Table 2 Climate factors and GDM study table

Meek, 2020 [37]

n = 23,375 United Kingdom

Cohort: enrolled pregnant women with singleton pregnancies from 1/2004 to 12/2008

30.7 ± 5.6 years

A) Season: day of delivery

B) Ambient temperature (°C): daily mean temp on day of screening

2-step approach: 1) 50-g GCT at 28 weeks, if > 7.7 mmol/l

2) 75-g OGTT

OGTT criteria:

WHO 1999 (1/2004–8/2007)

Modified WHO 1999 (8/2007–12/2008)

Maternal age, BMI, parity, ethnicity

•GDM incidence varied significantly by day of glucose screening throughout the year (p = 0.031). GDM incidence was highest in births during Sept/Oct and lowest in births during March

•Daily mean temperature on day of GCT screening were associated with increased risk of abnormal GCT (OR 1.21, 95% CI: 1.10, 1.32 per 5 °C increase) and increased odds of GDM (OR 1.13, 95% CI: 1.02, 1.25 per 5 °C increase)

Molina-Vega, 2020 [38]

n = 2366 Malaga, Spain

Cohort: retrospective cohort of women referred to a Pregnancy and Diabetes clinic for GDM screening

32 ± 5.2 years

A) Ambient temperature:

1) mean ∆ temp 2) mean temp

Day of OGTT, 14 days pre-OGTT, and 28 days pre-OGTT

B) Season Winter (Dec 21st – Mar 20th) Spring (Mar 21st – Jun 20th) Summer (Jun 21st – Sep 20th) Autumn (Sep 21st – Dec 20th)

NDDG criteria

Maternal age

•Odds of GDM were highest in summer (OR 1.78 CI: 1.34, 2.37) compared to autumn

•Higher mean temperature on the day of OGTT screening and 14- and 28-days pre-OGTT were associated with increased risk of GDM diagnosis (e.g. Mean temp on day of OGTT: OR 1.03, 95% CI: 1.01, 1.05)

•When stratified, these associations were only present in the seasons where temperatures were increasing (Mar-Aug)

Su, 2020 [21]

n = 371,131 Taiwan

Cohort: population-based cohort study of pregnant women with deliveries between 2013 and 2014 in Taiwan

Not provided

A) Season

B) Ambient temperature (°C): 1) mean temp

2) daily temp ∆. Mean temperature: day of OGTT, 7, 14, 21, 28, 35 days pre-OGTT. Temperature ∆: daily difference between min and max temp on OGTT day and average ∆ 7, 14, 21, 28, 35 days pre-OGTT

IADPSG & Carpenter and Coustan criteria

Maternal age

•Age-adjusted odds of GDM were highest is summer (OR 1.05, 95% CI: 1.04, 1.07) and fall (OR 1.04, 95% CI: 1.02, 1.06) compared to winter

•Increased mean daily temperature (per 1 °C increase) was associated with increased age-adjusted odds of GDM for mean temperatures between 14 and 17 °C (OR 1.03, 95% CI: 1.02, 1.03) and even more strongly for temperatures between 28 and 30 °C (OR 1.54, 95% CI: 1.48, 1.60)

•Increased daily temperature difference (per 1 °C increase) was associated with lower odds of GDM (OR 0.90, 95% CI: 0.87, 0.92)

Petry, 2019 [39]

n = 1074. Cambridge, United Kingdom

Cohort: Cambridge Baby Growth Study, enrolled pregnant women during early pregnancy between 4/2001–3/2009

33.4 years

Season: Winter (Dec-Feb) Spring (Mar-May) Summer (Jun-Aug) Autumn (Sep-Nov)

IADPSG (based on fasting and 1 h only)

•Season of OGTT was not associated with GDM

Shen, 2019 [40]

n = 2120. Brisbane & Newcastle, Australia

Cohort: Women enrolled at Australian sites of the Hyperglycemia and Adverse Pregnancy Outcome (HAPO) from 2001 to 2006

29.6 ± 5.4 years

Season: Winter (Jun-Aug) Spring (Sep-Nov) Summer (Dec-Feb) Autumn (Mar-May)

WHO criteria

•No significant difference in GDM prevalence by season

n = 1464. Toronto, Canada

Cohort: enrolled pregnant women at time of GDM screening

34 ± 4 years

Ambient temperature (°C): 1) mean temp

2) daily temp change. Mean temperature: day of OGTT, 7, 14, 21, 28, 35, 42, 49, 56 days pre-OGTT. Temperature ∆: daily difference between min and max temp on OGTT day and average ∆ 7, 14, 21, 28, 35, 42, 49, 56 days pre-OGTT

NDDG criteria (All women received OGTT)

Maternal age, ethnicity, FH of diabetes, pp-BMI, GWG up to OGTT, weeks gestation at OGTT

•Temperature ∆ was associated with increased risk of GDM, only in the season where daily temperature was increasing

•For example, in Feb-July temperature ∆ in the preceding 14 days was associated with GDM (OR 1.20, 95% CI: 1.05, 1.37)

Vasileiou, 2018 [27]

A) n = 7618

B) n = 768

Athens, Greece

Two cohorts: A) Retrospective cohort: pregnant women who underwent a 100 g OGTT from 2002 to 2012.

