Figure shows the distribution of total caffeine intake as registered in the FFQ. Coffee, black tea, soft drinks and chocolate accounted for more than 98% of daily caffeine intake but, interestingly, the dominant source differed in the low-intake and high-intake groups, with chocolate dominating in the first quintile, black tea in the second and third quintiles and coffee in the upper quintiles (Figure ). Self-reported pre-pregnancy caffeine intake from coffee, black tea and soft drinks in Q1 and Q3 revealed a median intake of 126 mg/day (IQR 40 to 254 mg/day) for all 59,123 women, including 7,406 women who did not consume any caffeine at all. At gestational week 17 the number of non-consumers was nearly doubled (14,012 women) and the median caffeine intake had decreased to 44 mg/day (13 to 104 mg/day). At gestational week 30, the median caffeine intake had increased again to 62 mg/day (21 to 130 mg/day) and 9,792 women remained non-consumers. Caffeine intake related to maternal characteristics is presented in Table : older, unmarried and smoking women with higher parity, history of PTD, lower pre-pregnancy BMI, less nausea, higher energy intake and higher household income had significantly higher caffeine intake.
Percentage of total caffeine intake per caffeine source. This figure shows the percentage of total caffeine intake per caffeine source (food frequency questionnaire data), n = 59,123, in the Norwegian Mother and Child Cohort Study 2002 to 2009.
Figure 2 Sources of caffeine intake according to quintiles of total caffeine intake. This figure shows sources of caffeine intake (food frequency questionnaire data) according to quintiles of total caffeine intake, n = 59,123, in the Norwegian Mother and Child (more ...)
Caffeine intake according to maternal characteristics
Gestational length and spontaneous PTD
A total of 49,102 women delivered spontaneously with a median gestational length of 282 days (IQR 276 to 287 days). Gestational length as well as its residuals were left skewed, but we preferred to use the original data scale for easier interpretation.
Caffeine intake from different sources (FFQ data)
Total caffeine intake was associated with slightly increased gestational length, that is, 5 h/100 mg/day (95% CI 3 to 8 h, P <10-4) (Table ). However, linear regression with all different caffeine sources included in the same model revealed that only coffee caffeine was significantly associated with gestational length. When the different caffeine sources were studied individually, it emerged that the association for total caffeine intake resembled that for coffee caffeine intake (8 h/100 mg coffee caffeine/day, 95% CI 5 to 10 h, P <10-7). If all sources were studied individually without mutual adjustment, only coffee caffeine remained significantly associated with gestational length (8 h/100 mg coffee caffeine/day; 95% CI 5 to 10 h, P <10-6). As we found that coffee is the dominant source of caffeine in high-caffeine consumers, these findings could be explained by a threshold model implying that only coffee drinkers reach the threshold associated with altered gestational length. To rule out this possibility, we compared the coffee caffeine intake categorized into five groups (no intake and for the remaining subjects quartiles 0 to 8.38, 8.39 to 40.71, 40.72 to 110.52, >110.52 mg/day) finding that compared to the fifth group even group one and two had a significantly shorter gestational length (first group: regression coefficient β = -3.2, P = 0.04, second group: β = -4.2, P = 0.02). After excluding all coffee consumers from the analysis, black tea and chocolate were still not associated with gestational length while caffeinated soft drinks were associated with a 13 h decreased gestational length/100 mg additional caffeine/day (95% CI 1 to 24 h, P = 0.032) in the remaining 17,491 women. In the subgroup of non-coffee consumers, total caffeine intake was significantly associated with 10 h decreased gestational length/100 mg additional caffeine/day (95% CI 1 to 18 h, P = 0.017, adjusted models). When performing the linear regression in only non-smokers (n = 45,053), coffee caffeine was still the only caffeine source significantly associated with gestational length (total caffeine intake 7 h/100 mg caffeine/day, 95% CI 4 to 10 h, P <10-5, coffee caffeine 10 h/100 mg caffeine/day, 95% CI 7 to 13 h, P <10-9, adjusted models).
Gestational length in pregnancies with spontaneous delivery and caffeine intake from different sources
There were 1,451 cases of spontaneous PTD in the study population (240 early spontaneous PTDs and 1,211 late spontaneous PTDs), compared to 27,498 controls, according to our strict inclusion and exclusion criteria. There was no significant association between total or coffee caffeine intake and the odds for overall, early or late spontaneous PTD (Table ). Black tea caffeine was associated with increased risk of early spontaneous PTD (OR 1.61, 95% CI 1.10 to 2.35, P = 0.01, adjusted model).
Odds for spontaneous preterm delivery (PTD) and caffeine intake from different sources
Caffeine intake over time (Q1 and Q3 data)
At all timepoints studied, total and coffee caffeine intake was consistently associated with increased gestational length. The association was strongest for caffeine consumption reported at gestational week 17 (3 h/100 mg total caffeine/day, 95% CI 1 to 4 h; 4 h/100 mg coffee caffeine/day, 95% CI 2 to 5 h, both P <0.001), followed by that reported at gestational week 30 (2 h/100 mg total caffeine/day, 95% CI 0 to 3 h, P = 0.02; 3 h/100 mg coffee caffeine/day, 95% CI 1 to 5 h, P <0.001) and reported pre-pregnancy caffeine consumption (2 h/100 mg total caffeine/day, 95% CI 1 to 3 h; 2 h/100 mg coffee caffeine/day, 95% CI 1 to 3 h, both P <10-6); all adjusted models.
Birth weight and SGA
The diagnosis of SGA varied considerably depending on the growth curve and SGA definition applied (Figure ).
Overlap of small for gestational age (SGA) definitions according to Marsal (ultrasound based), Skjaerven (population based) and Gardosi (customized), n = 59,123, in the Norwegian Mother and Child Cohort Study 2002 to 2009.
Caffeine intake from different sources (FFQ data)
Total caffeine intake, as well as caffeine intake from the individual sources, was associated with lower BW (Table ). The dependent variable in the linear regression was the difference between reported actual BW and expected BW, calculated as percentage of the expected BW. For easier understanding and interpretation, results are presented as a change in BW per 100 mg additional caffeine/day for a child with an expected BW of 3,600 g, the rounded-out median of BW in our study population (median 3,620 g). In the adjusted model, intake of an additional 100 mg total caffeine/day was associated with a 21 to 28 g BW decrease, depending on the growth curve. The opposite effect of chocolate caffeine in the Gardosi model was no longer significant after adjustment.
Birth weight and caffeine intake from different sources
When studied exclusively in non-smokers (n = 54,136), these associations remained significant, again with the exception of chocolate caffeine in the Gardosi model. However, the decrease in BW was somewhat lower: Marsal 18 g (95% CI 15 to 21 g) instead of 28 g, Skjaerven 15 g (95% CI 12 to 18 g) instead of 25 g, Gardosi 12 g (95% CI 9 to 15 g) instead of 21 g per 100 mg additional total caffeine/day (all significant with P <10-15; adjusted models).
Total and coffee caffeine intake was significantly associated with higher odds for SGA, based on logistic regression in all three SGA models, both unadjusted and adjusted (Table ). Energy drink and black tea caffeine intake were associated with a significant increase in two of the three SGA models, while there was no significant association with chocolate caffeine. The association of total caffeine intake and BW remained significant when analyses were limited to the non-smoker subgroup (n = 54,136).
Odds for small for gestational age (SGA) and caffeine intake from different sources
To test if there was a threshold effect, we performed the same logistic regression with sextiles of total caffeine intake (0 to 14.645, 14.646 to 32.093, 32.094 to 57.265, 57.266 to 96.029, 9603 to 163.806, >163.806 mg/day). In all three models the caffeine intake categories were associated with increasing odds ratios for SGA as compared to the lowest intake group (see Figure ). According to FFQ data, 10.8% of all women exceeded the NNR recommendation of less than 200 mg/day caffeine intake during pregnancy and 3.3% also exceeded the WHO recommendation of less than 300 mg/day. If the odds for SGA were studied with reference to these recommendations, those 7.7% of women with a daily caffeine intake of 200 to 300 mg had significantly higher odds for SGA (1.27 to 1.62, depending on the SGA definition), in comparison with the lowest (0 to 50 mg/day) caffeine intake group. The odds of giving birth to a SGA infant were 1.62 to 1.66 in the 3.3% of women consuming >300 mg caffeine/day (Table ).
Figure 4 Small for gestational age (SGA) risk depending on total caffeine intake. Relative risk for SGA in sextiles of total caffeine intake with the lowest sextile as reference category (0 to 14.645, 14.646 to 32.093, 32.094 to 57.265, 57.266 to 96.029, 9603 (more ...)
Odds for small for gestational age (SGA) and total caffeine intake according to official guidelines
Caffeine intake over time (Q1 and Q3 data)
Total and coffee caffeine intake at all timepoints studied was significantly associated with decreased BW for all applied SGA models. The association with caffeine intake from black tea was the strongest with the Skjaerven and Marsal models. Tea caffeine was not significantly associated with BW if defined according to Gardosi, though. For caffeine from coffee and soft drinks, intake reported at gestational week 17 had the strongest impact on BW (Table ).
Birth weight and maternal caffeine intake at different timepoints before and during pregnancy