Accepted Manuscript Prenatal Depression and Birth Mode Sequentially Mediate Maternal Education’s Influence on Infant Sleep Duration Brittany A. Matenchuk, Sukhpreet K. Tamana, Wendy Y.W. Lou, Diana L. Lefebvre, Malcolm R. Sears, Allan B. Becker, Meghan B. Azad, Theo J. Moraes, Stuart E. Turvey, Padmaja Subbarao, Anita L. Kozyrskyj, Piush J. Mandhane, CHILD Study Investigators PII: S1389-9457(18)30445-3 DOI: https://doi.org/10.1016/j.sleep.2019.01.015 Reference: SLEEP 3960 To appear in: Sleep Medicine Received Date: 20 July 2018 Revised Date: 15 November 2018 Accepted Date: 10 January 2019 Please cite this article as: Matenchuk BA, Tamana SK, Lou WYW, Lefebvre DL, Sears MR, Becker AB, Azad MB, Moraes TJ, Turvey SE, Subbarao P, Kozyrskyj AL, Mandhane PJ, CHILD Study Investigators, Prenatal Depression and Birth Mode Sequentially Mediate Maternal Education’s Influence on Infant Sleep Duration, Sleep Medicine, https://doi.org/10.1016/j.sleep.2019.01.015. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. ACCEPTED MANUSCRIPT 1 Title: Prenatal Depression and Birth Mode Sequentially Mediate Maternal Education’s Influence on 2 Infant Sleep Duration 3 Authors: Brittany A. Matenchuk1, Sukhpreet K. Tamana1, Wendy Y.W. Lou2, Diana L. Lefebvre3, 5 Malcolm R. Sears3, Allan B. Becker4, Meghan B. Azad4, Theo J. Moraes5, Stuart E. Turvey6, Padmaja 6 Subbarao5, CHILD Study Investigators, Anita L. Kozyrskyj1* and Piush J. Mandhane1* 7 Author Affiliations: 1 Department of Pediatrics, University of Alberta, Edmonton, AB, Canada 8 2 Dalla Lana School of Public Health, University of Toronto, Toronto, Canada 9 3 Department of Medicine, McMaster University, Hamilton, Canada 10 4 Department of Pediatrics & Child Health, Children’s Hospital Research Institute of Manitoba, 11 University of Manitoba, Winnipeg, MB, Canada 12 5 Department of Pediatrics, Hospital for Sick Children, University of Toronto, Toronto, ON, Canada 13 6 Department of Pediatrics, Child & Family Research Institute, BC Children’s Hospital, University of 14 British Columbia, Vancouver, BC, Canada 15 *Senior co-authors TE D M AN US C RI PT 4 16 Corresponding Author: Piush Mandhane, Department of Pediatrics, University of Alberta, 4-468 18 Edmonton Clinic Health Academy, 11405-87th Avenue, Edmonton, AB T6G 1C9, Canada 19 P: 780-248-5650 20 E-mail address: mandhane@ualberta.ca 22 23 24 25 26 AC C 21 EP 17 ACCEPTED MANUSCRIPT Author Contributions 28 Drs. Kozyrskyj and Mandhane had full access to all the data in the study and takes responsibility for the 29 integrity of the data and the accuracy of the data analysis. 30 Study concept and design: Matenchuk, Mandhane, Kozyrskyj. 31 Acquisition, analysis, or interpretation of data: Matenchuk, Tamana, Lou, Becker, Mandhane, Turvey, 32 Subbarao, Sears, Kozyrskyj. 33 Drafting of the manuscript: Matenchuk, Kozyrskyj 34 Critical revision of the manuscript for important intellectual content: Matenchuk, Lou, Becker, 35 Mandhane, Turvey, Subbarao, Moraes, Azad, Sears, Kozyrskyj, Tamana. 36 Statistical analysis: Matenchuk 37 Obtained funding: Becker, Mandhane, Turvey, Subbarao, Sears, Kozyrskyj. 38 Administrative, technical, or material support: Matenchuk, Lou, Becker, Mandhane, Turvey, Subbarao, 39 Lefebvre, Kozyrskyj. 40 Study supervision: Kozyrskyj, Mandhane D M AN US C RI PT 27 TE 41 All sources of support: This research was specifically funded by the CIHR Canadian Microbiome 43 Initiative (Grant No. 227312). The Canadian Institutes of Health Research (CIHR) and the Allergy, Genes 44 and Environment (AllerGen) Network of Centres of Excellence provided core support for the CHILD 45 study. 46 Declarations of interest: none. 47 Running head: Depression & birth mode mediate influence on sleep 49 50 51 52 AC C 48 EP 42 ACCEPTED MANUSCRIPT 53 Abstract 55 Rationale: Sleep duration is critical to growth, learning, and immune function development in infancy. 56 Strategies to ensure that national recommendations for sleep duration in infants are met require 57 knowledge of perinatal factors that affect infant sleep. 58 Objectives: To investigate the mechanistic pathways linking maternal education and infant sleep. 59 Methods: An observational study was conducted on 619 infants whose mothers were enrolled at the 60 Edmonton site of the CHILD birth cohort. Infant sleep duration at 3 months was assessed using the Brief 61 Infant Sleep Questionnaire. Maternal education was collected via maternal report. Prenatal and postnatal 62 depression scores were obtained from the 20-item Center for Epidemiologic Studies Depression Scale 63 (CES-D). Birth records and maternal report were the source of covariate measures. Mediation analysis 64 (PROCESS v3.0) was used to examine the indirect effects of maternal education on infant sleep duration 65 mediated through prenatal depression and birth mode. 66 Measurements and Main Results: At 3 months of age, infants slept on average 14.1 hours. Lower 67 maternal education and prenatal depression were associated with significantly shorter infant sleep 68 duration. Emergency cesarean section birth was associated with 1-hour shorter sleep duration at 3 months 69 compared to vaginal birth [without intrapartum antibiotic prophylaxis] (β: -0.99 hours; 95% CI: -1.51, - 70 0.48). Thirty percent of the effect of lower maternal education on infant total sleep duration was mediated 71 sequentially through prenatal depression and birth mode (Total Indirect Effects: -0.12, 95%CI: -0.22, - 72 0.03, p<0.05). 73 Conclusions: Prenatal depression and birth mode sequentially mediate the effect of maternal education on 74 infant sleep duration. 75 Keywords: pediatric sleep, maternal education, prenatal depression, birth mode, emergency caesarean 76 section 77 78 AC C EP TE D M AN US C RI PT 54 ACCEPTED MANUSCRIPT 79 80 81 82 RI PT 83 84 85 86 M AN US C 87 88 89 90 91 92 93 94 D 95 TE 96 97 EP 98 99 101 102 103 AC C 100 104 Abbreviations 105 CHILD - Canadian Healthy Infant Longitudinal Development; CS - Caesarean section; HPA – 106 Hypothalamic-pituitary-adrenal; IAP - Intrapartum antibiotic prophylaxis; SES – Socioeconomic status 107 ACCEPTED MANUSCRIPT 108 109 3.1 Introduction Globally, over 50% of 3 month old infants obtain less than the recommended 14 hours of sleep per 24 hour period (1,2). Sleep plays a central role in growth, neurological development, learning and 111 processing of memory, and optimal functioning of the immune system (3). In childhood, shorter sleep 112 duration is associated with cognitive deficits (4), poor school performance and increased behavioral 113 problems (5). Infants who sleep less are at risk for overweight (6) and those with frequent nocturnal 114 awakenings are more likely to develop asthma (7). Sleep problems which arise in infancy and childhood 115 tend to persist (8). In adults, short sleep duration has been linked to increased risk of mortality, diabetes 116 mellitus, hypertension, obesity and coronary heart disease (4). Poor quality sleep in adulthood is 117 identified as a pathway by which low socioeconomic status (SES) ‘gets under the skin’ to cause disease 118 (9) and conceivably, may be a missing link in the intergenerational transmission of SES inequalities in 119 health (10). Hence, infant sleep is a logical target for government and public health agencies. M AN US C 120 RI PT 110 Be it short sleep in infants or sleep problems in toddlers, there is emerging evidence of the influence of family SES, including maternal educational attainment, on sleep in young children (11,12). 