B) Prospective cohort: pregnant women enrolled in 3rd trimester followed over 18 month period from 1/2013–6/2014.

Not provided

A) Season

B) Ambient Temperature: 1) Mean monthly temperature

2) Daily temperature @ 9 am

Three temp groups: 1) < 24.9 °C

2) 25–29.9 °C

3) > 30 °C

Study A: Carpenter and Coustan criteria

Study B: IADPSG criteria

Unadjusted

Study A:

•Odds of GDM were significantly higher in summer compared to winter (OR 1.65. 95% CI: 1.43, 1.90)

Study B:

•Temperature was not associated with GDM

Chiefari, 2017 [42]

n = 5473, Calabria, Italy

Cohort: Study population formed based on women who underwent an OGTT for GDM screening at a hospital in Calabria, Italy from 8/2011–12/2016.

33 (29–36) years

Seasons: Fall, Winter, Spring, Summer, Warm half & cold half of the year

IADPSG criteria

Unadjusted

•GDM incidence was significantly higher in summer (33.7%) and lower in the winter (23.3%) compared to the spring and fall

•GDM incidence was significantly lower in the cold (< 15 °C; 24.2%) compared to warm (> 15 °C; 31.4%) half of the year

Booth, 2017 [43]

n = 555,911, Toronto, Canada

Cohort: study population formed of births in greater Toronto area from 4/1/2002–3/31/2014 from administrative health databases.

30.9 ± 5.4 years

Ambient temperature: Average temperature 30-days pre-GDM screening (27 weeks)

ICD-10-CA codes (E10, E11, E13, E14, O24) or ≥ 2 diabetes insurance claims in the last 120 days of pregnancy

Maternal age, parity, neighborhood income, world region, year

•Significant association between higher ambient temperature and greater odds of GDM

•Each 10 °C increase in mean 30-day temp associated with a 6% increased odds of GDM (OR 1.06, 95% CI: 1.04–1.07)

Katsarou, 2016 [25]

n = 11,538, Skane county, Sweden

Cohort: Mamma Study, recruited women from 4 obstetric delivery departments in Skane county, Sweden from 2003 to 2005.

29.9 ± 5.1 years

Seasons: Winter (Dec-Feb) Spring (Mar-May) Summer (June-Aug) Fall (Sept-Nov) Mean monthly ambient temperature

WHO (1999) criteria, 2 h OGTT threshold

Maternal age

•GDM frequency differed significantly by month and season (highest in June/Summer and lowest in March/Spring)

•OGTT during summer was associated with increased frequency of GDM compared to all other seasons (OR 1.51, 95% CI: 1.24–1.83)

Verburg, 2016 [44]

n = 60,30, South Australia

Cohort: women with singleton births from South Australian Perinatal Statistics Collection (SAPSC) data from 2007 to 2011.

< 20 to > 40

Estimated date of conception (eDoC) Based on birth date and gestational age at birth (dating ultrasound and/or LMP)

*Note Australian Summer (Dec-Feb) Winter (June-Aug)

ADIPS (1998) criteria

Maternal age, BMI, parity, ethnicity, socioeconomic status, chronic hypertension

•GDM was significantly associated with season of eDoC (p < 0.001)

•Adjusted incidence of GDM was highest in pregnancies with eDoC in August (6.6%) and lowest in pregnancies with eDoC in January (5.41%)

Moses, 2016 [45]

n = 7343, Wollongong, Australia

Cohort: pregnant women with OGTT medical record data during 2012–2014 from both public and private pathology labs in the Wollongong, Australia area.

Not provided

Seasons: Summer (Dec-Feb) Fall, Winter, Spring

Modified WHO (2006) criteria

Unadjusted

•Prevalence of GDM was 28% lower in winter and 29% higher in summer, compared to the overall prevalence (p = 0.002)

Janghorbani, 2006 [46]

n = 4852, Plymouth, United Kingdom

Cohort: study population based on pregnant women in Plymouth, UK screened for GDM between 1/1996–12/1997 using data from Plymouth Child Health Database and laboratory and midwifery notes.

GDM: 30.9 ± 5.5 years, Non-GDM: 28.1 ± 5.4 years

Month and season

Modified WHO (1999) criteria

Maternal age, random plasma glucose, infant sex

•The prevalence of GDM was highest in June (2.9%) and Spring (2.3%) and lowest in November (1.1%) and Winter(1.4%), but the differences were not statistically significant (p = 0.82, month; p = 0.41, season)

Moses, 1995 [26]

n = 2749, Wollongong, Australia

Cohort: study population based on women with available OGTT data collected from clinics and obstetric offices from 1/1993 to 6/1994.

27 ± 5.1 years

Month & Season: Summer, Fall, Winter, Spring. Mean monthly ambient temperature (measured @ 9 am) *Note Australian Summer (Dec-Feb) Winter (June-Aug)

ADIPS (1991) criteria

Unadjusted

•Month/season and temperature were not associated GDM