University of Southern California Dissertations and Theses

Association of maternal and environmental factors with infant feeding behaviors in a birth cohort study

(USC Thesis Other)

Association of maternal and environmental factors with infant feeding behaviors in a birth cohort study

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content ASSOCIATION OF MATERNAL AND ENVIRONMENTAL FACTORS WITH INFANT FEEDING BEHAVIORS IN A BIRTH COHORT STUDY by Huihui Zhang A Thesis Presented to the FACULTY OF THE USC GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree MASTER OF SCIENCE (APPLIED BIOSTATISTICS AND EPIDEMIOLOGY) August 2016 Copyright 2016 Huihui Zhang i TABLE OF CONTENTS Acknowledgements ii Abbreviations iii Abstract iv Introduction 1 Methods 4 Results 9 Discussion 22 References 26 ii ACKNOWLEDGEMENTS I would like to thank my advisor Dr. Carrie Breton for being a great mentor, allowing me to join her research and for her guidance and support throughout my research and writing process. In addition, I would like to thank my committee members Dr. Rob S. McConnell, Dr. Genevieve Dunton and Dr. Sandrah P. Eckel for their guidance and valuable suggestions. Finally, I would like to thank my family and friends for all their love, support, and encouragement. iii ABBREVIATIONS ABF: Any Breastfeeding BMI: Body Mass Index NSVD: Normal Spontaneous Vaginal Delivery iv ABSTRACT Background: Breastfeeding is well known to be beneficial to both infant and maternal health. Complementary feeding can also have direct or later consequences on infant health. WHO recommends that infants be exclusively breastfed for the first six months post-partum. However, adherence to this guideline in the U.S. is low. It is critical to understand determinants of breastfeeding duration and timing of complementary food introduction. Methods: This is a secondary analysis using data collected from the Maternal and Child Health Study. Forty-three mother-infant pairs followed through 9-months post-partum were assessed for this analysis. Maternal and infant characteristics (e.g. maternal age, maternal educational level, pre-pregnancy BMI, birth weight, baby’s gender as well as maternal smoking status) were obtained from baseline questionnaire and electronic medical records. Infant feeding practices were collected through the 9-month follow-up telephone interview. The relationships of potential predictors with breastfeeding duration and introducing time of formula milk as well as complementary foods were analyzed using Cox proportional hazards model while dichotomous outcomes such as early introduction of baby cereal were studied using an exact logistic model. Results: Babies with at least one reported occurrence of diarrhea during the 9 months (HR=2.6, 95% CL: 0.9, 7.2, p=0.06), mothers’ exposure to secondhand smoking (HR=3.0, 95% CL: 1.1, 8.6, p=0.04) and mothers who drank caffeinated beverages during pregnancy (HR=2.4, 95% CL: 0.7, 8.2, p=0.18) were more likely to stop any breastfeeding in our analysis. As for the start of formula milk, baby boys (HR=0.5, 95% CL: 0.2, 1.2, p=0.14) and babies with history of vomiting (HR=0.4, 95% CL: 0.2, 1.2, p=0.12) were less likely to start having formula milk early while higher pre-pregnancy BMI seemed to be positively associated with early introduction of formula milk (p=0.23). Babies’ history of diarrhea (p=0.29), pre-pregnancy BMI (p=0.56) and v mothers’ exposure to secondhand smoking (p=0.08) were positively related to early introduction of baby cereal. 1 INTRODUCTION Breastfeeding is beneficial to both infant and maternal health [3]. Breast milk is arguably the natural first and best food to be offered to infants at the initial stage of life since it has all nutrients necessary for growth and development of infants like water, protein, lactose, fat, minerals, vitamins and also provides numerous immune factors in preventing morbidity and mortality [2]. Infants who are breastfed for longer periods are at lower risk of developing respiratory tract infections and diarrhea [4]. They have fewer dental malocclusions and higher intelligence than do those who are breastfed for shorter periods or not breastfed [4]. Growing evidence also suggests that breastfeeding might protect against overweight later in life [5]. Studies of the overall effect of breastfeeding on maternal outcomes suggest that it decreases the risk of developing breast cancer [6] ovarian cancer [7] hypertension, cardiovascular disease and diabetes [8]. Women who breastfeed are more likely to return to pre-pregnancy weight earlier [8]. Complementary food is needed when breast milk (or infant formula) alone is no longer sufficient to meet late infant’s nutritional requirements [9]. It is defined as foods other than breastmilk, infant formula or follow-on formula given to infants and these can be liquids, semi-liquids, and solids [10]. Like breastfeeding, complementary feeding can also have direct or later consequences on infant health. Recent studies reveal that the early introduction of solid foods before six months of age can lead to an increased risk of diarrheal disease or gastro-intestinal infection in infancy [11], food allergies [12], and overweight or higher Body Mass Index (BMI) in childhood. Systematic review of available evidence suggests that early solid feeding may increase the risk of eczema [13]. The World Health Organization (WHO) recommends that infants be exclusively breastfed for the first six months postpartum, and that parents introduce nutritionally-adequate 2 and safe complementary (solid) foods at six months together with continued breastfeeding for up to two years and beyond [14], which aims to ensure infant’s optimal growth, development and health. However, adherence to these guidelines in the United States is low. The National Immunization Survey report showed that the rate of breastfeeding initiation was 77% and the rate for continued breastfeeding at 6 months was 49% in 2010, up from 35% in 2000. But only 13.3% of women exclusively breastfeed through six months postpartum [15]. The prevalence of early introduction of solid foods (4 months of age) in the United States has been estimated to range from 19% to 29%. Although there have been improvements in recent years, rates of feeding breast milk have still fall short of which recommended by WHO [16]. The best timing of complementary food introduction is still debated. In 2006, after reviewing 52 studies, the American College of Allergy, Asthma and Immunology concluded that early solid food introduction could increase the risk of food allergy and agreed with the recommendations on delayed introduction of certain foods until 6 months after birth for children at risk [17]. While another study revealed that children exposure to cereal grains after the age of 6 months may increase the risk of developing wheat allergy [18]. Findings from a large prospective birth cohort of more than 2500 infants showed delayed introduction of foods to be related with eczema, and atopy at age of two [19]. Introducing solid foods after the recommended 6 months of age is also not optimal due to the fact that it may cause deficiencies of zinc, protein, iron and vitamin D that in turn suppress growth, and cause feeding problems [20]. As a matter of fact, some studies suggest complementary feeding not be introduced before 17 weeks (four months of age) and not later than 26 weeks (six months of age) [26]. 3 For promotion of healthy infant feeding practices, it is critical to understand determinants of breastfeeding duration and timing of complementary food introduction. According to prior research, the most frequent factors associated with shorter duration of any breastfeeding were: smaller maternal age, lower infant birth weight [21], maternal smoking status [22], cesarean section delivery [23], lower maternal education level [24]. Additionally, maternal and infant sickness were significant contributors influencing early cessation of breastfeeding [25]. The predictors identified to be associated with early introduction of complementary food included smaller maternal age, mother smoking prior to pregnancy, maternal race/ethnicity [26], lower education level [27], baby’s well-being (i.e. diarrhea, wheeze, cough) [28]. Knowledge of these factors is a key first step for the formulation of effective interventions for improving the rates of breastfeeding. To gain additional insights into predictors of infant feeding practices in a primarily Hispanic population, we analyzed data from an ongoing prospective birth cohort study, the Maternal and Child Health Study. The main aim of this study was to examine the time of breastfeeding cessation as well as the status of early introduction of formula or baby cereal or pureed foods in 9 month old infants and to investigate whether infant diet and breast feeding are related to any individual-, household- and environmental-level factors. We also tested the association between early introduction of complementary foods and maternal health as well as infant physical status. 