122 When examining the various household factors that affect infant sleep, such as parental sleep (11,13,14), 123 the upstream factor which is most likely candidate to mediate the relationship between SES and infant 124 sleep duration is maternal prenatal depression (15). Mothers in distress have sleep problems during 125 pregnancy (16), which can be ‘transmitted’ to the fetus via the maternal suprachiasmatic nucleus or 126 melatonin levels (17–19). Maternal depression also leads to elevated free cortisol levels during pregnancy 127 (20), which in turn, appear to increase infant cortisol levels in response to stress (21,22). When cortisol 128 levels are elevated, they preferentially bind to norepinephrine and glucocorticoid receptors and ultimately 129 increase sleep EEG frequency, light sleep and frequent waking, and decrease short-wave sleep via 130 stimulation of corticotropin releasing hormone (CRH) (23). Interestingly, maternal psychological health 131 also plays a role in the birth process and birth outcomes (24–26). If birth is stressful and/or leads to 132 unexpected events such as cesarean delivery, infant sleep can be impacted through newborn exposure to 133 hypothalamic pituitary adrenal (HPA) axis hormones (27), reduced mother-infant bonding (28), or AC C EP TE D 121 ACCEPTED MANUSCRIPT 134 additional birth-associated medical interventions. There is a gap in the literature regarding the influence 135 of the birth process on infant sleep duration beyond the second postnatal day (29). 136 While maternal education status has been linked to infant sleep duration (11) and maternal psychological health (30), the relatedness of these factors has not been studied nor has the birth process 138 been taken into account. Reported associations between cesarean delivery and postpartum depression (31) 139 may in fact be secondary to existing prenatal depression. Importantly, potential causes of childhood sleep 140 duration are often examined after 1 year of age, excluding the first 6 months of life when critical 141 development of the circadian rhythm, neurological function, and behavior takes place (3,19,32). We 142 examined the association between maternal educational attainment and infant sleep duration at 3 months 143 of age in the Canadian healthy Infant Longitudinal Development (CHILD) birth cohort. Second, we 144 assessed whether prenatal depression and birth mode sequentially mediated the association between 145 maternal education status and infant sleep duration. The CHILD cohort also provided a unique 146 opportunity to test independence from putative confounding factors such as colic, often a suspected cause 147 of sleep problems in infants (33) and antibiotic exposure, which has been found to induce transient 148 insomnia (34) and decrease slow wave sleep (35). TE D M AN US C RI PT 137 149 3.2 Methods 151 Study Design 152 This study involved a subsample of 619 Canadian infants from the Edmonton site of the Canadian 153 Healthy Infant Longitudinal Development (CHILD) birth cohort (http://www.childstudy.ca) (36). The 154 Human Research Ethics board at the University of Alberta approved this study. Written informed consent 155 was obtained from the mother at enrollment. Mothers of studied infants were enrolled during pregnancy 156 between 2008 and 2012. Data on covariates were obtained from hospital records (birth mode, gestational 157 age at birth, birth weight and direct antibiotic exposure) or standardized questionnaires (maternal age, 158 maternal race/ethnicity, household income, infant sex, breastfeeding status, weight at 3 months, and solid 159 food intake before 3 months) (see Appendix). AC C EP 150 ACCEPTED MANUSCRIPT Infant sleep at 3 months of age (outcome of interest). Infant total sleep duration was obtained from the 161 parent self-reported BISQ (Appendix B) administered at 3 months of age (37). Parent self-report of infant 162 day (7 am until 7 pm) and night (7 pm until 7 am) sleep duration in hours and minutes were combined to 163 obtain infant total sleep duration per 24-hour period. 164 Maternal education (exposure of interest). Maternal education level was collected from a standardized 165 questionnaire. Mothers chose from: “1-high school or less”, “2-some university or college”, and “3- 166 university degree obtained”. Categories 1 and 2 were then combined. 167 Depression symptoms. Depression symptoms were measured using the 20-item Center for Epidemiologic 168 Studies Depression Scale (CES-D) (38) at 36 weeks of gestation and 6 months postpartum. Women self- 169 reported how often they experienced various depressive cognitions, affect, and behaviors during the past 170 week. Responses were given on a score ranging from 0 (None of the time; less than 1 day) to 3 (Most or 171 all of the time; 5-7 days). Responses were summed, with higher scores indicating higher depressive 172 symptoms (min=0, max=60). CES-D scores of 16 or greater represent significant risk for clinical 173 depression (39). 