;167:107456.
Coker ES, Martin J, Bradley LD, Sem K, Clarke K, Sabo-Attwood T. A time series analysis of the ecologic relationship between acute and intermediate PM2.5 exposure duration on neonatal intensive care unit admissions in Florida. Environ Res. 2021;196:110374.
Rishikeshan CA, Katiyar SK, Mahesh VNV, Ieee, editors. Detailed evaluation of DEM interpolation methods in GIS using DGPS data. 6th International Conference on Computational Intelligence and Communication Networks; 2014 Nov 14–16; Bhopal, INDIA2014.
Jones KH. A comparison of algorithms used to compute hill slope as a property of the DEM. Comput Geosci. 1998;24(4):315–23.
Poursafa P, Keikha M, Kelishadi R. Systematic review on adverse birth outcomes of climate change. J Res Med Sci. 2015;20(4):397–402.
Tam WH, Sahota DS, Lau TK, Li CY, Fung TY. Seasonal variation in pre-eclamptic rate and its association with the ambient temperature and humidity in early pregnancy. Gynecol Obstet Invest. 2008;66(1):22–6.
Moriyama M, Hugentobler WJ, Iwasaki A. Seasonality of Respiratory Viral Infections. Annu Rev Virol. 2020;7(1):83–101.
Chadha M, Hirve S, Bancej C, Barr I, Baumeister E, Caetano B, et al. Human respiratory syncytial virus and influenza seasonality patterns-Early findings from the WHO global respiratory syncytial virus surveillance. Influenza Other Respir Viruses. 2020;14(6):638–46.
Tamerius JD, Shaman J, Alonso WJ, Bloom-Feshbach K, Uejio CK, Comrie A, et al. Environmental predictors of seasonal influenza epidemics across temperate and tropical climates. PLoS Pathog. 2013;9(3):e1003194.
Steinhoff MC, Omer SB, Roy E, El Arifeen S, Raqib R, Dodd C, et al. Neonatal outcomes after influenza immunization during pregnancy: a randomized controlled trial. CMAJ. 2012;184(6):645–53.
Hills T, Sharpe C, Wong T, Cutfield T, Lee A, McBride S, et al. Fetal Loss and Preterm Birth Caused by Intraamniotic Haemophilus influenzae Infection, New Zealand. Emerg Infect Dis. 2022;28(9):1749–54.
Morgan TK. Role of the Placenta in Preterm Birth: A Review. Am J Perinatol. 2016;33(3):258–66.
Lawlor DA, Leon DA, Davey Smith G. The association of ambient outdoor temperature throughout pregnancy and offspring birthweight: findings from the Aberdeen Children of the 1950s cohort. BJOG. 2005;112(5):647–57.
Currie J, Schwandt H. Within-mother analysis of seasonal patterns in health at birth (vol 110, pg 12265, 2013). P Natl Acad Sci USA. 2013;110:30.
Wang S, Xu M, Song Aw, Dai J. Climate changes in 24 solar terms and responses of winter wheat and single-season rice phenophases in Anhui Province of East China in 1961–2012. Shengtaixue Zazhi. 2013;32(9):2390–7.
Jia Q, Zhang H, Jiang S, Qiao T, Qin Q, Hua M. Establishment and analysis of database of birth date and easy to emergency department-solar terms in Beijing. China J Traditional Chin Med Pharm. 2021;36(8):5037–67.
Ji C, Wang S, Shang K. Climate Changes of 24 Solar Terms in the Middle and Lower Reaches of the Yellow River during 1961–2010. Clim Environ Res. 2015;20(2):220–8.
Lu LN, Zhang JJ, Wu FH, Xie GF, Shan ZC, Liu XQ. Flavor profile variations of Huangjiu brewed in different traditional Chinese solar terms. Food Qual Saf. 2022;6.
Wu X, Peng J, Li G, Zhang W, Liu G, Liu B. Efficacy evaluation of summer acupoint application treatment on asthma patients: a two-year follow-up clinical study. J Tradit Chin Med. 2015;35(1):21–7.
Khezri B, Chan YY, Tiong LY, Webster RD. Annual air pollution caused by the Hungry Ghost Festival. Environ Sci Process Impacts. 2015;17(9):1578–86.
Xu T, Wang Q. On the formation of taiji diagram from the perspective of twenty-four solar terms. Saudi J Biol Sci. 2018;25(8):1670–7.

No competing interests reported.

Supplementarymaterial.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Role for environmental factors on the relationship between seasons of conception and preterm birth: An observational study in China

Status:

Version 1

Abstract

Figures

1. Background

2. Methods

2.1. Study population

2.2. Dependent variable and independent variable

2.3. Assessment of exposure to environmental factors

2.4. Covariates

2.5. Statistical analysis

2.6. Sensitive analysis

2.7. Quality control

3. Results

3.1. The characteristics of the subjects

3.2. The distributions of preterm birth in the population

3.3. Preterm birth in different seasons of conception

3.4. The exposure levels of the environmental factors during the first 30 days of gestation

3.5. The relationship between seasons of conception and the risk of preterm birth

3.7. The result for the mediating effect of humidity

3.8. The result of the sensitivity analysis

4. Discussion

5. Conclusions

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1