4 METHODS Study design This study is a secondary analysis of the Maternal and Child Health Study (AKA Cord Blood Study), which is an ongoing birth cohort study starting in 2012. The birth cohort was initiated to prospectively investigate the influence of environmental exposures on birth and health outcomes in their early life. Information was obtained from the baseline questionnaire and maternal medical record at delivery. Baseline questionnaire with 42 questions was answered by mothers either antepartum or during the postpartum recovery period through an interview conducted by a trained staff member. The questionnaire consisted of socio-demographic characteristics, maternal lifestyle factors as well as tobacco smoke variables. Data on infant and maternal characteristics, for instance, baby’s gender, birth weight, maternal age, gestational age, race/ethnicity, parity, delivery mode and mother’s health history were available via electronic medical records abstracted by a trained abstractor. Mother-infant pairs were followed by structured telephone interviews with mothers at 3, 9 and 12 months postpartum. During follow- up, data on infant feeding practices, such as breastfeeding experiences, the consumption of other types of milk and introduction of complementary foods was collected. It also included questions about infant’s health and medical history as well as residential moves. This analysis was conducted using data on infant feeding practices collected only at the 9-month follow-up visit. Study Population Expectant mothers were recruited from the labor and delivery ward of the LAC + USC Medical Center by a trained research coordinator. Mothers who agreed to join the study were 5 interviewed face-to-face to complete a baseline questionnaire prior to discharge from hospital. The 201 women who completed the baseline questionnaire were available for this study. Of these women, 43 mother-newborn pairs who also responded to the 9-month questionnaire were included in this analysis. The population is mostly Hispanic and of low socioeconomic status. Outcomes Breastfeeding in this study refers to any breastfeeding (ABF), which is defined as in addition to directly breastfeeding, the infant may also receive formula milk, other fluids or complementary foods. The main outcome measure is the duration of any breastfeeding during the first 9 months of life. The duration of breastfeeding in weeks was calculated from the dates of babies’ birth and the dates of breastfeeding cessation as stated by the mothers. Other outcome variables include the time when mothers introduced milk-based formula milk, baby cereal and baby pureed foods to infant diet. There were three variables related to baby cereal -- oat cereal, wheat cereal and rice cereal, which were combined together because of their similar distribution and the small sample size. Pureed fruits and vegetables were treated together as pureed foods. The age of the infant at the time that any food was introduced was recorded in weeks. In order to learn the relationship between early introduction of complementary foods and infant/maternal characteristics, time of starting having baby cereal or pureed foods was dichotomized into either ‘equal or less than 24 weeks’ or ‘more than 24 weeks’ where early introduction of other foods was considered to be before 24 weeks of age. 6 Explanatory Variables Maternal and infant characteristics known or suspected to be related with either breastfeeding practices or introduction of solid foods were investigated. The explanatory variables attributed to infant included their gender, birth weight, gestational age, birth order and health status including diarrhea, vomiting, high temperature (above 101 ℉/38 ℃), cough, runny nose/cold, rash/eczema and wheezing/trouble breathing. Maternal variables evaluated included maternal age, pre-pregnancy BMI, education level, marital status, employment condition, stress experience in the last month, delivery mode and the history of health outcome such as anemia, preeclampsia or high blood pressure and history of diabetes during pregnancy. Other variables included annual income and household exposure to tobacco smoke. Given the significance of infant feeding for child’s health, we also studied indicators such as taking folic acid tablets, drinking caffeinated beverages, or keeping dogs inside home during this pregnancy, which in general have rarely been assessed by previous studies. Among all the variables mentioned above, Delivery type was categorized as Normal spontaneous vaginal delivery (NSVD), Cesarean section; Birth order was coded as First, Second, Third and later; All the health condition variables were dichotomized as Yes or No; Continuous pre-pregnancy BMI was also categorized as Normal (BMI less than 25), Overweight (BMI from 25 to 29.9), Obese (BMI no less than 30) on the basis of height and weight measures from electrical medical record; Maternal education was broken down into three categories, consisting of Less than high school (<12 th grade), High school education (=12 th grade), Any college or higher; Family income for last year was counted as <$15,000, From $15,000 to $29,999, From $30,000 to $49,999, Do not know; Marital status was analyzed as Married, Living together, Never married or divorced; Employment condition was recorded as Homemaker, Student or 7 employed, Temporary medical leave, Unemployed. The frequency of having caffeinated drinks was defined as either <1 serving per day or >=1 serving per day. Maternal perceived stress was measured using a four-item version of the Perceived Stress Scale [26]. The scale contains the following questions: “In the last month, how often have you felt that you were unable to control the important things in your life”, “In the last month, how often have you felt confident about your ability to handle your personal problems”, “In the last month, how often have you felt that things were going your way”, “In the last month, how often have you felt difficulties were piling up so high that you could not overcome them”. As a dichotomous indicator of perceived stress, we chose a cut-point of ‘>=7’ to represent high stress [26]. We studied three variables regarding to maternal smoking status, including whether the mother has smoked directly during pregnancy or has any history of previous smoking, while second-hand smoking refers to whether anyone else living in the mother’s home smoked during this pregnancy. Statistical analyses Descriptive analyses were performed to examine the distribution of subject characteristics and socio-demographic variables. Additional analysis of baseline characteristics was conducted to see whether the 43 mothers who participated in the 9-month visit were different from the mothers who did not. Mean and standard deviation were calculated for approximately normally distributed variables while median was given for other continuous variables. Number and percentages were estimated for categorical factors. All the outcome variables were examined against a set of independent factors that were identified from the literature as having possible effects on them. The associations of any breastfeeding duration with these variables were 8 assessed using Cox proportional hazards model. This model allows joint estimation of the influence of independent variables on the risk of cessation of breastfeeding and can be used to analyze data that contain censored observations. The participants who were still breastfeeding when they completed the 9-month questionnaire (51.2%) were considered as right censored. The