174 Statistical analysis. The Student’s t-test and ANOVA test with Tukey post-hoc test were used to examine 175 the association between maternal education level and covariates. Univariate and multivariate linear 176 regression modelling was performed with total sleep duration as the outcome and maternal education as 177 the exposure of interest. A final model was chosen using purposeful selection as described by Hosmer and 178 Lemeshow (40). Multinomial regression was used to predict having postnatal depression only, prenatal 179 depression only, or prenatal and postnatal depression together (ref: no prenatal or postnatal depression) 180 from maternal education level. Statistical significance of the difference in sleep duration according to 181 birth mode was determined by ANOVA with Tukey post hoc test. Mediation analysis was conducted 182 using the Hayes PROCESS v3.0 macro for SPSS, version 23.0 (SPSS Inc) (41). A multiple mediation 183 path model was evaluated to determine the indirect effects of sequential mediators: prenatal depression 184 (mediator 1) and birth mode (mediator 2) in the path between maternal education and infant total sleep 185 duration at 3 months of age. Bootstrapping (5000 bootstrap resamples) was used to generate to 95% CIs AC C EP TE D M AN US C RI PT 160 ACCEPTED MANUSCRIPT 186 in mediation models. Sensitivity analyses were conducted to explore the potential confounding effect of 187 postpartum depression on the multiple mediation model. 188 3.3 Results 190 191 Table 1. Associations between infant and maternal characteristics, maternal education level and total sleep duration at 3 months. RI PT 189 Maternal Education - AC C Total Sleep Duration (hours/24 hours) Mean (SD) M AN US C University Degree n = 335/619 (54.11%) Observations 14.17 (2.14) - N.A. 284 335 13.94 (2.20) 14.36 (2.08) 28 (9.96%) 138 (49.11%) 115 (40.93%) 44 (13.25%) 148 (44.58%) 140 (42.17%) 0.357 72 286 255 14.32 (2.25) 14.33 (2.27) 13.93 (1.93) 6 (2.12%) 147 (51.94%) 127 (44.88) 3 (1.06%) 14 (4.22%) 173 (52.11%) 138 (41.57%) 7 (2.11%) 0.332 20 320 265 10 14.09 (2.23) 14.19 (2.23) 14.14 (2.05) 13.95 (1.44) 6 (2.17%) 59 (21.38%) 185 (67.03%) 26 (9.42%) 13 (3.93%) 73 (22.05%) 203 (61.33%) 42 (12.69%) 0.302 19 132 388 68 14.26 (2.24) 14.34 (2.26) 14.16 (2.11) 13.91 (2.10) 150 (52.82%) 134 (47.18%) 161 (48.06%) 174 (51.94%) 0.259 311 308 14.25 (2.13) 14.09 (2.16) 149 (53.41%) 130 (46.59%) 161 (49.09%) 167 (50.91%) 0.291 310 297 13.94 (2.16) 14.37 (2.10) 144 (51.25%) 56 (19.93%) 36 (12.81%) 45 (16.01%) 179 (53.92%) 81 (24.40%) 35 (10.54%) 37 (11.14%) 0.177 323 137 71 82 14.39 (2.08) 13.98 (2.18) 14.36 (2.07) 13.40 (2.20) 138 (48.59%) 79 (27.82%) 67 (23.59%) 204 (61.08%) 95 (28.44%) 35 (10.48%) <0.001 342 174 102 14.26 (2.06) 14.09 (2.16) 14.00 (2.38) 8 (2.92%) 266 (97.08%) 7 (2.12%) 323 (97.88%) 0.604 15 589 14.83 (1.54) 14.16 (2.16) 43 (15.30%) 52 (15.66%) 0.911 95 13.89 (2.27) TE D p-value EP Maternal education, No. (%) No University Degree University Degree Infant Characteristics Gestational age, No. (%) Below 38 weeks 38 to 39 weeks Over 40 weeks Birth weight, No. (%) < 2500g 2500-3499g 3500-4499g > 4500g Weight at 3 months, No. (%) < 5000g 5000-5999g 6000-7999g >8000g Gender, No. (%) Boy Girl Antibiotic exposure, No. (%) Yes No Birth mode, No. (%) Vaginal –IAP Vaginal +IAP Scheduled CS Emergency CS Breastfeeding status, No. (%) Exclusive Partial Zero Solids, No. (%) Yes No Colic, No. (%) Yes No University Degree n = 284/619 (45.89 %) ACCEPTED MANUSCRIPT <0.001 185 409 25 14.07 (2.08) 14.20 (2.17) 14.32 (2.20) <0.001 38 138 92 308 13.86 (1.70) 14.01 (2.20) 14.04 (2.34) 14.41 (1.97) 42 13.59 (2.76) 17 (5.07%) RI PT 7 (2.09%) 55 (16.42%) 49 (14.63%) 207 (67.79%) 14.22 (2.13) 258 (77.01%) 50 (14.93%) 27 (8.06%) <0.001 473 68 76 14.09 (2.09) 14.64 (2.46) 14.21 (2.18) 41 (13.62%) 260 (86.38%) <0.001 110 444 13.65 (2.11) 14.23 (2.14) 30 (10.71%) 250 (89.29%) 0.028 70 437 13.43 (2.54) 14.26 (2.03) 332 (99.40%) 2 (0.60%) <0.001 588 27 14.18 (2.14) 13.87 (1.94) 152 (45.65%) 181 (54.35%) 0.685 270 340 13.70 (2.06) 14.54 (2.13) Notes: -IAP: no intrapartum antibiotics; +IAP: with intrapartum antibiotics; CS: caesarean section. TE 193 194 60 (17.91%) 259 (77.31%) 16 (4.78%) 518 D 192 280 (84.34%) M AN US C No 238 (84.70%) Maternal Characteristics Maternal age, No. (%) 18 to 29 125 (44.01%) 30 to 39 150 (52.82%) Over 40 9 (3.17%) Annual household income, No. (%) Less than 39,999 31 (10.95%) 40,000-79,999 83 (29.33%) 80,000-99,9999 43 (15.19%) Greater than 100,000 101 (35.69%) Preferred not to answer 25 (8.83%) Maternal race, No. (%) White 215 (76.24%) Asian 18 (6.38%) Other 49 (17.38%) Prenatal depression, No. (%) Yes 69 (27.27%) No 184 (72.73%) Postnatal depression, No. (%) Yes 40 (17.62%) No 187 (82.38%) Maternal prenatal smoking, No. (%) No 256 (91.10%) Yes 25 (8.90%) Siblings in the home, No. (%) No 124 (43.82%) Yes 159 (56.18%) In our population-based cohort of 619 mother-infant dyads, 54% of mothers had a university degree. Significant differences in maternal age, annual household income, maternal race, breastfeeding 196 status, prenatal smoking, and pre- and post- natal depression were found between mothers with and 197 without a university degree (See Table 1). Women with a university degree were older than those without 198 a university degree. The majority of mothers with a university degree (67.79%) had an annual household 199 income greater than $100,000. Annual household income greater than $100,000 was less prevalent when 200 mothers did not have a university degree (35.69%). Two percent of mothers with a university degree and 201 10.95% of mothers without a university degree had a household income below $39,999. Mothers with a 202 university degree were less likely to identify as a race other than white or Asian (8.1% vs 17.4%). In 203 contrast, 6.4%of mothers without a university degree were Asian, compared to 14.9% of mothers with a AC C EP 195 ACCEPTED MANUSCRIPT university degree. Exclusive breastfeeding was higher at 3 months of age in mothers with a university 205 degree (61.1% vs 48.6%). 207 M AN US C RI PT 204 208 Figure 1. Infant total sleep duration at 3 months of age according to birth mode. 209 Note:-IAP: no intrapartum antibiotics; +IAP: with intrapartum antibiotics; CS: caesarean section. TE D 206 211 EP 210 Mean infant total sleep duration at 3 months of age was 14.2 hours (Standard deviation [SD]: 2.14; Table 1). The National Sleep Foundation recommends 14-17 hours of sleep in this age group (2). 213 Infants born to mothers without a university degree slept an average of 13.94 hours (SD: 2.20) compared 214 to 14.36 hours (SD: 2.08) in infants born to mothers with university degrees. Both prenatal and postnatal 215 depression were associated with significantly shorter infant sleep duration. Total sleep duration was 216 significantly different according to birth mode. Tukey post-hoc test showed that infants born by 217 emergency caesarian section (CS) slept significantly shorter than infants born vaginally without IAP or by 218 scheduled CS (Figure 1). 219 AC C 212 ACCEPTED MANUSCRIPT 220 221 Maternal education is associated with prenatal and postnatal depression. Twenty-seven percent of mothers without a university degree had prenatal depression (CES-D score ≥ 16), while 13.6% of mothers with a university degree had prenatal depression. Similarly, 17.6% 223 of mothers without a university degree had postnatal depression while only 10.7% of mothers with a 224 university degree had postnatal depression. Women without a university degree had an almost 2 times 225 higher relative risk of prenatal depression without postnatal depression (relative risk [RRR]: 1.91, 95% 226 CI: 1.06, 3.43, p=0.03; figure E1 online supplement), 4.4 times higher relative risk of both prenatal and 227 postnatal depression (RRR: 4.39, 95% CI: 1.82, 10.62, p<0.001), but no difference in relative risk of 228 postnatal depression without prenatal depression (RRR: 0.95, 95% CI: 0.43, 2.08, p=0.89) compared to 229 women with a university degree. M AN US C RI PT 222 230 231 232 Maternal postsecondary education is positively associated with infant sleep duration. Infants of mothers without a university degree had reduced sleep duration at 3 months of age (β: 0.42 hours, 95% Confidence Interval [CI]: -0.76, -0.08, p<0.01; Table 2) compared to mothers with a 234 university degree. This association remained significant following adjustment for infant factors including 235 gestational age at birth, gender, birth mode, breastfeeding status, solids, and colic (β: -0.42 hours, 95% 236 CI: -0.78, -0.07, p<0.05; Model 2). However, the difference in sleep duration by maternal education 237 status was not significant (β: -0.28 hours, 95% CI: -0.67, 0.11, p=NS; Model 3) when controlling for 238 maternal characteristics including prenatal depression, maternal age, maternal race, siblings in the home, 239 maternal prenatal smoking and all Model 2 variables. TE EP AC C 240 D 233 Table 2. Crude and multivariate linear regression analyses predicting infant total sleep duration at 3 months of age. Crude Model 1 Model 2 Model 3 Model 4 95% CI 95% CI 95% CI 95% CI 95% CI Β Β Β Β Β Coeff. Lower, Upper Coeff. Lower, Upper Coeff. Lower, Upper Coeff. Lower, Upper Coeff. Lower, Upper Maternal education (ref = university degree) Infant Characteristics -0.42* -0.76, -0.08 -0.44** -0.78, -0.10 -0.42* -0.78, -0.07 -0.37 -0.75, 0.02 -0.29 -0.65, 0.06 ACCEPTED MANUSCRIPT Block 1 -0.12† -0.24, 0.01 -0.13† -0.26, 0.001 -0.15* -0.28, -0.02 -0.03 -0.18, 0.12 - - 0.16 -0.18, 0.50 0.12 -0.22, 0.46 0.09 -0.25, 0.44 0.16 -0.20, 0.52 - - Vag +IAP Scheduled CS Emergency CS -0.41† -0.02 -0.99*** -0.83, -0.02 -0.57, 0.52 -1.51, -0.48 -0.49* -0.11 -0.94*** -0.92, -0.07 -0.67, 0.45 -1.45, -0.42 -0.50* -0.09 -0.98*** -0.93, -0.06 -0.66, 0.48 -1.51, -0.44 -0.48* -0.22 -0.68* -0.93, -0.02 -0.81, 0.37 -1.26, -0.10 Partial Zero -0.17 -0.56, 0.22 - - -0.20 -0.60, 0.21 -0.27 -0.74, 0.21 - - -0.24 -0.74, 0.26 Solids 0.68 -0.43, 1.78 - - 0.89 -0.23, 2.00 Colic -0.33 -0.81, 0.14 - - -0.40 -0.88, 0.08 -0.03** -0.06, -0.01 - - - - -0.14 0.12 -0.51, 0.24 -0.75, 0.99 - - - - 0.55* 0.12 0.01, 1.10 -0.40, 0.64 - - - -0.85**** 0.51, 1.18 - - -1.14, 0.51 - - Gestational age at birth (continuous) Gender (ref = female) Birth mode (ref = vaginal - IAP) -0.35 -0.24 -0.70* -0.80, 0.09 -0.81, 0.33 -1.25, -0.15 Maternal Characteristics Prenatal depression (CES-D score) Maternal age 18-29 (ref = 30-39) 40+ Maternal race (ref = white) Asian Other Siblings in the home (ref = no) Maternal prenatal smoking -0.31 241 - - -0.83, 0.23 - - 0.96 -0.59, 0.43 - - -0.08 -0.59, 0.43 - - -0.03* -0.06, 0.00 -0.03* -0.05, -0.004 -0.36† -0.46 -0.78, 0.05 -1.41, 0.49 - - - 0.66* 0.29 0.06, 1.25 -0.28, 0.87 - - - - 0.79*** 0.41, 1.17 0.79*** 0.42, 1.16 - - 0.29 -0.66, 1.24 - - Notes: Model 1: maternal education, gestational age at birth, gender, and birth mode. Model 2: Model 1 with breastfeeding status, solids, and colic. Model 3: Model 2 with prenatal depression, maternal age, and maternal race. Model 4: maternal education, birth mode, prenatal depression, and siblings in the home [chosen by purposeful selection]. IAP: intrapartum antibiotics; CS: caesarean section. p<0.05*; p<0.01**; p<0.001***; p<0.0001****. D 248 -0.77, 0.07 -0.30 Emergency CS was associated with shorter sleep duration at 3 months of age compared to the TE 242 243 244 245 246 247 -0.35† M AN US C Breastfeeding status (ref = exclusive) RI PT Block 2 reference group of infants born vaginally without IAP (Crude β: -0.99 hours, 95% CI: -1.51, -0.48, 250 p<0.001). Each 1-point increase in mothers’ prenatal CES-D score was associated with a 0.03-hour 251 decrease in infant sleep duration (Crude β: -0.03 hours, 95% CI: -0.06, -0.01, p<0.01). Infants of Asian 252 mothers slept on average 0.59 hours more than infants of white mothers (Crude β: 0.59 hours, 95% CI: 253 0.05, 1.13, p<0.05). AC C 254 EP 249 Purposeful selection resulted in the inclusion of birth mode, prenatal depression (CES-D score) 255 and siblings in the home in the regression model predicting infant sleep duration (Model 4). Maternal 256 education was included as an exposure of interest. Emergency CS (β: -0.70 hours, 95% CI: -1.25, -0.15, 257 p<0.05; Model 4), prenatal depression (CES-D score) (β: -0.03 hours, 95% CI: -0.05, -0.004, p<0.05) and 258 siblings in the home (β: 0.79 hours, 95% CI: 0.42, 1.16, p<0.001), significantly contributed to the ACCEPTED MANUSCRIPT prediction of infant sleep duration. Maternal education did not contribute to the model predicting infant 260 sleep duration (β: -0.29 hours, 95% CI: -0.65, 0.06, p=NS) when adjusting for birth mode, prenatal 261 depression and siblings in the home. Interactions between maternal education, prenatal depression, birth 262 mode and siblings in the home did not significantly contribute to the model. RI PT 259 263 Prenatal depression and birth mode sequentially mediate the relationship between maternal 265 education level and infant sleep duration. 266 M AN US C 264 Regression analysis was used to investigate the hypothesis that prenatal depression and birth mode sequentially mediate the effect of maternal education on infant total sleep duration (Figure 2). 