same model was applied to the age at which formula milk was first introduced and age of first introduction of baby cereal or baby pureed foods. Additional analyses were conducted where baby cereal and baby pureed foods outcomes were dichotomized into two categories with cut-off point of 24 weeks to explore effects of explanatory factors on early introduction of complementary foods. For the reason that the number of observations in certain category was less than 5, an Exact logistic regression model was fitted for these dichotomous outcomes. Due to the small sample size, the model selection strategy for all models denoted above included a first evaluation of the univariate associations between individual variables and selected outcomes. Any variable with a P-value of <0.25 according to the Wald statistics was then included in a multivariate model which was reduced using a procedure similar to a backward elimination method. Since small sample size makes it harder to find statistical significance, which means it may fail to meet the p-value criterion even though there is an effect. Unlike the traditional implementation of backward selection, the statistic that was mainly used to determine whether to drop an effect was not the significance level but the change of standard error. At any step, the variable was dropped when its standard error increased ≥20% in the multivariate model as compared to the univariate model. Analyses were conducted with SAS software (Version 9.4; SAS Institute Inc.). 9 RESULTS Baseline Characteristics of Participants Two hundred and one mother-infant pairs in total were recruited at the beginning of the study. Forty-three observations followed at 9-month were assessed for this analysis. Table 1A shows the baseline characteristics of the 43 mothers and their babies. Table 1B displays the characteristics for the rest of mother-infant pairs who did not participate in the analysis. No significant difference was found among them except for pre-pregnancy BMI (p-value=0.027) and lifetime smoking status (p-value=0.033). Mothers who did not participate were of lower pre- pregnancy BMI (median [IQR]=27.5 [7.1]) and fewer of them had ever smoked during their lifetime (23.2%). Participants were primarily Hispanic Whites (90.7%). The majority of the infants were the firstborn baby (51.2%) and 16.3% of them were the second born baby of the mother. Most of the mothers had attended high school (37.2%) and college or higher (30.2%), with 32.6% mothers whose educational level lower than high school (12 th Grade). A little less than half of the children were male (44.2%). The median [IQR] of continuous, non-normally distributed variables were: birth weight (3120 grams [545]), gestational age (39 weeks [2]), maternal age (27 years [10]) and pre-pregnancy BMI (29.6 [10.4]). Forty-seven percent of the mothers were obese and 25% were overweight while only 28% mothers were normal weight. Forty percent of women reported ever having smoked in their lifetime but only 3 of these mothers said they have smoked during this pregnancy. Nineteen percent of the mothers were exposed to secondhand smoking during this pregnancy. Taking folic acid tablets during pregnancy (41.9%) and drinking caffeinated beverages (69.8%) were prevalent in this population. 10 Four of the 43 mothers have never breastfed during the first 9 months of infant’s life. Twenty-two were still breastfeeding at the 9-month interview. Thirty-three of 43 babies had drunk formula milk (milk-based) during their first 9 months of life. All of the 43 infants were singleton birth. Table 1A. Selected demographic and clinical characteristics of study subjects Baby factors N (%) Birth weight, grams *, † 3120 [545] Gestational age, weeks *, † 39 [2] Baby's gender Boy 19 (44.2) Girl 24 (55.8) Delivery type † Normal spontaneous vaginal delivery (NSVD) 25 (65.8) Cesarean section 13 (34.2) Birth order First 22 (51.2) Second 7 (16.3) Third and later 14 (32.5) Baby ever had Diarrhea 16 (37.2) Baby ever had vomiting 11 (25.6) Baby ever had high temperature (above 101F/38C) 18 (41.9) Baby ever had cough 22 (51.2) Baby ever had runny nose/cold 30 (69.8) Baby ever had rash/eczema 19 (44.2) Baby ever had wheezing/trouble breathing 8 (18.6) Baby ever been taken to the doctor/urgent room 25 (58.1) Maternal factors N (%) Maternal Age, years *, † 27 [10] Pre-pregnancy BMI *, † 29.6 [10.4] Pre-pregnancy BMI † Normal 10 (27.8) Overweight 9 (25.0) Obese 17 (47.2) Maternal education <high school (<12th grade) 14 (32.6) High school (12th grade) 16 (37.2) Any college 13 (30.2) Have ever smoked in lifetime 17 (39.5) Have smoked during pregnancy 3 (15.8) Exposed to secondhand smoking during pregnancy 8 (18.6) Anemia or low hemoglobin in blood during pregnancy 12 (27.9) Preeclampsia or high blood pressure during pregnancy 7 (16.3) History of diabetes(types I,II or gestational) † 7 (17.1) Total household family income for last year <$15,000 15 (34.9) $15,000 to $29,999 11 (25.6) 11 $30,000 to $49,999 4 (9.3) Do not know 13 (30.2) Stress experience in the last month 24 (55.8) Marital status Married 17 (39.5) Living together 16 (37.2) Never married or divorced 10 (23.3) Work during this pregnancy 20 (46.5) Current employment status Homemaker 15 (34.9) Student or employed 7 (16.3) Temporary medical leave 9 (20.9) Unemployed 12 (27.9) Take folic acid tablets during pregnancy 18 (41.9) Keep dog inside your home during pregnancy 10 (23.3) Mother's Race Hispanic White 39 (90.7) Other 4 (9.3) Father's Race Hispanic White 36 (83.7) Non-Hispanic White 1 (2.3) Other 6 (14.0) Drink caffeinated beverages during this pregnancy 30 (69.8) Frequency of drinking caffeinated beverages (missing=14) >=1/day 9 (31.0) <1/day 20 (69.0) * Median [IQR] was reported for continuous variables; categorical data was presented as N (frequency). † Numbers do not add up due to missing data. Table 1B. Selected demographic and clinical characteristics of people who did not participate. Baby factors N (%) Birth weight, grams *, † 3230 [655] Gestational age, weeks *, † 39 [3] Baby's gender Boy 59 (53.2) Girl 52 (46.9) Delivery type † Normal spontaneous vaginal delivery (NSVD) 56 (60.2) Cesarean section 37 (39.8) Birth order First 74 (49.3) Second 40 (26.7) Third and later 36 (24.0) Maternal factors N (%) Maternal Age, years *, † 27 [10] Pre-pregnancy BMI *, † 27.5 [7.1] Pre-pregnancy BMI † Normal 32 (34.4) Overweight 31 (33.3) 12 Obese 30 (32.3) Maternal education <high school (<12th grade) 51 (33.1) High school (12th grade) 53 (34.4) Any college 50 (32.5) Have ever smoked in lifetime 36 (23.2) Exposed to secondhand smoking during pregnancy 22 (14.2) Anemia or low hemoglobin in blood during pregnancy 44 (28.4) Preeclampsia or high blood pressure during pregnancy 22 (14.2) History of diabetes(types I,II or gestational) † 19 (17.1) Total household family income for last year <$15,000 47 (30.7) $15,000 to $29,999 47 (30.7) $30,000 to $49,999 12 (7.8) Do not know 47 (30.7) Stress experience in the last month 91 (58.7) Marital status Married 41 (26.8) Living together 70 (45.8) Never married or divorced 42 (27.4) Work during this pregnancy 20 (46.5) Current employment status Homemaker 70 (45.7) Student or employed 35 (22.9) Temporary medical leave 19 (12.4) Unemployed 29 (19.0) Take folic acid tablets during pregnancy 107 (69.0) Keep dog inside your home during pregnancy 40 (25.6) Mother's Race Hispanic White 139 (89.7) Other 16 (10.3) Father's Race Hispanic White 136 (87.7) Non-Hispanic White 5 (3.2) Other 14 (9.0) Drink caffeinated beverages during this pregnancy 112 (73.2) Frequency of drinking caffeinated beverages (missing=14) >=1/day 36 (33.3) <1/day 72 (66.7) * Median [IQR] were reported for continuous variables; categorical data were presented as N (frequency). † Numbers do not add up due to missing data. Association of potential factors with time to stop breastfeeding Table 2 shows the univariate associations of numerous potential predictors with duration of ABF and time to introduction of milk-based formula milk. Mothers whose baby had at least one 13 reported occurrence of diarrhea during the 9 months were 4.2 times as likely to stop breastfeeding as mothers whose baby had not had diarrhea (HR=4.2, 95% CL=1.7, 10.4, p=0.002). Birth weight was slightly positively significantly related to risk of stopping any breastfeeding (HR=1.001, 95% CL=1.000, 1.002, p=0.08). The continuous pre-pregnancy BMI was statistically significantly associated with time of breastfeeding cessation (p=0.04). There is a 4% increase in the expected risk of breastfeeding cessation for each unit increase in pre- pregnancy BMI. Mothers who were exposed to secondhand smoking during pregnancy were 5.7 times more likely to stop breastfeeding (HR=5.7, 95% CL=2.2, 14.9, p=0.0003). There was a marginally significant association between drinking caffeinated beverages during pregnancy and duration of any breastfeeding (p=0.06). Mothers who drank caffeinated beverages were 3.2 times as likely to stop breastfeeding compared to mothers who did not (HR=3.2, 95% CL=1.0, 11.0). Moreover, the mother who drank more than one cup of caffeinated beverage per day was 2.7 more likely to stop breastfeeding than mothers who drank less than one cup a day (HR=2.73, 95% CL=1.02, 7.31, p=0.046). Table 3 displays the adjusted model for time to breastfeeding cessation. In the model, secondhand smoking was statistically significantly associated with time to breastfeeding cessation (adjusted p=0.04), indicating that mothers exposed to secondhand smoking during pregnancy were 3.0 times more likely to stop breastfeeding than mothers who did not after adjusting for baby had diarrhea, and mother drank caffeinated beverages during this pregnancy (adjusted HR=3.0, 95% CL=1.1, 8.6). There was a marginal association between baby having at least one reported occurrence of diarrhea and duration of any breastfeeding (Adjusted p=0.06). Moreover, women who drank caffeinated beverages during pregnancy tended to stop breastfeeding after controlling for baby had diarrhea and secondhand smoking (adjusted p=0.18). 14 Table 2. Univariate analysis of potential predictors with time to stop ABF and time to the start of formula milk Time to stop ABF † Time to the start of formula milk † Crude HR (95% CL) P value Nobs/ Censored* Crude HR P value Nobs Anemia or low hemoglobin in blood 1.1 (0.4, 2.9) 0.82 43/22 1.3 (0.6, 2.9) 0.51 33 Baby had Diarrhea 4.2 (1.7, 10.4) 0.002 43/22 1.2 (0.6, 2.4) 0.66 33 Baby had vomiting 0.8 (0.3, 2.2) 0.69 43/22 0.5 (0.2, 1.3) 0.15 33 Baby had high temperature 1.01 (0.43, 2.40) 0.98 43/22 1.1 (0.5, 2.2) 0.84 33 Baby had cough 0.6 (0.3, 1.5) 0.31 43/22 1.3 (0.6, 2.7) 0.45 33 Baby had runny nose/cold 1.6 (0.6, 4.3) 0.38 43/22 2.7 (1.1, 6.7) 0.04 33 Baby had rash/eczema 1.5 (0.6, 3.5) 0.37 43/22 1.5 (0.7, 3.0) 0.31 33 Baby had wheezing/trouble breathing 1.5 (0.6, 4.2) 0.40 43/22 1.6 (0.7, 3.9) 0.28 33 Baby been taken to the doctor 1.1 (0.5, 2.7) 0.79 43/22 1.1 (0.6, 2.3) 0.74 33 Birth order 0.94 43/22 0.77 33 First 0.9 (0.4, 2.5) 0.90 0.9 (0.4, 2.0) 0.79 Second 1.2 (0.3, 4.0) 0.81 0.7 (0.2, 2.0) 0.47 Third and later ‡ Birth weight (grams) 1.001 (1.000, 1.002) 0.08 42/22 1.00 (0.99, 1.00) 0.54 32 Boy 1.5 (0.6, 3.4) 0.39 43/22 0.5 (0.3, 1.1) 0.09 33 Current employment status 0.40 43/22 0.54 33 Homemaker ‡ Student or employed 2.0 (0.6, 6.5) 0.26 0.5 (0.2, 1.4) 0.22 Temporary medical leave 1.9 (0.6, 5.9) 0.26 0.9 (0.3, 2.2) 0.74 Unemployed 0.8 (0.2, 2.9) 0.77 0.5 (0.2, 1.5) 0.23 Drink caffeinated beverages 3.2 (1.0, 11.0) 0.06 43/22 0.8 (0.4, 1.8) 0.59 33 Frequency of drinking caffeinated beverages 29/12 24 >=1/day 2.73 (1.02, 7.31) 0.046 0.99 (0.42, 2.33) 0.97 <1/day ‡ Gestational age (weeks) 1.1 (0.8, 1.4) 0.53 42/22 0.9 (0.8, 1.1) 0.53 32 History of diabetes(types I,II or gestational) 0.8 (0.2, 2.9) 0.77 41/22 1.02 (0.41, 2.52) 0.97 31 Keep dog inside your home 1.4 (0.5, 3.6) 0.48 43/22 1.4 (0.6, 3.2) 0.45 33 Marital status 0.07 43/22 0.75 33 Living together ‡ Married 0.8 (0.3, 2.3) 0.64 1.1 (0.5, 2.4) 0.91 Never married or divorced 2.5 (0.9, 6.9) 0.08 1.4 (0.5, 3.5) 0.49 Maternal Age (years) 0.99 (0.92, 1.07) 0.86 42/22 1.04 (0.99, 1.10) 0.16 32 Maternal education 0.97 43/22 0.19 33 <high school (<12th grade) ‡ High school (12th grade) 0.89 (0.32, 2.46) 0.82 0.4 (0.2, 1.1) 0.07 Any college 0.90 (0.30, 2.69) 0.86 0.7 (0.3, 1.7) 0.45 Mode of birth (Delivery type) 38/21 28 NSVD 1.2 (0.4, 3.4) 0.73 0.9 (0.4, 1.9) 0.70 Cesarean section ‡ Preeclampsia or high blood pressure 0.19 (0.03, 1.38) 0.10 43/22 1.2 (0.4, 3.5) 0.74 33 Pre-pregnancy BMI 1.04 (1.00, 1.07) 0.04 36/16 1.02 (0.99, 1.05) 0.30 28 15 Pre-pregnancy BMI (categorical) 0.47 36/16 0.24 28 Normal ‡ Overweight vs. normal 1.9 (0.5, 7.1) 0.34 2.1 (0.6, 7.0) 0.22 Obese vs. normal 2.0 (0.6, 6.3) 0.23 2.5 (0.9, 7.4) 0.09 Smoker 1.04 (0.43, 2.50) 0.94 43/22 1.5 (0.7, 3.1) 0.27 33 Secondhand smoking 5.7 (2.2, 14.9) 0.0003 43/22 1.3 (0.6, 3.1) 0.48 33 Stress experience in the last month 0.9 (0.4, 2.0) 0.72 43/22 1.0 (0.5, 2.1) 1.00 33 Take folic acid tablets 0.6 (0.3, 1.6) 0.34 43/22 1.5 (0.7, 3.1) 0.30 33 Total household family income last year 0.21 43/22 0.03 33 <$15,000 ‡ $15,000 to $29,999 2.2 (0.6, 7.8) 0.23 0.3 (0.1, 0.8) 0.02 $30,000 to $49,999 2.6 (0.5, 14.0) 0.28 1.7 (0.5, 5.6) 0.38 Do not know 3.6 (1.1, 11.7) 0.03 0.38 (0.14, 1.03) 0.06 Work during this pregnancy 1.4 (0.6, 3.3) 0.44 43/22 0.6 (0.3, 1.3) 0.20 33 * There were 22 of 43 mothers still breastfeeding when filling the 9-month questionnaire. † Crude Hazard ratio, 95% confidence limits and p-value were obtained from Cox proportional hazard model. ‡ Reference group. Table 3. Multivariate analysis of potential predictors with time of ABF cessation Unadjusted analysis Adjusted analysis * Variables Crude HR (95% CL) Std † Err P value Adjusted HR (95% CL) Std † Err P value Baby had diarrhea 4.2 (1.7, 10.4) 0.46 0.002 2.6 (0.9, 7.2) 0.52 0.06 Drink caffeinated beverages 3.2 (1.0, 11.0) 0.62 0.06 2.4 (0.7, 8.2) 0.64 0.18 Secondhand smoking 5.7 (2.2, 14.9) 0.49 0.0003 3.0 (1.1, 8.6) 0.53 0.04 * Adjusted analysis was based on other two variables in this table using Cox proportional hazard model. † Standard error. Association of potential factors with time of starting formula milk (milk-based) In univariate analyses (Table 2), babies with a history of runny noses/colds in the first 9 months of life were 2.7 times as likely to start drinking formula milk compared to babies who had not had colds (HR=2.7, 95% CL=1.1, 6.7, p=0.04) while babies who had at least one occurrence of vomiting were less likely to start formula milk (HR=0.5, 95% CL=0.2, 1.3, p=0.15). Mothers tended to give boys formula milk earlier than girls (p=0.09). As for pre- pregnancy BMI, no significant association was found for continuous BMI. However, for 16 categorical BMI, obese mothers were 2.5 (HR=2.5, 95% CL=0.9, 7.4) times more likely to give formula milk to their babies than mothers with normal weight while overweight mothers had 2.1 (HR=2.1, 95% CL=0.6, 7.0) times more probability to introduce formula milk than mothers with normal weight (overall p=0.24). Compared to duration of ABF, although some variables had no significant association with time to the start of formula milk, the direction of association was the same with each other. For instance, babies with history of diarrhea were more likely to stop any breastfeeding (p=0.002) as well as start having formula milk (p=0.66). Same direction of association was also found for babies who had vomiting (p=0.69 for ABF, p=0.15 for formula milk), babies had runny noses/colds (p=0.38 for ABF, p=0.04 for formula milk) and for mothers who drank caffeinated beverages (p=0.06 for ABF, p=0.59 for formula milk). Moreover, mothers who were exposed to secondhand smoking during pregnancy were more likely to stop ABF (p=0.0003) and start giving formula milk to their babies (p=0.48). These results show consistent associations across the two outcomes as expected. In a multivariable model, no variables remained statistically significant at a 0.05 type I error although direction of effects were consistent with univariate analyses. Categorical BMI was positively correlated to time to the start of baby formula milk after controlling for baby ever had vomiting and baby’s gender (Table 4, adjusted p=0.23). Obese mothers were 2.2 times as likely to give their babies formula milk than mothers who were normal weight (adjusted HR=2.2, 95% CL=0.7, 6.4). However, mothers who were overweight before pregnancy had higher probability of introducing baby formula milk than those with normal weight (adjusted HR=3.1, 95% CL=0.8-11.8). Babies with history of vomiting as well as boys were more likely to be given formula milk later after considering other factors (Table 4, adjusted p=0.12, 0.14 respectively). 