268 Lower maternal education was a significant predictor of prenatal depression (CES-D score) (β=2.68, 269 SE=0.63, p<0.0001; Online Supplement Table E1 ). Furthermore, prenatal depression (CES-D score) 270 (β=0.01, SE=0.01, p=0.05), but not lower maternal education (β=0.15, SE=0.09, p=0.11) was a significant 271 predictor of birth mode (classified as 1 = vaginal no IAP, 2 = vaginal IAP, 3 = scheduled CS, and 4 = 272 emergency CS) when modelled concurrently. When evaluated in regression analysis together, prenatal 273 depression (CES-D score) (β=-0.03, SE=0.01, p=0.04) and birth mode (β=-0.27, SE=0.08, p<0.01) but not 274 maternal education (β=-0.27, SE=0.18, p=0.15) predicted infant total sleep duration. AC C EP TE D 267 M AN US C RI PT ACCEPTED MANUSCRIPT 275 276 Figure 2. Sequential mediation model of associations between maternal education, prenatal depression, 277 birth mode, and infant sleep duration. -IAP: no intrapartum antibiotics; +IAP: with intrapartum 278 antibiotics; CS: caesarean section. p<0.1†; p<0.05*; p<0.01**; p<0.001***; p<0.0001****. D 279 Prenatal depression and birth mode sequentially mediate the relationship between maternal 281 education and infant sleep duration. Following sequential mediation, the direct association of lower 282 maternal education with infant total sleep duration (path c`) was no longer significant (Effect: -0.27, 95% 283 CI: -0.63, 0.09, p=0.15; Table 3); however, the total indirect effects of lower maternal education on infant 284 total sleep duration mediated sequentially through prenatal depression and birth mode were significant 285 (Effect: -0.12, 95% CI: -0.22, -0.03, p<0.05). Combined, the direct and indirect effects of lower maternal 286 education on infant sleep duration were significant (Effect: -0.38, 95% CI: -0.74, -0.03, p<0.05). Eighteen 287 percent of the effect of lower maternal education on infant total sleep duration was mediated through 288 prenatal depression alone (Effect: -0.07, 95% CI: -0.15, -0.01, p<0.05). The indirect effect of maternal 289 education through birth mode alone was not significant (Effect: -0.04, 95% CI: -0.11, 0.01, p=NS). The 290 effect of lower maternal education on infant total sleep duration (2.3% of the total effect) was mediated AC C EP TE 280 ACCEPTED MANUSCRIPT sequentially through prenatal depression and birth mode directly (Effect: -0.01, 95% CI: -0.02, -0.0004, 292 p<0.05). These associations were robust to sensitivity analyses for imputed missing values. Due to the 293 high correlation between prenatal and postnatal CES-D scores, sequential mediation of the relationship 294 between maternal education and infant sleep duration through prenatal and subsequently postnatal 295 depression, as well as postnatal depression alone, was explored. Postnatal CES-D score was not found to 296 mediate the relationship between maternal education and infant sleep duration with prenatal CES-D score 297 or on its own. M AN US C RI PT 291 298 299 300 Table 3. Breakdown of direct and indirect effects of maternal education on infant sleep duration at 3 months of age through prenatal depression (CES-D score) and birth mode. A) Total effect of maternal education (indirect + direct effects) B) Indirect effect 1 Maternal education prenatal depression C) Indirect effect 2 Maternal education birth mode EP Direct effect of maternal education 303 304 birth mode Effect SE p 95% CI 100% -0.38* 0.18 0.04 -0.74, -0.03 17.7% -0.07* 0.03 <0.05 -0.15, -0.01 10.5% -0.04 0.03 NS -0.11, 0.01 2.3% -0.01* 0.006 <0.05 -0.02, 0.0004 30.4% -0.12* 0.05 <0.05 -0.22, -0.03 69.5% -0.27 0.18 NS -0.63, 0.09 AC C 302 Notes: p <0.05*. D prenatal depression Total indirect effects (1 + 2 + 3) 301 sleep duration TE D) Indirect effect 3 Maternal education sleep duration sleep duration % Effect Explained 3.4 Discussion In our general population cohort of infants from an urban center in Canada, 38% of infants slept 305 less than the recommended 14 hours per day; lower than global estimates of infant short sleep at 3 months 306 of age (1). Infants born to mothers with a university degree slept an average of 0.42 hours longer than 307 infants of mothers without a university degree. The association between maternal level of education and ACCEPTED MANUSCRIPT infant sleep duration persisted following adjustment for infant factors but diminished with additional 309 adjustment for maternal characteristics, notably maternal prenatal depression (14). Further, we found that 310 birth mode independently predicted infant sleep duration, with infants delivered by emergency cesarean 311 sleeping approximately one hour less than infants born by vaginal birth. When combined, we found that 312 maternal prenatal depression status and birth mode jointly mediated the association between maternal 313 level of education and infant sleep duration. Previously, prenatal depression was found to be associated 314 with shorter sleep duration in 1-2 year olds independent of household SES, and postnatal depressive 315 symptoms in caregivers reported to influence the relationship between family demographics and sleep 316 problems in toddlers (11,42). Our study is the first to suggest that prenatal depression has the capacity to 317 mediate the relationship between maternal education level and infant sleep in the 3 months of age. Almost 318 one-third of the indirect effect of maternal education was mediated through the joint action of prenatal 319 depression and emergency cesarean. M AN US C 320 RI PT 308 The additional novelty of our study is the reduction in infant sleep three months after emergency cesarean delivery; this was not observed with scheduled cesarean or in vaginal deliveries with maternal 322 antibiotic prophylaxis. Compared to vaginal delivery, both emergency and scheduled cesarean delivery 323 have been shown to reduce active sleep in newborns on the first but not second postnatal day; however, 324 an observed lack of diurnal rhythms in infant sleep/wakefulness with both surgical groups seems to 325 persist (29). Netsi et al did not find an association between birth mode and sleep duration at age 3 months 326 in a Brazilian cohort, in which many of the cesarean births would have been scheduled (43). One aspect 327 of modern birth, the induction and augmentation of labour using synthetic oxytocin, is very common in 328 birth by emergency cesarean (44). In animal studies, synthetic oxytocin increases wakefulness (45), 329 hypothesized to occur due to oxytocin’s influence on the HPA axis through an excitatory action on CRH 330 (23,46,47). Upregulation of the CRH system has been implicated in the impairment of sleep quality in 331 both human and animal studies (46). AC C EP TE D 321 332 Furthermore, emergency cesarean co-mediated with prenatal depression, the association between 333 maternal SES and infant sleep. Little is known about the maternal physiological impact of emergency CS ACCEPTED MANUSCRIPT on the infant (48). Unexpected cesarean delivery can be a traumatic birth experience for the mother (49), 335 interfering with parenting behaviours that promote self-soothing in the infant and longer sleep duration 336 (50). Interestingly, both maternal depression during pregnancy and emergency cesarean birth have the 337 capacity to disrupt development of the infant HPA axis and alter regulation of circadian rhythm (27,32). 