17 Adjusting for pre-pregnancy BMI (categorical) and babies’ history of vomiting, boys were half as likely to be given formula milk than girls (adjusted HR=0.5, 95% CL=0.2, 1.2). Table 4. Multivariate analysis of potential predictors with time to the start of formula milk Unadjusted analysis Adjusted analysis * Variables Crude HR (95% CL) Std † Err P value Adjusted HR (95% CL) Std † Err P value Pre-pregnancy BMI_categorical 0.24 0.23 Overweight vs. normal 2.1 (0.6, 7.0) 0.61 0.22 3.1 (0.8, 11.8) 0.69 0.10 Obese vs. normal 2.5 (0.9, 7.4) 0.55 0.09 2.2 (0.7, 6.4) 0.55 0.16 Baby had vomiting 0.5 (0.2, 1.3) 0.44 0.15 0.4 (0.2, 1.2) 0.53 0.12 Baby's gender--Boy 0.5 (0.3, 1.1) 0.38 0.09 0.5 (0.2, 1.2) 0.43 0.14 * Adjusted analysis was based on other two variables in this table using Cox proportional hazard model. † Standard Error. Association of potential factors with introduction of baby cereal We studied time to the introduction of baby cereal and whether an infant starts having baby cereal before 24 weeks of life versus after 24 weeks (Table 5). On the left hand of the table is time to start having baby cereal using Cox proportional hazard model, on the right side is dichotomous outcome modeling on introduction of baby cereal earlier than 24 weeks of babies’ life using exact logistic regression model. Before adjusting for potential risk factors, mothers who were exposed to secondhand smoking were 2.4 times as likely to give the infant cereal as women who were not exposed (p=0.04), which is consistent with dichotomous outcome that mothers exposed to secondhand smoking tended to introduce baby cereal before 24 weeks (p=0.04). A history of baby having diarrhea increased the risk of introducing baby cereal (p=0.21). A similar association was seen with cereal introduced earlier than 24 weeks of babies’ life (p=0.16). Pre-pregnancy BMI was also associated with introduction of cereal (p=0.18 from Cox proportional hazard model, p=0.17 from exact logistic regression model). The heavier the 18 mothers were before pregnancy, the more likely were they to introduce baby cereal earlier (HR=1.02, 95% CL=0.99, 1.05 in Cox proportional hazard model; OR=0.96, 95% CL=0.89, 1.02 in exact logistic regression model). Although we were interested in the effect of stress experience and caffeinated beverages on introduction of baby cereal, no significant results were observed. After adjustment for pre-pregnancy BMI and baby history of diarrhea during the 9 months of life, secondhand smoking was marginally associated with introduction of baby cereal (Table 6, p=0.08). Mothers who were exposed to secondhand smoking during pregnancy were 2.2 times as likely to give their infants cereal compared to mothers who were not (adjusted HR=2.2, 95% CL=0.9, 5.4). Even though not significant, babies with history of diarrhea were 1.6 times as likely to have cereal as babies without a history after controlling for pre-pregnancy BMI and secondhand smoking factors (Table 6, HR=1.6, 95% CL=0.7, 4.0, adjusted p=0.286). Similar results were shown from analyses that were conducted using exact logistic regression for categorizing cereal, which further strengthened the findings. Table 5. Univariate analysis of potential predictors with time to the start of baby cereal Time to the start of baby cereal * Early vs. late introduction of baby cereal † Crude HR (95% CL) P value Crude OR (95% CL) P value Nobs Anemia or low hemoglobin in blood 0.8 (0.4, 1.6) 0.51 0.2 (0, 0.9) 0.08 40 Baby had Diarrhea 1.5 (0.8, 2.9) 0.21 4.1 (0.6, 32.1) 0.16 40 Baby had vomiting 1.01 (0.5, 2.1) 0.97 1.2 (0.1, 8.9) 1.00 40 Baby had high temperature 1.1 (0.6, 2.1) 0.81 0.9 (0.1, 5.5) 1.00 40 Baby had cough 1.3 (0.7, 2.4) 0.45 1.9 (0.3, 14.0) 0.69 40 Baby had runny nose/cold 1.2 (0.6, 2.4) 0.65 0.7 (0.1, 5.1) 0.90 40 Baby had rash/eczema 1.2 (0.6, 2.3) 0.56 2.4 (0.4, 18.1) 0.47 40 Baby had wheezing/trouble breathing 1.1 (0.5, 2.5) 0.80 1.8 (0.1, 14.7) 0.86 40 Baby been taken to the doctor 1.2 (0.6, 2.2) 0.64 0.7 (0.1, 4.4) 0.93 40 Birth order 0.50 0.23 40 First 0.66 (0.33, 1.34) 0.25 0.3 (0.02, 2.43) 0.35 Second 0.69 (0.25, 1.88 0.47 1.2 (0.1, 13.8) 1.00 Third and later ‡ Birth weight (grams) 1 (0.999, 1.001) 0.75 1.0 (0.998, 1.001) 0.66 39 Boy 1.05 (0.6. 2.0) 0.89 1.5 (0.2, 9.3) 0.93 40 Current employment status 0.51 0.95 40 Homemaker ‡ Student or employed 0.8 (0.3, 2.1) 0.71 1.4 (0.1, 17.4) 1.00 Temporary medical leave 0.9 (0.4, 2.2) 0.90 0.5 (0.008, 7.2) 0.97 19 Unemployed 0.5 (0.2, 1.3) 0.14 0.9 (0.1, 10.2) 1.00 Drink caffeinated beverages 1.1 (0.5, 2.1) 0.89 1.4 (0.2, 16.0) 1.00 40 Frequency of drinking caffeinated beverages 27 >=1/day 1.6 (0.7, 3.9) 0.28 1.7 (0.1, 19.6) 0.94 <1/day ‡ Gestational age (weeks) 0.9 (0.8, 1.2) 0.62 0.8 (0.5, 1.3) 0.33 39 History of diabetes(types I,II or gestational) 1.1 (0.5, 2.5) 0.85 0.6 (0.01, 6.2) 1.00 40 Keep dog inside your home 1.02 (0.5, 2.2) 0.95 0.4 (0.01 4.4) 0.82 40 Marital status 0.42 0.70 40 Married 1.6 (0.7, 3.6) 0.22 1.4 (0.1, 19.1) 1.00 Living together 1.6 (0.7, 3.7) 0.30 2.5 (0.2, 36.5) 0.66 Never married or divorced ‡ Maternal Age (years) 1.00 (0.95, 1.05) 0.95 0.9 (0.8, 1.1) 0.28 39 Maternal education 0.17 0.13 40 <high school (<12th grade) ‡ High school (12th grade) 0.5 (0.2, 1.0) 0.06 0.1 (0.002, 1.5) 0.13 Any college 0.6 (0.3, 1.4) 0.27 0.3 (0.02, 2.5) 0.38 Mode of birth (Delivery type) 35 NSVD 0.9 (0.4, 1.8) 0.74 0.2 (0.2, 1.7) 0.18 Cesarean section ‡ Preeclampsia or high blood pressure 0.8 (0.3. 2.1) 0.62 1.0 (0.02, 12.5) 1.00 40 Pre-pregnancy BMI 1.02 (0.99, 1.05) 0.18 1.05 (0.98, 1.12) 0.17 33 Pre-pregnancy BMI (categorical) 0.34 0.07 33 Normal ‡ Overweight 1.3 (0.5, 3.4) 0.62 1.3 (0, 23.8) 1.00 Obese 1.9 (0.8, 4.4) 0.16 5.6 (0.5, 306.0) 0.24 Smoker 1.3 (0.7, 2.5) 0.46 0.9 (0.1, 5.5) 1.00 40 Secondhand smoking 2.4 (1.0, 5.5) 0.04 8.9 (1.1, 87.0) 0.04 40 Stress experience in the last month 1.3 (0.7, 2.5) 0.42 1.3 (0.2, 9.7) 1.00 40 Take folic acid tablets 0.8 (0.4, 1.6) 0.54 0.1 (0.003 1.2) 0.09 40 Total household family income last year 0.69 0.69 40 <$15,000 ‡ $15,000 to $29,999 1.5 (0.6, 3.6) 0.35 1.8 (0.2, 18.0) 0.87 $30,000 to $49,999 0.85 (0.28, 2.61) 0.78 0.8 (0, 6.5) 0.89 Do not know 0.94 (0.44, 2.00) 0.88 0.7 (0.05 7.2) 1.00 Work during this pregnancy 0.98 (0.5, 1.8) 0.94 0.6 (0.1 3.8) 0.82 40 * Hazard ratio, 95% confidence limits and p-value were obtained using Cox proportional hazard model. † Odds ratio, 95% confidence limits and p-value were derived from exact logistic regression model. ‡ Reference group. Table 6. Multivariate analysis of potential predictors with time to the start of baby cereal Unadjusted analysis Adjusted analysis * Variables Crude HR (95% CL) Std † Err P value Adjusted HR (95% CL) Std † Err P value Baby had Diarrhea 1.5 (0.8, 2.9) 0.34 0.21 1.6 (0.7, 4.0) 0.40 0.29 Pre-pregnancy BMI 1.02 (0.99, 1.05) 0.015 0.18 1.01 (0.98, 1.05) 0.017 0.56 Secondhand smoking 2.4 (1.0, 5.5) 0.43 0.04 2.2 (0.9, 5.4) 0.46 0.08 * Adjusted analysis was based on other two variables in this table using Cox proportional hazard model. † Standard error. 20 Association of potential factors with introduction of pureed foods According to unadjusted models (Table 7), no significant relationship was found for pureed foods not only in Cox proportional hazard models or in exact logistic regression models. In univariate analysis, the only variable that met the criteria to include in the adjusted model was lifetime maternal smoking status. Mothers who had ever smoked were more likely to give pureed foods to their babies earlier during first 9 months of life (p=0.18). Table 7. Univariate analysis of potential predictors with time to the start of pureed foods Time to the start of pureed foods * Early vs. late introduction of pureed foods † Crude HR (95% CL) P value Crude OR (95% CL) P value Nobs Anemia or low hemoglobin in blood 1.01 (0.51, 1.99) 0.99 0.5 (0.05, 3.3) 0.71 40 Baby had Diarrhea 1.1 (0.6, 2.1) 0.81 1.3 (0.2, 7.2) 0.98 40 Baby had vomiting 1.1 (0.5, 2.3) 0.83 2.6 (0.4, 15.9) 0.39 40 Baby had high temperature 1.2 (0.6, 2.3) 0.57 1.0 (0.2, 5.3) 1.00 40 Baby had cough 1.1 (0.6, 2.1) 0.72 1.0 (0.2, 5.4) 1.00 40 Baby had runny nose/cold 1.3 (0.6, 2.7) 0.45 1.7 (0.3, 19.5) 0.86 40 Baby had rash/eczema 1.2 (0.6, 2.2) 0.61 3.4 (0.6, 24.4) 0.20 40 Baby had wheezing/trouble breathing 0.9 (0.4, 2.2) 0.90 2.7 (0.3, 20.6) 0.46 40 Baby been taken to the doctor 1.3 (0.7, 2.6) 0.36 2.0 (0.4, 14.3) 0.59 40 Birth order 0.60 0.89 40 First 0.72 (0.35, 1.49) 0.38 0.8 (0.1, 4.9) 1.00 Second 0.67 (0.27, 1.71) 0.41 0.4 (0.01, 5.4) 0.81 Third and later ‡ Birth weight (grams) 1 (0.999, 1.001) 0.74 1.0 (0.998, 1.001) 0.97 39 Boy 1.02 (0.5, 2.0) 0.96 2.5 (0.5, 15.1) 0.36 40 Current employment status 0.38 1.00 40 Homemaker ‡ Student or employed 0.6 (0.2, 1.5) 0.28 1.0 (0.1, 10.3) 1.00 Temporary medical leave 1.1 (0.5, 2.5) 0.87 0.7 (0.1, 6.9) 1.00 Unemployed 0.5 (0.2, 1.2) 0.14 0.6 (0.05, 5.9) 1.00 Drink caffeinated beverages 0.9 (0.4, 1.8) 0.71 1.0 (0.2, 7.4) 1.00 40 Frequency of drinking caffeinated beverages 28 >=1/day 1.2 (0.5, 2.9) 0.61 1.0 (0.1, 8.6) 1.00 <1/day ‡ Gestational age (weeks) 0.995 (0.8, 1.2) 0.96 1.3 (0.8, 2.2) 0.26 39 History of diabetes(types I,II or gestational) 1.2 (0.5, 2.7) 0.69 1.1 (0.1, 8.9) 1.00 40 Keep dog inside your home 1.2 (0.6, 2.5) 0.68 0.3 (0.01, 2.5) 0.41 40 Marital status 0.53 0.15 40 Married 1.5 (0.7, 3.3) 0.26 1.4 (0.1, 19.1) 1.00 Living together 1.3 (0.6, 2.9) 0.57 5.5 (0.6, 76.8) 0.15 Never married or divorced ‡ Maternal Age (years) 0.99 (0.95, 1.04) 0.78 0.9 (0.8, 1.0) 0.17 39 Maternal education 0.32 0.66 40 <high school (<12th grade) ‡ High school (12th grade) 0.5 (0.2, 1.2) 0.13 0.5 (0.1, 3.5) 0.68 21 Any college 0.7 (0.3, 1.6) 0.42 0.4 (0.03, 3.0) 0.52 Mode of birth (Delivery type) 35 NSVD 1.2 (0.6, 2.5) 0.66 0.9 (0.1, 6.9) 1.00 Cesarean section ‡ Preeclampsia or high blood pressure 0.97 (0.4, 2.3) 0.94 0.6 (0.01, 6.1) 1.00 40 Pre-pregnancy BMI 1.01 (0.98, 1.04) 0.55 1.02 (0.96, 1.08) 0.58 33 Pre-pregnancy BMI (categorical) 0.71 0.61 33 Normal ‡ Overweight 1.39 (0.5, 3.7) 0.52 0.4 (0.01, 5.7) 0.76 Obese 1.40 (0.6, 3.2) 0.43 1.2 (0.2, 10.0) 1.00 Smoker 1.6 (0.8, 3.1) 0.18 1.7 (0.3, 9.3) 0.70 40 Secondhand smoking 1.3 (0.6, 2.9) 0.46 2.1 (0.3, 14.4) 0.62 40 Stress experience in the last month 1.3 (0.7, 2.5) 0.47 0.6 (0.1, 3.2) 0.70 40 Take folic acid tablets 0.95 (0.5, 1.8) 0.87 0.8 (0.1, 4.0) 1.00 40 Total household family income last year 0.75 0.77 40 <$15,000 ‡ $15,000 to $29,999 1.3 (0.6, 3.1) 0.54 1.2 (0.1, 10.5) 1.00 $30,000 to $49,999 0.7 (0.2, 2.2) 0.56 0.6 (0, 4.1) 0.65 Do not know 0.9 (0.4, 1.9) 0.72 0.8 (0.1, 5.8) 1.00 Work during this pregnancy 0.9 (0.5, 1.7) 0.78 0.8 (0.1, 4.0) 1.00 40 * Hazard ratio, 95% confidence limits and p-value were obtained using Cox proportional hazard model. † Odds ratio, 95% confidence limits and p-value were obtained using exact logistic regression model. ‡ Reference group. 22 DISCUSSION In our analysis, we investigated the associations of maternal and infant characteristics with timing of ABF cessation, time to the start of formula milk, time of introduction of baby cereal as well as pureed foods among 43 mother-infant dyads recruited from LAC-USC Medical Center. The results regarding ABF cessation are as follows. Our analyses show that mothers whose baby had at least one reported occurrence of diarrhea were more likely to stop ABF both in univariate and multivariate analysis, which is consistent with the finding of a study carried out by Nassar et al [25]. A few infants could get diarrhea because of breast milk, such as lactose intolerance. When the baby has diarrhea, the mother tends to try other infant milk and might stop breastfeeding. Studies have shown BMI to be one factor underlying duration of ABF [29, 30]. The study in Denmark [30] showed a significant negative association between high maternal BMI and duration of ABF. Obese women also had a higher probability of experiencing early postpartum breastfeeding difficulty because of milk-supply problems as compared to normal- weight women [31]. The results of our study supports the idea suggesting that continuous pre- pregnancy BMI is related to timing of ABF cessation, although no statistically significant association was found in multivariate analysis. We also found that second-hand smoking may have positive association with early discontinuation of ABF. This result agrees with a study in United States, in which maternal smoking status was found to be associated with early discontinuation of ABF [22]. Chronic smoking could compromise breast feeding by suppressing prolactin secretion and thereby lowering breast milk volume [36]. In our study, women who drank caffeinated beverages during pregnancy tended to stop ABF after controlling for other factors. Caffeine ingestion is a frequent concern for infant’s health. A study conducted in Bovaria found that coffee consumption had a significant negative influence on the breastfeeding 23 duration (OR=1.5, 95% CI: 1.3, 1.8) [35]. One study has indicated that chronic coffee drinking might decrease iron content of breastmilk [34]. But they also mentioned that moderate maternal caffeine consumption during gestation and lactation has no measurable consequences on the fetus and newborn infant. According to our analyses, baby’s history of vomiting, baby’s gender and categorical pre- pregnancy BMI combined were associated with time to the start of formula milk in the adjusted model. Boys were less likely to receive formula milk than girls, which is inconsistent with prior research in which male infant was shown to be more likely to be given formula milk [32]. The reason is not clear. However, this study was conducted in West Africa. Cultural difference might play a role. Our data indicates that there is a negative relationship between baby’s occurrence of vomiting and start formula milk. Conversely, early evidence showed that infants who suffered illnesses, such as diarrhea and acute respiratory infection, were significantly more likely to be introduced to formula milk [32]. It is possible that babies who vomited were taken good care and still received breastfeeding or this result was detected by chance due to the small sample size. No significant association was found for pre-pregnancy BMI, however, categorical BMI met the criteria to be included in the multivariate model. Obese and overweight women were more likely to start giving formula milk than those with normal weight. Based on the data from this study, mother’s exposure to secondhand smoking, baby’s history of diarrhea, pre-pregnancy BMI were together associated with early introduction of baby cereal. As expected, the heavier the mothers were, the more likely were they to introduce baby cereal earlier. Similar finding has been reported in previous study [33], where obese mothers were 1.4 (95% CL: 1.1, 1.9) times more likely to add cereal to infant formula than their normal BMI counterparts. In agreement with previous study [26] we found that there was a higher 24 likelihood for introduction of baby cereal among mothers who were exposed to secondhand smoking during the pregnancy. Infants with at least one occurrence of diarrhea appeared to be more likely to receive baby cereal earlier, which is also in line with other literature [28]. In order to further explore the relationship between early introduction of complementary foods and infant/maternal characteristics, time of starting having baby cereal or pureed foods were dichotomized into either ‘equal or less than 24 weeks’ or ‘more than 24 weeks’ where before 24 weeks of age was considered as early introduction. Consistent results were shown for dichotomous cereal outcome, which further strengthened the findings. The use of different statistical analysis approaches could provide a richer understanding of factors affecting infant complementary feeding. However, no meaningful association was found for timing of pureed foods introduction not only in Cox proportional hazard models but in exact logistic regression models. There are several limitations in the present study. First of all, the sample size in our analyses is small which is also reflected in the wide confidence limits among the hazard ratios and odds ratios reported. Limitation of the small cohort may have prevented us from detecting potential significant results. Second, only 43 of 201 subjects were analyzed in this article. Additional analysis on the 43 participated mother-infant pairs and the rest of the 201 pairs recruited