338 Smith et al. found that infants born by emergency but not scheduled cesarean, had higher levels of free 339 cortisol in umbilical cord blood samples than vaginally born infants (27). Elevated cortisol levels can 340 increase CRH, which are associated with reduced sleep quality (23). However, due to the development of 341 the circadian clock genes and HPA axis during late gestation and early infancy, elevated cortisol at birth 342 may have a lasting effect on the programming of these systems (32). Furthermore, infants born by 343 emergency but not scheduled cesarean, have been found to have elevated C-reactive protein in the cord 344 blood following birth (51). The administration of pro-inflammatory cytokines in animal studies promotes 345 non-REM sleep, which is more common after sleep deprivation (52). Lastly, infants born by emergency 346 cesarean are also more likely to exhibit gut microbial dysbiosis than infants born vaginally or by 347 scheduled CS (53), compositional changes that may ultimately alter circadian rhythm and sleep patterns 348 (54). TE D M AN US C RI PT 334 Our results also support the thesis that prenatal depression influences infant sleep through a fetal 350 programming pathway (15). Infants born to mothers with prenatal depression slept on average 0.56 hours 351 shorter than infants born to mothers without prenatal depression. The prenatal stress model, which is an 352 approximate animal model of stress and depression in pregnancy, results in prolonged corticosterone 353 production after acute stress and reduced expression of glucocorticoids in the hippocampus in adult 354 offspring (32). As a result, infants of mothers with prenatal depression may have an exaggerated stress- 355 response which negatively impacts their sleep duration after birth. Prenatal depression is strongly linked 356 to low SES (55); stressful life events during pregnancy and concern over finances have both been 357 associated with frequent nocturnal awakening in toddlers (7). In our study, women without a university 358 degree were much more likely to experience prenatal depression with or without postnatal depression but 359 not postnatal depression without prenatal depression. AC C EP 349 ACCEPTED MANUSCRIPT 360 Strengths and Limitations 362 Our study has several strengths, including the ability to investigate birth mode in greater detail than 363 previously examined in a birth cohort with a representative and large sample size. Also, the universal 364 healthcare context of the Canadian populace provides an opportunity to study SES independent of 365 accessibility to prenatal care healthcare (56). Limitations of this study include the unavailability of 366 measures on maternal prenatal sleep, parenting behaviour and depressive symptoms in the postpartum 367 period prior to 6 months. M AN US C RI PT 361 368 369 370 3.5 Conclusions Socioeconomic factors in early life have a strong influence on virtually all aspects of early human development (57). In our general population cohort from the CHILD study, infant sleep duration at 3 372 months of age was predicted by maternal education level, prenatal depression and birth mode. The 373 maternal educational association with infant sleep was sequentially mediated by prenatal depression and 374 birth mode. Our study provides evidence for a prenatal-birth pathway by which parental SES can impact 375 infant sleep. Mothers who experience prenatal depression or emergency cesarean birth may benefit from 376 advice on parenting style and infant stimulus control to increase infant sleep duration (50), so that these 377 problems do not persist in childhood. While we are at an early stage to discern the underlying biologic 378 mechanisms, this study identifies prenatal depression and birth mode as targets for policy makers to 379 improve infant sleep duration. Future work is required to determine if the impact of these exposures is 380 mediated by oxytocin administration, cortisol level, maternal sleep, postpartum depression or parental 381 behaviours. AC C EP TE D 371 382 383 384 3.6 Acknowledgements ACCEPTED MANUSCRIPT The authors would like to thank the CHILD Study participants, without whom this research study would 386 not be possible. The authors also thank the CHILD team, which includes interviewers, computer and 387 laboratory technicians, clerical workers, research scientists, volunteers, managers, receptionists and 388 nurses. 389 3.7 Appendix 390 Covariate Measures 391 Birth mode. Birth mode was collected from maternal hospital records. Birth mode was categorized a 392 vaginal birth without intrapartum antibiotic prophylaxis (IAP), vaginal birth with IAP, scheduled 393 caesarean section (CS) and emergency CS. 394 Colic. Infant colic status was determined from parent-reported infant medication questionnaires. Infants 395 were classified as having colic before 3 months of age if they were taking a medication indicated for the 396 treatment of colic or if colic was listed as the reason for taking a medication. 397 Breastfeeding status. Infant feeding status was collected from parental report at 3 months of age. 398 Breastfeeding status was categorized as exclusive (breast milk only following hospital discharge from 0-3 399 months), partial (both breastmilk and formula consumed from 0-3 months), and formula (formula only 400 from 0-3 months). 401 Household income. Household income was collected from maternal report at 18-36 weeks gestational age 402 and was categorized as: 1) less than or equal to 39,999; 2) 40,000 to 79,999; 3) 80,000 to 99,999; 4) 403 100,000 or greater; 5) prefer not to answer. 404 Maternal race/ethnicity. Maternal race/ethnicity was collected from maternal report at 18-36 weeks 405 gestational age. For the purposes of this study, maternal race was categorized as Caucasian, Asian (East 406 Asian, South Asian and South East Asian) or other (Black, Hispanic, Middle Eastern and First Nations). AC C EP TE D M AN US C RI PT 385 407 408 CHILD Study Investigators 409 Subbarao P (Director), The Hospital for Sick Children & University of Toronto; Turvey SE (co‐ 410 Director), University of British Columbia; Anand SS, McMaster University; Azad MB, University of ACCEPTED MANUSCRIPT Manitoba; Becker AB, University of Manitoba; Befus AD, University of Alberta; Brauer M, University of 412 British Columbia; Brook JR, University of Toronto; Chen E, Northwestern University, Chicago; Cyr 413 MM, McMaster University; Daley D, University of British Columbia; Dell SD, The Hospital for Sick 414 Children & University of Toronto; Denburg JA, McMaster University; Duan QL, Queen's University; 415 Eiwegger T, The Hospital for Sick Children & University of Toronto; Grasemann H, The Hospital for 416 Sick Children & University of Toronto; HayGlass K, University of Manitoba; Hegele RG, The Hospital 417 for Sick Children & University of Toronto; Holness DL, University of Toronto; Hystad P, Oregon State 418 University; Kobor M, University of British Columbia; Kollmann TR, University of British Columbia; 419 Kozyrskyj AL, University of Alberta; Laprise C, Université du Québec à Chicoutimi; Lou WYW, 420 University of Toronto; Macri J, McMaster University; Mandhane PJ, University of Alberta; Miller G, 421 Northwestern University, Chicago; Moraes TJ, The Hospital for Sick Children & University of Toronto; 422 Paré P, University of British Columbia; Ramsey C, University of Manitoba; Ratjen F, The Hospital for 423 Sick Children & University of Toronto; Sandford A, University of British Columbia; Scott JA, University 424 of Toronto; Scott J, University of Toronto; Sears MR, (Founding Director), McMaster University; 425 Silverman F, University of Toronto; Simons E, University of Manitoba; Takaro T, Simon Fraser 426 University; Tebbutt SJ, University of British Columbia; To T, The Hospital for Sick Children & 427 University of Toronto. References 430 1. 431 432 AC C 429 EP 428 TE D M AN US C RI PT 411 Galland BC, Taylor BJ, Elder DE, Herbison P. Normal sleep patterns in infants and children : A systematic review of observational studies. Sleep Med Rev. Elsevier Ltd; 2016;16(3):213–22. 2. Hirshkowitz M, Whiton K, Albert SM, Alessi C, Bruni O, DonCarlos L, et al. National sleep 433 foundation’s sleep time duration recommendations: Methodology and results summary. Sleep 434 Heal. National Sleep Foundation; 2015;1:40–3. 435 436 3. Davis KF, Parker KP, Montgomery GL. Sleep in Infants and Young Children: Part One: Normal Sleep. J Pediatr Heal care. 2004;18(2):65–71. ACCEPTED MANUSCRIPT 437 4. 438 439 Smithson L, Baird T, Tamana S, Lau A, Mariasine J, Chikuma J, et al. Shorter sleep duration is associated with reduced cognitive development at 2 years of age. Sleep Med. 2018;0(0). 5. Astill RG, Van der Heijden KB, Van Ijzendoorn MH, Van Someren EJW. Sleep, cognition, and behavioral problems in school-age children: A century of research meta-analyzed. Psychol Bull. 441 2012;138(6):1109–38. 442 6. RI PT 440 Halal CSE, Matijasevich A, Howe LD, Santos IS, Barros FC, Nunes ML. Short Sleep Duration in the First Years of Life and Obesity/Overweight at Age 4 Years: A Birth Cohort Study. J Pediatr. 444 2016;168:99–103. 7. 446 early life is associated with nonatopic asthma in children. Eur Respir J. 2009;34:1288–95. 8. 448 Sleep Problems in the First 3 Years of Life. Pediatrics. 2012;129(2):e276–84. 9. 450 sleep. Psychosom Med. 2002;64(2):337–44. 10. 452 453 11. 12. 13. 462 Gellis LA. Children’s Sleep in the Context of Socioeconomic Status, Race, and Ethnicity. In: ElSheikh M, editor. Sleep and Development: Familial and Socio- Cultural Considerations. 2011. 14. 460 461 Barazzetta M, Ghislandi S. Family Income and Material Deprivation : Do They Matter for Sleep Quality and Quantity in Early Life ? Evidence From a Longitudinal Study. Sleep. 2017;40(3). 458 459 Nevarez MD, Rifas-shiman SL, Kleinman KP, Gillman MW, Taveras EM. Associations of Early Life Risk Factors With Infant Sleep Duration. Acad Pediatr. Elsevier Ltd; 2010;10:187–93. 456 457 Cheng TL, Johnson SB, Goodman E. Breaking the Intergenerational Cycle of Disadvantage: The Three Generation Approach. Pediatrics. 2016;137(6):e20152467. 454 455 D 451 Moore PJ, Adler NE, Williams DR, Jackson JS. Socioeconomic status and health: The role of TE 449 Byars KC, Yolton K, Rausch J, Lanphear B, Beebe DW. Prevalence, Patterns, and Persistence of EP 447 Kozyrskyj AL, Kendall GE, Zubrick SR, Newnham JP, Sly PD. Frequent nocturnal awakening in AC C 445 M AN US C 443 Seifer R. Parental Psychopathology and Children’s Sleep. In: El-Sheikh M, editor. Sleep and Development: Familial and Socio- Cultural Considerations. 2011. 15. Connor TGO, Caprariello P, Gregory AM, Fleming P, the ALSPAC Study Team. Prenatal Mood Disturbance Predicts Sleep Problems in Infancy and Toddlerhood. Early Hum Dev. ACCEPTED MANUSCRIPT 463 16. 465 466 Okun ML, Tolge M, Hall M. Low socioeconomic status negatively affects sleep in pregnant women. J Obstet Gynecol Neonatal Nurs. 2014;43(2):160–7. 17. Torres-Farfan C, Rocco V, Monsó C, Valenzuela FJ, Campino C, Germain A, et al. Maternal RI PT 464 2007;83(7):451–8. 467 melatonin effects on clock gene expression in a nonhuman primate fetus. Endocrinology. 468 2006;147(10):4618–26. 18. Mendez N, Abarzua-Catalan L, Vilches N, Galdames HA, Spichiger C, Richter HG, et al. Timed M AN US C 469 470 maternal melatonin treatment reverses circadian disruption of the fetal adrenal clock imposed by 471 exposure to constant light. PLoS One. 2012;7(8):e42713. 472 19. 473 474 Med Hypotheses. 1997;49:229–34. 20. 475 Shariat M, Abedinia N, Noorbala AA, Raznahan M. The Relationship between Sleep Quality, Depression, and Anxiety in Pregnant Women: A Cohort Study. J Sleep Sci. 2017;2(1):20–7. 21. Davis EP, Glynn LM, Schetter CD, Hobel C, Chicz-demet A, Sandman CA. Prenatal Exposure to D 476 Recio J, Míguez JM, Buxton OM, Challet E. Synchronizing circadian rhythms in early infancy. Maternal Depression and Cortisol Influences Infant Temperament. J Am Acad Child Adolesc 478 Psychiatry. The American Academy of Child and Adolescent Psychiatry; 2007;46(6):737–46. 22. Simons SSH, Beijers R, Cillessen AHN, Weerth C De. Psychoneuroendocrinology Development EP 479 TE 477 of the cortisol circadian rhythm in the light of stress early in life. Psychoneuroendocrinology. 481 2016;62(2015):292–300. 482 23. AC C 480 Buckley TM, Schatzberg AF. Review: On the interactions of the hypothalamic-pituitary-adrenal 483 (HPA) axis and sleep: Normal HPA axis activity and circadian rhythm, exemplary sleep disorders. 484 J Clin Endocrinol Metab. 2005;90(5):3106–14. 485 24. Szegda K, Markenson G, Bertone-Johnson ER, Chasan-Taber L. Depression during pregnancy: a 486 risk factor for adverse neonatal outcomes? A critical review of the literature. J Matern Neonatal 487 Med. 2014;27(9):960–7. 488 25. Ding XX, Wu Y Le, Xu SJ, Zhu RP, Jia XM, Zhang SF, et al. Maternal anxiety during pregnancy ACCEPTED MANUSCRIPT 489 and adverse birth outcomes: A systematic review and meta-analysis of prospective cohort studies. 490 J Affect Disord. 2014;159(81):103–10. 491 26. Accortt EE, Cheadle ACD, Dunkel Schetter C. Prenatal Depression and Adverse Birth Outcomes: An Updated Systematic Review. Vol. 19, Maternal and Child Health Journal. Springer US; 2015. 493 1306-1337 p. 494 27. RI PT 492 Smith AK, Newport DJ, Ashe MP, Brennan PA, Laprairie JL, Calamaras M, et al. Predictors of neonatal hypothalamic-pituitary-adrenal axis activity at delivery. Clin Endocrinol (Oxf). 496 2011;75(1):90–5. 497 28. M AN US C 495 Hobbs AJ, Mannion CA, McDonald SW, Brockway M, Tough SC. The impact of caesarean 498 section on breastfeeding initiation, duration and difficulties in the first four months postpartum. 499 BMC Pregnancy Childbirth. BMC Pregnancy and Childbirth; 2016;16(1):1–9. 500 29. 501 delivery mode. Dev Psychobiol. 1998;32:293–303. 30. Hein A, Rauh C, Engel A, Häberle L, Dammer U, Voigt F, et al. Socioeconomic status and D 502 Freudigman KA, Thoman EB. Infants’ earliest sleep/wake organization differs as a function of depression during and after pregnancy in the Franconian Maternal Health Evaluation Studies 504 (FRAMES). Arch Gynecol Obstet. 2014;289(4):755–63. 506 507 the evidence examining the link. Psychosom Med. 2006;68(2):321–30. 32. 508 509 Carter FA, Frampton CMA, Mulder RT. Cesarean section and postpartum depression: A review of EP 31. Astiz M, Oster H. Perinatal Programming of Circadian Clock-Stress Crosstalk. Neural Plast. AC C 505 TE 503 2018;2018. 33. Sung V, Collet S, de Gooyer T, Hiscock H, Tang M, Wake M. Probiotics to Prevent or Treat 510 Excessive Infant Crying Systematic Review and Meta-analysis. JAMA Pediatr. 511 2016;167(12):1150–7. 512 34. Levaquin [package insert]. Titusville, NJ: Janssen Pharmaceuticals, Inc.; 2012. 513 35. Nonaka K, Nakazawa Y, Kotorii T. Effects of antibiotics, minocycline and ampicillin, on human 514 sleep. Brain Res. 1983;288(1–2):253–9. ACCEPTED MANUSCRIPT 515 36. Subbarao P, Anand SS, Becker AB, Befus AD, Brauer M, Brook JR, et al. The Canadian Healthy 516 Infant Longitudinal Development (CHILD) Study: examining developmental origins of allergy 517 and asthma: Table 1. Thorax. 2015;70(10):998–1000. 519 520 Internet sample. Pediatrics. 2004;113(6):e570-7. 38. 521 522 Sadeh A. A brief screening questionnaire for infant sleep problems: validation and findings for an Radloff LS. A Self-Report Depression Scale for Research in the General Population. Appl Psychol Measement. 1977;1(3):385–401. 39. RI PT 37. M AN US C 518 Lewinsohn PM, Seeley JR, Roberts RE, Allen NB. Center for Epidemiologic Studies Depression 523 Scale (CES-D) as a screening instrument for depression among community-residing older adults. 524 1997;12(2):277–87. 525 40. Hosmer D, Lemeshow S. Applied Logistic Regression. New York: Wiley; 2000. 526 41. Hayes AF, Little TD. Introduction to mediation, moderation, and conditional process analysis: a 527 42. Reid GJ, Hong RY, Wade TJ. The relation between common sleep problems and emotional and D 528 regression-based approach. Second Edi. New York: The Guilford Press; 2018. behavioral problems among 2- and 3-year-olds in the context of known risk factors for 530 psychopathology: Sleep in toddlers. J Sleep Res. 2009;18(1):49–59. 43. Netsi E, Santos IS, Stein A, Barros FC, Barros AJD, Matijasevich A. A different rhythm of life: EP 531 TE 529 sleep patterns in the first 4 years of life and associated sociodemographic characteristics in a large 533 Brazilian birth cohort. Sleep Med. Elsevier Ltd; 2017;37:77–87. 534 44. AC C 532 Kaul B, Vallejo MC, Ramanathan S, Mandell G, Phelps AL, Daftary AR. Induction of labor with 535 oxytocin increases cesarean section rate as compared with oxytocin for augmentation of 536 spontaneous labor in nulliparous parturients controlled for lumbar epidural analgesia. J Clin 537 Anesth. 2004;16(6):411–4. 538 45. 539 540 Lancel M, Kro S, Neumann ID. Intracerebral oxytocin modulates sleep – wake behaviour in male rats. Regul Pept. 2003;114:145–52. 46. Opp MR. Corticotropin-releasing hormone involvement in stressor-induced alterations in sleep ACCEPTED MANUSCRIPT 541 542 and in the regulation of waking. Adv Neuroimmunol. 1995;5(2):127–43. 47. Neumann ID, Wigger A, Torner L, Holsboer F, Landgraf R. Brain oxytocin inhibits basal and stress-induced activity of the hypothalamo-pituitary-adrenal axis in male and female rats: Partial 544 action within the paraventricular nucleus. J Neuroendocrinol. 2000;12(3):235–43. 48. 546 547 period: Can mode of delivery impact on the future health of the neonate? J Physiol. 2018;0:1–14. 49. 548 549 Furuta M, Sandall J, Cooper D, Bick D. Predictors of birth-related post-traumatic stress symptoms: secondary analysis of a cohort study. Arch Womens Ment Health. 2016;19(6):987–99. 50. 550 551 Tribe RM, Taylor PD, Kelly NM, Rees D, Sandall J, Kennedy HP. Parturition and the perinatal M AN US C 545 RI PT 543 Lau AA. Sleep and sleep disordered breathing in the first year of life: a Canadian birth cohort study. University of Alberta; 2014. 51. Logan CA, Thiel L, Bornemann R, Koenig W, Reister F, Brenner H, et al. Delivery mode, 552 duration of labor, and cord blood adiponectin, leptin, and C-reactive protein: Results of the 553 population-based Ulm Birth Cohort Studies. PLoS One. 2016;11(2). 555 556 Irwin MR, Opp MR. Sleep Health: Reciprocal Regulation of Sleep and Innate Immunity. D 52. Neuropsychopharmacol Rev. Nature Publishing Group; 2016; 53. TE 554 Azad MB, Konya T, Persaud RR, Guttman DS, Chari RS, Field CJ, et al. Impact of maternal intrapartum antibiotics , method of birth and breastfeeding on gut microbiota during the first year 558 of life: a prospective cohort study. BJOG. 2016;123:983–93. 54. 560 561 Matenchuk BA, Mandhane PJ, Kozyrskyj AL. Sleep, Circadian Rhythm, and the Gut Microbiota: AC C 559 EP 557 Considerations for Human Development. Sleep Med Rev. 55. Kozyrskyj AL, Kendall GE, Jacoby P, Sly PD, Zubrick SR. Association between socioeconomic 562 status and the development of asthma: Analyses of income trajectories. Am J Public Health. 563 2010;100(3):540–6. 564 56. 565 566 Organization WH. Births attended by skilled health personnel (%): Latest available year since 2005. Maternal health. 57. Shonkoff JP, Phillips DA. From Neurons to Neighborhoods: The Science of Early Childhood ACCEPTED MANUSCRIPT 567 Development. Washington, D.C.: National Academy Press; 2000. AC C EP TE D M AN US C RI PT 568 AC C EP TE D M AN U SC RI PT ACCEPTED MANUSCRIPT ACCEPTED MANUSCRIPT Highlights: EP TE D M AN US C RI PT Mean sleep duration at 3 months of age was 14.1 hours in 619 infants in Canada Maternal education & prenatal depression were associated with infant sleep duration Emergency cesarean section babies slept 1-hour less than those born vaginally Prenatal depression & birth mode mediate maternal education impact on infant sleep AC C • • • •