at the beginning of the study showed significant difference between them when looking at pre- pregnancy BMI (p-value=0.027) and maternal lifetime smoking status (p-value=0.033). Mothers who did not participate were of lower pre-pregnancy BMI and fewer of them had ever smoked during their lifetime. This suggested that the results may not be representative of the population from which it was drawn. And the results do not necessarily reflect the practices of women from hospitals other than LAC-USC Medical Center. Furthermore, the information regarding infant’s 25 diet behaviors and health condition was obtained based on a self-reported recall of the mother, which could cause recall bias. What’s more, we were not able to evaluate exclusive breastfeeding as it was not explicitly asked on the questionnaire. Lastly, the exact time that any health related condition happened was not clear. So that we could not make firm conclusion about whether the occurrence of disease affected time to the start of foods or vice versa. In conclusion, in our sample of 43 mothers, 91% of whom are Hispanic White, we observed that baby’s history of diarrhea and mother’s exposure to secondhand smoking were associated with early discontinuation of ABF as well as early introduction of baby cereal. Babies with a history of vomiting, baby boys and babies’ mothers with normal BMI were less likely to receive formula milk early. In addition, the present study also pointed out that drinking caffeinated beverages during pregnancy might negatively affect the duration of ABF. Maternal pre- pregnancy BMI was positively associated with early introduction of baby cereal. Although power in this study was limited, the findings in many cases align with current literature and may be useful in guiding additional investigations specifically in the Hispanic population. More research is needed in a larger cohort to replicate these results and clarify the factors influencing infant feeding behaviors in a low income, Hispanic population. 26 REFERENCES 1. Fewtrell MS, Morgan JB, Duggan C, et al. Optimal duration of exclusive breastfeeding: what is the evidence to support current recommendations. Am J Clin Nutr. 2007; 85:635S-8. 2. Marques RFSV , Lopez FA, Braga JAP. Growth of infants fed exclusively breast milk for the first 6 months of life. Rev Bol Ped. 2006; 45 (1): 4653. 3. Gartner LM, Morton J, Lawrence RA, et al. Breastfeeding and the use of human milk. Pediatrics. 2005; 115: 496-506. 4. Grummer-Strawn LM, Scanlon KS, Fein SB. Infant feeding and feeding transitions during the first year of life. Pediatrics. 2008; 122: 36–42. 5. Victora CG, et al. Breastfeeding in the 21st century: epidemiology, mechanisms, and lifelong effect. The Lancet. 2016; 387: 475–90. 6. Ferreira MC, Franca JL, Franca EL, et al. Breastfeeding and its relationship with reduction of breast cancer: a review. Asian Pac J Cancer Prev. 2012; 13: 5327-32. 7. Li D-P, Du C, Zhang Z-M, et al. Breastfeeding and ovarian cancer risk: a systematic review and meta-analysis of 40 epidemiological studies. Asian Pac J Cancer Prev. 2014;15(12):4829–37. 8. Schwarz EB, Ray RM, Stuebe AM, et al. Duration of lactation and risk factors for maternal cardiovascular disease. Obstet Gynecol. 2009; 113: 974-82. 9. Shumey A, Demissie M, Berhane Y . Timely initiation of complementary feeding and associated factors among children aged 6 to 12 months in Northern Ethiopia: an institution-based cross-sectional study. BMC Public Health. 2013; 13: 1050. 10. Indicators for Assessing Infant and Young Child Feeding Practices: Part 1, Definitions. World Health Organization: Geneva, Switzerland, 2008; 1–20. 11. Kramer MS, Kakuma R. Optimal Duration of Exclusive Breastfeeding. Cochrane Database Syst. Rev. 2009; 1: 3517. 12. Symon B, Bammann M. Feeding in the first year of life—Emerging benefits of introducing complementary solids from 4 months. Aust. Fam. Phys. 2012; 41: 226–229. 13. Tarini BA, Carroll AE, Sox CM, et al. Systematic review of the relationship between early introduction of solid foods to infants and the development of allergic disease. Arch Pediatr Adolesc Med. 2006; 160(5):502–7. 14. Infant and young child nutrition. Global strategy on infant and young child feeding. World Health Organization. 2002. 15. Breastfeeding Report Card—United States. Centers for Disease Control and Prevention. 2013. 16. Clayton HB, Li R, Perrine CG, et al. Prevalence and Reasons for Introducing Infants Early to Solid Foods: Variations by Milk Feeding Type. Pediatrics. 2013; 131(4): 1108– 14. 17. Fiocchi A, Assa’ad A, Bahna S; Adverse Reactions to Foods Committee; American College of Allergy, Asthma and Immunology. Food allergy and the introduction of solid 27 foods to infants: a consensus document. Ann Allergy Asthma Immunol 2006; 97(1):10-20. 18. Poole JA, Barriga K, Leung DY , et al. Timing of initial exposure to cereal grains and the risk of wheat allergy. Pediatrics. 2006; 117(6): 2175-82. 19. Snijders BE, Thijs C, van Ree R, et al. Age at first introduction of cow milk products and other food products in relation to infant atopic manifestations in the first 2 years of life: the KOALA Birth Cohort Study. Pediatrics. 2008; 122(1): 115-22. 20. Butte, N F, Lopez-Alarcon M G, Garza C. Nutrient adequacy of exclusive breastfeeding for the term infant during the first six months of life. World Health Organization. 2002. 21. Turcksin R, Bel S, Galjaard S, et al. Maternal obesity and breastfeeding intention, initiation, intensity and duration: a systematic review. Matern Child Nutr. 2014; 10(2): 166–83. 22. Mulready-Ward C, Sackoff J. Outcomes and factors associated with breastfeeding for <8 weeks among preterm infants: findings from 6 states and NYC, 2004-2007. Maternal Child Health J. 2013; 17(9): 1648–57. 23. Thulier D, Mercer J. Variables associated with breastfeeding duration. J Obstet Gynecol Neonatal Nurs. 2009; 38: 259-68. 24. Liu P, Qiao L, Xu F, et al. Factors associated with breastfeeding duration: a 30-month cohort study in northwest China. J Hum Lact. 2013; 29(2): 253–9. 25. Nassar MF, et al. Breastfeeding practice in Kuwait: determinants of success and reasons for failure. Eastern Mediterranean Health Journal. 2009; 20 (7), 409–15. 26. Scott JA, Binns CW, Graham KI, et al. Predictors of the early introduction of solid foods in infants: results of a cohort study. BMC Pediatr. 2009; 9: 60. 27. Kavlashvili N, Kherkheulidze M, Kandelaki E, et al. Infants’ complementary feeding and factors influencing its timing. Georgian Med News. 2014; 234: 112–6. 28. Forsyth JS, et al. Relation between early introduction of solid food to infants and their weight and illnesses during the first two years of life. BMJ (Clinical research ed.). 1993; 306 (6892): 1572–1576. 29. Laura E H, Stephanie A L, Kathleen M R. Associations of maternal obesity and psychosocial factors with breastfeeding intention, initiation, and duration. Am J Clin Nutr. 2014; 99: 524–34. 30. Kronborg H, Vaeth M, Rasmussen KM. Obesity and early cessation of breastfeeding in Denmark. Eur J Public Health. 2013; 23: 316–22. 31. Leonard SA, Labiner-Wolfe J, Geraghty SR, et al. Associations between high pre- pregnancy body mass index, breast-milk expression, and breast-milk production and feeding. Am J Clin Nutr. 2011; 93: 556–63. 32. Issaka, AI, et al. Factors associated with early introduction of formula and/or solid, semi- solid or soft foods in seven Francophone West African countries. Nutrients. 2015; 7(2): 948–969. 33. Kitsantas P, Gallo S, Palla H, et al. Nature and nurture in the development of childhood obesity: early infant feeding practices of overweight/obese mothers differ compared to 28 mothers of normal body mass index. J Matern Fetal Neonatal Med. 2016; 29 (2): 290–3. 34. Nehlig A, Debry G. Consequences on the newborn of chronic maternal consumption of coffee during gestation and lactation: a review. J Am Coll Nutr. 1994 Feb; 13(1): 6-21. 35. Rebhan B, Kohlhuber M, Schwegler U, et al. Smoking, alcohol and caffeine consumption of mothers before, during and after pregnancy: results of the study “breast-feeding habits in Bavaria”. Gesundheitswesen. 2009 Jul; 71(7): 391–8. 36. Bahadori B, Riediger ND, Farrell SM, et al. Hypothesis: smoking decreases breast feeding duration by suppressing prolactin secretion. Med Hypotheses. 2013 Oct; 81(4): 582–6.

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Prenatal environmental exposures and fetal growth in the MADRES cohort

PDF

Effect of biomass fuel exposure on infant respiratory health outcomes in Bangladesh

PDF

Native American ancestry among Hispanic Whites is associated with higher risk of childhood obesity: a longitudinal analysis of Children’s Health Study data

PDF

Disparities in exposure to traffic-related pollution sources by self-identified and ancestral Hispanic descent in participants of the USC Children’s Health Study

PDF

Surgical aortic arch intervention at the time of extended ascending aortic replacement is associated with increased mortality

PDF

The environmental and genetic determinants of cleft lip and palate in the global setting

PDF

Associations of cumulative pollution burden and environmental health vulnerabilities with gestational weight gain in a cohort of predominantly low-income Hispanic women

PDF

sFLT-1 gene polymorphisms and risk of severe-spectrum hypertensive disorders of pregnancy

PDF

Determinants of menarche discordance in fraternal and identical twins

PDF

Study of maternal free thyroxine and thyroid-stimulating hormone's relationship with infant birthweight

PDF

Predictive factors of breast cancer survival: a population-based study

PDF

Pregnancy in the time of COVID-19: effects on perinatal mental health, birth, and infant development

PDF

Effectiveness of individual- and household-level protective actions in reducing symptoms associated with hydrogen sulfide chronic low-level exposure

PDF

Association of single nucleotide polymorphisms in GCK, GCKR and PNPLA3 with type 2 diabetes related quantitative traits in Mexican-American population

PDF

Supporting a high value maternity system of care: prioritizing resilience of and relationships with mothers to improve maternal and child health

PDF

Genetic epidemiological approaches in the study of risk factors for hematologic malignancies

PDF

The association of prediagnostic metformin use with prostate cancer in the multiethnic cohort study

Asset Metadata

Creator Zhang, Huihui (author)

Core Title Association of maternal and environmental factors with infant feeding behaviors in a birth cohort study

School Keck School of Medicine

Degree Master of Science

Degree Program Applied Biostatistics and Epidemiology

Publication Date 07/22/2016

Defense Date 06/24/2016

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag breastfeeding duration,infant feeding,maternal factors,OAI-PMH Harvest,smoking behaviors

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Breton, Carrie (committee chair), Dunton, Genevieve (committee member), Eckel, Sandrah (committee member), McConnell, Rob (committee member)

Creator Email huihuizh@usc.edu,zhanghuihui.tju@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-272546

Unique identifier UC11280527

Identifier etd-ZhangHuihu-4580.pdf (filename),usctheses-c40-272546 (legacy record id)

Legacy Identifier etd-ZhangHuihu-4580.pdf

Dmrecord 272546

Document Type Thesis

Format application/pdf (imt)

Rights Zhang, Huihui

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA