Abstract
Objective
To describe weight-for-length (WFL) trajectories in the children (birth-12 months) of mothers with and without eating disorders.
Methods
This study is based on the Norwegian Mother and Child Cohort Study (MoBa) conducted by the Norwegian Institute of Public Health. We categorized women (N=57,185) based on diagnosis prior to and during pregnancy: anorexia nervosa (AN), bulimia nervosa (BN), eating disorder not otherwise specified-purging subtype (EDNOS-P), binge eating disorder (BED), or no eating disorder (no-ED). The primary analysis included a shape invariant model fitted with non-linear mixed effects to compare growth rates across eating disorder subtypes.
Results
The children of mothers reporting any eating disorder had a lower WFL growth rate from birth--12 months than the children of mothers without eating disorders, even after adjusting for relative birth weight and some confounders known to affect growth.
Discussion
In this cohort, child WFL was related to maternal eating disorder status before and/or during pregnancy. These differences in growth trajectories warrant further study of long-term health outcomes and, if replicated, tailoring counseling to mothers with eating disorders during pregnancy.
Growth trajectories from birth to early childhood relate to chronic disease risk later in life.(1-7) Rapid weight gain during infancy or the first years of life is associated with the development of overweight in childhood and adulthood and with chronic health problems including cardiovascular disease and insulin resistance.(1-4; 6; 7) Maternal health and parenting behaviors predict infant and early childhood growth and weight trajectories, and maternal smoking, drinking before pregnancy, and dieting during pregnancy all predict rapid infant weight gain.(8) In young children, maternal emotional eating and overweight are associated with child emotional eating and overweight, (9) but restrictive maternal feeding styles also predict child overeating in the absence of hunger and child obesity.(10)
Two large prospective population-based samples suggest that children of mothers with eating disorders may be at particular risk for altered growth trajectories. Based on data from the Norwegian Mother and Child Cohort Study (MoBa) conducted by the Norwegian Institute of Public Health, mothers with bulimia nervosa (BN) and binge eating disorder (BED) reported more restrictive feeding with their 36-month-old children than women with no eating disorders.(11) Maternal eating disorder attitudes and behaviors including greater drive for thinness and bulimic symptoms are also associated with more restrictive feeding practices with young children.(12)
In another large prospective study, the Avon Longitudinal Study of Parents and Children (ALSPAC), women with anorexia nervosa (AN) histories reported that their one and six-month-old infants had more feeding difficulties including slow eating and taking in only small quantities at each feeding. However, the growth trajectories of children of mothers with eating disorders did not differ from children of mothers with no eating disorder histories in the first year of life. In contrast, mothers with BN histories reported no infant feeding difficulties, except refusal of solids, but by 9 months of age, their infants had higher WFL percentiles and faster growth rates than children in the referent group.(13; 14) However, to our knowledge, no studies have investigated longitudinal weight trajectories from birth to 12 months of age in infants of mothers with eating disorders or have adjusted for maternal breastfeeding status, a known predictor of childhood WFL and growth rates.(15)
An explication of early growth trajectories could help guide child feeding practices in mothers with eating disorders and prevent early unhealthy weight trajectories in children at increased risk for feeding difficulties and later eating psychopathology.(16; 17) Thus, we analyzed data from MoBa to describe the WFL trajectories over the first 12 months in the children of mothers with and without eating disorders. Based on the literature, we hypothesized that weight trajectories in the children of mothers with eating disorders would vary by eating disorder subtype. We hypothesized: a) WFL change over time for the children of women with AN or purging without binge eating (eating disorders not otherwise specified; EDNOS-P) would not differ from children of mothers in the referent group; b) WFL change over time for the children of mothers with BN and BED would be greater than children of mothers in the referent group.
METHODS
The Study Sample
The sample is based on women who participated in MoBa conducted by the Norwegian Institute of Public Health.(18) Participants were recruited from all over Norway from 1999-2008, and 38.5% of invited women consented to participate. The cohort now includes 108,593 children, 97,540 mothers (with one or more children) and 71,533 fathers. Follow-up is conducted by questionnaires at regular intervals and by linkage to national health registries.
Participants were recruited through a postal invitation in connection with a routine ultrasound examination sent to all pregnant women in Norway at approximately 15 weeks of gestation (www.fhi.no/morogbarn). This study is based on version 7 of the quality-assured data files released for research in January 2013, and we used information from Questionnaire 1 (~17 weeks gestation), Questionnaire 4 (6 months postpartum), and Questionnaire 5 (18 months postpartum). Informed consent was obtained from each participant and approved by The Regional Committee for Medical Research Ethics, the Norwegian Data Inspectorate, and the University of North Carolina Institutional Review Board.
All pregnancies included in the analyses met the following criteria: (1) non-missing pregnancy identification values, sex of the infant, infant weight and height measurements at any time point, and eating disorder classification, (2) maternal weight between 30 and 300 kg (66 to 660 pounds) before and during pregnancy, (3) maternal height above 100 cm (39.4 inches), (4) returned questionnaires to be completed during pregnancy before the infant's birth, (5) first pregnancy occurring during the study period (N.B. This did not have to be the mother's first birth. If two infants from the same mother were enrolled in MoBa during the study period, only the first birth was included in these analyses), (6) resulting in a live, singleton birth, and (7) resulting in a birth weight of infants >1500 g or 3 pounds and 4.9 ounces) since infants with low birth weight are likely to be premature or growth restricted with different resulting growth patterns. Additionally, we excluded individuals with missing data on eating disorder subtype; missing weight or height at any time point; weight, length, or WFL percentile flagged as an outlier; and with fewer than four available time points. Figure 1 demonstrates how we reached the analytic sample of 57,185 mothers.
Figure 1.
Flow chart for sample size
Child weight and length determination
In Norway, the child's “health card” is a running record of the child's measurements and immunizations that is completed at each doctor's visit. On the 6-month questionnaire (Questionnaire 4: http://www.fhi.no/dokumenter/9ecca1c459.pdf), mothers reported their child's weight in centimeters and length in grams at 6 weeks, 3 months, and 6 months of age using the “health card” measurements. On the 18-month questionnaire (Questionnaire 5: http://www.fhi.no/dokumenter/2640dd4bcc.pdf), mothers again reported their child's weight in centimeters and length in grams at 8 months and 12 months of age using the “health card” measurements. Normative infant WFL values were defined by World Health Organization (WHO) standards.(19) We used the definitions from WHO to determine weight, length, and WFL percentiles. We also used the following criteria from the WHO to omit values that would be biologically implausible(20): 1) a z-score for weight less than −6 or greater than 5 using WHO percentiles, 2) a z-score for length less than −6 and greater than 6 using WHO percentiles, 3)any weight-for-length z-score less than −5 or greater than 5 using WHO percentiles. The omission of these values corresponds to a less than 1% decrease in person-time observations for the sample.
Eating disorder categorization
Diagnostic algorithms and hierarchies were constructed from the questionnaire items to define the presence of eating disorders in the six months before pregnancy and/or during pregnancy and have been described elsewhere.(21; 22) Our final categories included: broadly defined AN, defined as meeting DSM IV criteria for AN (excluding amenorrhea) prior to pregnancy; broadly defined BN, endorsing at least weekly frequency of binge eating and either purging (vomiting, laxatives) or non-purging (exercise, fasting) compensatory behaviors; broadly defined BED, at least weekly frequency of binge eating in the absence of compensatory behaviors; and EDNOS-P, purging at least weekly in the absence of binge eating. Binge eating questions included eating an unusually large amount of food and loss of control. Purging was distinguished from nausea and vomiting of pregnancy.
As the symptom picture for many women changed in the interval before pregnancy and during pregnancy, our diagnostic hierarchy was: AN, BN, EDNOS-P, BED, and no-ED. All individuals who met AN criteria before pregnancy were categorized as AN regardless of presentation during pregnancy. Those who met BN criteria either before or during pregnancy and who did not meet AN criteria before pregnancy were categorized as BN. If not classified as AN or BN, those who met criteria for EDNOS-P before or during pregnancy and did not endorse binge eating at either time were categorized as EDNOS-P. Similarly, individuals who endorsed BED and did not endorse purging during or before pregnancy were included in the BED group. Group assignment was only made when all responses were available to ensure accurate classification.
Breastfeeding status
Breastfeeding was based on maternal report at five months. Predominantly breastfeeding was defined as breastfeeding without any supplements of formula milk or solid food. Although the current WHO recommendation is for exclusive breastfeeding (without any drinks, including water etc.), we applied the less restrictive category of predominant breastfeeding up to 5 months postpartum. If other feeding types had been used (either solid foods or formulas) then they were categorized as ‘partially breastfeeding’, and a third category was ‘not breastfeeding’ if the child was not receiving any breast milk.
Statistical methods
A cross sectional component of the analyses included weight-for-length z-scores as an outcome in separate models at birth and 12 months. A simple linear regression was fitted to the data to obtain unadjusted and adjusted coefficients indicating differences in mean z-scores for each of the eating disorder subtypes compared to the referent group. Confounders in the adjusted estimates included sex of child, maternal education, income index, parity, breastfeeding status at 5 months (predominant, partial or not breastfeeding, as defined above), smoking during pregnancy and gestational age. All p-values for this group of tests were adjusted using the Benjamini-Hochberg FDR method.(23)
To obtain a comparison of growth rates across eating disorder subtypes we used an infant growth model described by Beath.(24) Weight-for-length is modeled as a function of time consisting of a shape invariant model (SIM) with a natural spline fitted through a non-linear mixed effects model. This particular model includes a common shape for the growth curve for each eating disorder subtype accompanied by person-level shifts on the vertical and horizontal axes as well as changes in the scale of the growth rate.(25) Parameter estimates indicate both relative differences in the outcome at birth and group differences from the referent in outcome for a one-unit change in time – the estimated rate of growth. Model advantages include better fit and interpretability than linear regression models. The model with all three random effects resulted in a 59% decrease in the residual standard deviation. Subsequent to fitting the velocity curve, the acceleration (change in velocity over time) was plotted with differentiation of the velocity curve and interpreted via visual inspection. Confounders for the relative birth WFL included sex of child, gestational age at birth, income and smoking status of mother during pregnancy. The relative growth estimate also included breastfeeding status at five months and excluded smoking status during pregnancy (for model fit purposes). We used the R language (26) for SIM model fitting with the nlme package (27) and SAS/STAT software, version 9.2 of the SAS system for Windows (Cary, NC) to handle data and other analyses.
RESULTS
Demographics
Frequencies and eating disorder subtype differences for parity, smoking during pregnancy, education, economic status, large and small for gestational age, and breastfeeding status are reported in Table 1. Mothers with AN were younger, had lower incomes, were less likely to breastfeed, and were more likely to smoke during pregnancy than mothers with no-ED. The majority, 84.9% (N=48,399), of mothers breastfed their infants at 5 months, with 14.0% (N=7,974) breastfeeding predominantly. Bulik et al. (2009) have reported these demographic differences previously.(22)
Table 1.
Population characteristics n (percent) (unless otherwise indicated) overall and by eating disorder subtype
| Overall | Anorexia Nervosa | Bulimia Nervosa | EDNOS-P | Binge eating disorder | None | p-valuea | |
|---|---|---|---|---|---|---|---|
| Total N | 57,185 | 45 | 531 | 54 | 2,809 | 53,746 | |
| Age of mother, mean (SD) | 30.1 (4.5) | 27.0 (4.6) | 29.7 (4.6) | 28.8 (5.1) | 30.2 (4.6) | 30.1 (4.5) | <.01 |
| BMI of mother, mean (SD) | 24.0 (4.2) | 18.1 (0.6) | 24.4 (4.4) | 24.1 (4.7) | 25.9 (5.2) | 23.9 (4.1) | <.01 |
| Gestational age of child at birth, mean (SD) | 280.0 (11.1) | 280.7 (9.4) | 279.2 (10.6) | 280.1 (10.2) | 279.9 (11.3) | 280.0 (11.1) | 0.24 |
| Length of child at birth (cm), mean (SD) | 50.4 (2.3) | 50.3 (1.9) | 50.2 (2.4) | 50.1 (2.1) | 50.6 (2.3) | 50.4 (2.3) | <.01 |
| Weight of child at birth (g), mean (SD) | 3602.9 (523.2) | 3531.8 (482.6) | 3563.5 (527.1) | 3563.3 (494.0) | 3671.9 (533.5) | 3599.7 (522.5) | <.01 |
| Total number of live births | |||||||
| 0 | 32,607 (57.0) | 31 (68.9) | 306 (57.6) | 32 (59.3) | 1,446 (51.5) | 30,792 (57.3) | <0.01 |
| 1 | 16,235 (28.4) | 12 (26.7) | 139 (26.2) | 15 (27.8) | 890 (31.7) | 15,179 (28.2) | |
| 2+ | 8,343 (14.6) | 2 (4.4) | 86 (16.2) | 7 (13.0) | 473 (16.8) | 7,775 (14.5) | |
| Combined income | |||||||
| 0-$33or 0-200,000NOK | 4,279 (8.0) | 9 (20.5) | 61 (12.4) | 12 (26.1) | 248 (9.5) | 3,949 (7.9) | <0.01 |
| >$33or >200,000NOK | 26,339 (49.3) | 20 (45.5) | 225 (45.8) | 18 (39.1) | 1,413 (54.2) | 24,663 (49.1) | |
| >$82 or >500,000NOK | 14,325 (26.8) | 9 (20.5) | 125 (25.5) | 14 (30.4) | 648 (24.8) | 13,529 (26.9) | |
| >$114 or >700,000NOK | 8,523 (15.9) | 6 (13.6) | 80 (16.3) | 2 (4.3) | 300 (11.5) | 8,135 (16.2) | |
| Smoking during pregnancy | |||||||
| Yes | 4,368 (7.7) | 14 (31.1) | 62 (11.7) | 11 (20.4) | 324 (11.6) | 3,957 (7.4) | <0.01 |
| Mother's education | |||||||
| <3 year high school | 3,454 (6.4) | 3 (7.3) | 49 (9.8) | 7 (13.5) | 251 (9.4) | 3,144 (6.2) | <0.01 |
| Vocational high school | 6,940 (12.8) | 9 (22.0) | 75 (15.0) | 8 (15.4) | 434 (16.2) | 6,414 (12.6) | |
| 3-year high school/junior college | 8,006 (14.7) | 9 (22.0) | 95 (19.0) | 11 (21.2) | 465 (17.4) | 7,426 (14.5) | |
| Regional tech college, 4-year university degree | 23,053 (42.4) | 10 (24.4) | 183 (36.5) | 18 (34.6) | 1,065 (39.9) | 21,777 (42.6) | |
| University, technical college, 4+ years | 12,888 (23.7) | 10 (24.4) | 99 (19.8) | 8 (15.4) | 457 (17.1) | 12,314 (24.1) | |
| Large for gestational age | |||||||
| Yes | 5,326 (9.5) | 4 (9.5) | 49 (9.4) | 1 (1.9) | 344 (12.5) | 4,928 (9.4) | <0.01 |
| Small for gestational age | |||||||
| Yes | 4,629 (8.3) | 7 (16.7) | 62 (11.9) | 6 (11.5) | 173 (6.3) | 4,381 (8.3) | <0.01 |
| Breastfeeding status | |||||||
| Predominant breastfeeding | 7,974 (14.0) | 6 (13.3) | 78 (14.7) | 5 (9.4) | 343 (12.3) | 7,542 (14.1) | <0.01 |
| Any breastfeeding, not predominant | 40,425 (70.9) | 26 (57.8) | 356 (67.3) | 31 (58.5) | 1,896 (67.8) | 38,116 (71.1) | |
| None | 8,625 (15.1) | 13 (28.9) | 95 (18.0) | 17 (32.1) | 557 (19.9) | 7,943 (14.8) |
The test statistic is a Kruskal-Wallis statistic for continuous variables and a chi-square statistic (df) for categorical variables.
Note: All p-values adjusted with Benjamini-Hochberg false discovery rate (FDR) procedure.
Cross-sectional comparison of child weight-for-length across maternal eating disorder status
Cross-sectional comparisons of WFL z-scores across the eating disorder subtypes and the no-ED referent group were done at zero and 12 months with linear regression models (data not shown). At birth, infants of mothers with AN and BN had lower mean WFL z-scores than the referent whereas the infants of mothers with EDNOS-P (difference=0.08, ns) and BED (difference=0.06, p-value<0.05) had scores that were greater than the referent. After adjusting for confounders, no significant difference remained. The children of mothers with AN continued to have WFL z-scores below the referent at 12 months (difference=-0.44, p-value<0.05), after adjusting for confounders. The children of mothers with EDNOS-P remain above the referent group at 12 months.
Child weight-for-length trajectories across maternal eating disorder status
The nonlinear regression provides estimates in Table 2 of both differences across eating disorder subtypes in WFL at birth and the change of WFL over time, termed ‘relative growth’. Figure 2 shows infant velocity (weight-for-length change over time) predicted curves by eating disorder subtype superimposed on observed points. The results from the shape invariant models indicate that children of mothers with all of the eating disorder subtypes (AN, BN, EDNOS-P and BED) had significantly lower rates of growth—change in WFL over a unit of time—between 0 and 12 months compared to the referent group of children of mothers with no-ED. Although at birth the children of mothers with BED were significantly (2%) larger than the referent group, their growth rate over the first 12 months was significantly lower than the referent (16%) after adjusting for covariates. WFL for children of mothers with BN was significantly higher than the referent group at birth after adjusting for confounders but the adjusted growth rate remained lower than the referent (10%) over the first 12 months. Children of mothers with EDNOS had no significant differences in their birth WFL, but the adjusted growth rate remained lower than the referent (16%) over this time period. Lastly, children of mothers with AN had a significantly lower unadjusted (12%) and adjusted growth rate (17%) over the first 12 months than the referent.
Table 2.
Birth weight-for-length (WFL) and weight-for-length growth at 0-12 months by eating disorder subtype relative to group with no eating disorders
| Unadjusted |
Adjusted* |
|||
|---|---|---|---|---|
| Eating Disorder Subtype |
Relative birth WFL | Relative growth at 0-12 months | Relative birth WFL | Relative growth rate at 0-12 months |
| AN | 0.98 (0.96,1.01) | 0.88 (0.78,0.99)* | 0.98 (0.95,1.01) | 0.83 (0.73,0.94)** |
| BN | 1.01 (1.00,1.02) | 0.91 (0.88,0.94)*** | 1.01 (1.00,1.02)** | 0.90 (0.87,0.93)*** |
| EDNOS-P | 1.01 (0.99,1.04) | 0.90 (0.81,1.00)* | 1.01 (0.98,1.04) | 0.84 (0.75,0.95)** |
| BED | 1.02 (1.02,1.03)*** | 0.85 (0.84,0.86)*** | 1.02 (1.02,1.03)*** | 0.84 (0.82,0.85)*** |
| No ED | -- | -- | -- | -- |
* Relative birth WFL adjusted for sex of child, gestational age at birth, income, and smoking status of mother during pregnancy. Relative growth adjusted for sex of child, gestational age at birth and breastfeeding status at five months.
Note:
p<0.05
p<0.01
p<0.001
Figure 2.
Weight-for-length (WFL) value versus time (velocity) for observed points and predicted lines from the shape invariant model. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Visual examination of the acceleration by eating disorder subtype (Figure 3) suggests that the growth rates across eating disorder subtypes do not remain consistently lower than the referent group of women reporting no eating disorders (no-ED). The peak in acceleration before 1.5 months is greatest for the referent group providing evidence that children of mothers with eating disorders experience lower peaks in growth rate in the first two months postpartum than the referent group. Following this differential, a shift from lower acceleration for all eating disorder subtypes to one that exceeds the referent group occurs around 4 months and remains past 8 months. This change is also reflected in the growth curve (Figure 2) in which distance between the growth curves representing groups associated with eating disorders and the referent group becomes smaller over time.
Figure 3.
Weight-for-length (WFL)/month value versus time (acceleration) estimated from the shape invariant model. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
DISCUSSION
In this study of over 50,000 pregnancies, we found notable differences in children born to mothers with eating disorders for birth WFL and rate of WFL growth. Overall, over the first 12 months of life, the children of mothers with eating disorders had significantly lower WFL growth than children of mothers with no-ED. However, closely examining acceleration of WFL growth in the first year of life presents a more nuanced picture. In their first 2 months of life, children of mothers with eating disorders experience lower acceleration in their WFL growth rate in comparison to the referent group of children of mothers with no-ED. However, after 4 months of age, they experience a significantly higher acceleration in their WFL growth rate in comparison to the referent. After adjusting for confounders known to affect growth and relative birth weight, the children of mothers with BN and BED had higher WFL at birth than the children of mothers without eating disorders but lower growth rate over their 0-12 months of life.
In our first hypothesis, we anticipated that birth WFL and WFL change over time for children of mothers with AN and EDNOS-P would not differ from children of mothers without eating disorders. Indeed children born to mothers with AN and EDNOS-P did not have significantly different birth WFL, but in contrast to our hypothesis, they did have significantly lower WFL growth than mothers without eating disorders overall. In our second hypothesis, we anticipated that WFL change over time for the children of mothers with BN and BED would be greater than for the children of mothers with no-ED. Contrary to our second hypothesis, the children of women with BN and BED did not have greater change in weight over time. In fact, children of women with BN and BED had a higher relative birth WFL, but a lower relative growth rate overall.
There could be several possible reasons for the overall slower growth witnessed across eating disorder subtypes. First, breastfeeding in Norway has a remarkably high prevalence (98% in this sample) and women with eating disorders are at high risk for difficulties breastfeeding their infant children.(13) Women with eating disorders in this sample have been shown to be significantly more likely to rate their infants’ temperament as difficult (28) and thus, their infants may be more difficult to nurse successfully.(14) Women with eating disorders also frequently report heightened self-consciousness about the physical exposure that accompanies breastfeeding(29) and also frequently report breastfeeding “by the clock,” which may hamper the lactation let-down reflex or amount of fat-containing hind milk offered in more prolonged feeding sessions.(30) Alternatively, a recurrence of eating psychopathology in response to gestational weight gain including food restriction, purging behavior, and weight loss attempts might lead to an inability to consume the additional 500-600 kcal per day needed to sustain optimal breast milk. Thus, re-emergent maternal eating disorder symptoms, self-consciousness, and difficult infant temperament, may result in suboptimal quantity and quality of breast milk, particularly breast milk's fat content.(31) (32) However, data regarding breast milk quantity and quality were not available in the present sample and we were unable to test this potential contributor to slower growth in the children of mothers with eating disorders.
Additionally, further research is needed to examine the relative acceleration in growth rate for children of mothers with eating disorders after 4 months of age. We speculate that the acceleration in WFL growth rate for children of mothers with eating disorders witnessed at this time could be a consequence of the diversification of food intake and introduction of solid foods, particularly rice and oatmeal cereals.(33) However, further research with larger samples of women with eating disorders is necessary to determine the reasons and consequences of accelerated children WFL growth at this critical 4 months transition point. Research on mothers with eating disorder histories has found that they do not differ from mothers without eating disorders on the age they introduce their infants to solid food or the frequency with which they feed their infant children. Moreover, in previous MoBa results, women with BN and BED were more likely to engage in restrictive feeding practices with their 36-month-old children.(11) We speculate that engaging in restrictive feeding practices at this later time point (36 months) may be a consequence of mothers’ concerns about the acceleration in growth rate witnessed and subsequent worries overweight or obesity in their infants and toddlers. Thus, the consequences of greater growth acceleration after 4 months of age might only become evident after our observation period. Further research is needed to fully understand the contributing factors for WFL acceleration after 4 months of age and how these factors may affect parenting and child feeding later in development in mothers with eating disorders.
Limitations
Limitations must be considered in interpreting these findings. First, it is unknown whether these results generalize to other countries. For example, the Norwegian authorities advocate the WHO recommendation that infants should be exclusively breastfed for the first 6 months, and continue to breast-feed until the child is 12 months old. How that authoritative stance affects individuals' experience is unknown. Second, although our diagnostic items have been used in previous epidemiologic studies (34) they were self-report, and some questions were retrospective. Although the large sample sizes of cohort studies are invaluable, the necessary trade-off of self-report may yield less accurate information than face-to-face diagnostic interviews.(35) The estimates of self-reported eating disorders prior to pregnancy were also lower than lifetime estimates and point prevalence estimates reported elsewhere.(36; 37) Although, impossible to compare directly, it is possible that these self-report questionnaires missed some eating disorder cases. Women met diagnostic criteria for eating disorders peripartum, but we did not assess lifetime diagnoses. An additional limitation may have resulted from the fact that, rather than focus on symptoms, which may cut across diagnoses, we focused on diagnoses themselves.
Third, although a 38.5% response rate is typical for large population-based studies, participants may differ in meaningful ways from the general population. MoBa participants differ from the general Norwegian population of pregnant women by having more education and a lower prevalence of preterm births (7.2 vs. 7.7%) and low birth-weights (4.6 vs. 5.1%).(18) Fourth, our group sizes were small, so findings must be interpreted with caution and reassessed in larger samples. Our results also might not generalize to individuals with more severe AN. Women with AN who were healthy enough to conceive and participate in the MoBa could differ from treatment-seeking samples. Finally, our study only provides information on the relationship between eating disorders and growth for the first 12 months of life. Information for later time points is available, but about only half the sample (40% to 53% from 18 months through 3 years) remained for these time points. Due to the ongoing nature of the cohort study we decided to omit those values for this analysis with the intent to examine these values once a substantive proportion of the data from future waves of respondents are available. Also, while we found decreased relative growth of children born to mothers with eating disorders, on a population level we cannot assess their overall health either currently or in the future as a result of their growth trajectories.
Conclusion
The fact that the overall growth of children of women with eating disorders differs significantly from children born to women without eating disorders, at least in a Norwegian cohort of mothers, suggests that counseling for these mothers, then, should focus not only on their own health but also their children's growth and nutrition. Studies of clinical samples of women with eating disorders are required to determine whether these findings hold in treatment-seeking populations or if greater severity of illness impacts child growth trajectories even more markedly. Our results underscore the importance of supporting women with eating disorders through pregnancy and infant feeding and carefully following growth trajectories through early childhood.
Acknowledgements
This study was supported by R01HD047186 (Bulik: PI). The Norwegian Mother and Child Cohort Study is supported by the Norwegian Ministry of Health, NIH/NIEHS (grant no N01-ES-85433), NIH/NINDS (grant no.1 U01 NS 047537-01), and the Norwegian Research Council/FUGE (grant no. 151918/S10). Drs. Perrin, Skinner, and Zerwas were supported by a CTSA Grant (UL1RR025747) and Dr. Perrin is supported by an R01from the NIH/NICHD (R01 HD049794). Drs. Skinner and Zerwas are supported by an NIH Building Interdisciplinary Careers in Women's Health award (K12-HD01441). Dr. Reba-Harrelson was supported by F31MH083312. We are grateful to all the participating families in Norway who take part in this ongoing cohort study.
Footnotes
Disclosure of Conflicts
The authors have no conflicts of interest to disclose.
References
- 1.Reilly JJ, Armstrong J, Dorosty ARea. Early life risk factors for obesity in childhood: cohort study. BMJ. 2005;330:1357. doi: 10.1136/bmj.38470.670903.E0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Singhal A, Lucas A. Early origins of cardiovascular disease: is there a unifying hypothesis? Lancet. 2004;363:1642–1645. doi: 10.1016/S0140-6736(04)16210-7. [DOI] [PubMed] [Google Scholar]
- 3.Adair LS. Size at birth and growth trajectories to young adulthood. Am J Hum Biol. 2007;19:327–337. doi: 10.1002/ajhb.20587. [DOI] [PubMed] [Google Scholar]
- 4.Taveras EM, Rifas-Shiman SL, Belfort MB, Kleinman KP, Oken E, Gillman MW. Weight status in the first 6 months of life and obesity at 3 years of age. Pediatrics. 2009;123:1177–1183. doi: 10.1542/peds.2008-1149. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Gillman MW, Rifas-Shiman SL, Kleinman K, Oken E, Rich-Edwards JW, Taveras EM. Developmental origins of childhood overweight: potential public health impact. Obesity (Silver Spring) 2008;16:1651–1656. doi: 10.1038/oby.2008.260. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Baird J, Fisher D, Lucas P, Kleijnen J, Roberts H, Law C. Being big or growing fast: systematic review of size and growth in infancy and later obesity. BMJ. 2005;331:929–931. doi: 10.1136/bmj.38586.411273.E0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Dennison BA, Edmunds LS, Stratton HH, Pruzek RM. Rapid infant weight gain predicts childhood over-weight. Obesity. 2006;14:491–499. doi: 10.1038/oby.2006.64. [DOI] [PubMed] [Google Scholar]
- 8.Oyama M, Nakamura K, Tsuchiya Y, Yamamoto M. Unhealthy maternal lifestyle leads to rapid infant weight gain: prevention of future chronic diseases. Tohoku J Exp Med. 2009;217:67–72. doi: 10.1620/tjem.217.67. [DOI] [PubMed] [Google Scholar]
- 9.Jahnke DL, Warschburger PA. Familial transmission of eating behaviors in preschool-aged children. Obesity. 2008;16:1821–1825. doi: 10.1038/oby.2008.255. [DOI] [PubMed] [Google Scholar]
- 10.Birch LL, Fisher JO, Davison KK. Learning to overeat: maternal use of restrictive feeding practices promotes girls' eating in the absence of hunger. Am J Clin Nutr. 2003;78:215–220. doi: 10.1093/ajcn/78.2.215. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Reba-Harrelson L, Von Holle A, Hamer RM, Torgersen L, Reichborn-Kjennerud T, Bulik CM. Patterns of maternal feeding and child eating associated with eating disorders in the Norwegian Mother and Child Cohort Study (MoBa). Eat Behav. 2010;11:54–61. doi: 10.1016/j.eatbeh.2009.09.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Farrow CV, Blissett J. Do obsessive compulsive symptoms mediate the relationship between maternal eating psychopathology and restrictive feeding practices? Int J Eat Disord. 2009;42:76–80. doi: 10.1002/eat.20579. [DOI] [PubMed] [Google Scholar]
- 13.Micali N, Simonoff E, Treasure J. Infant feeding and weight in the first year of life in babies of women with eating disorders. J Pediatr. 2009;154:55–60. e51. doi: 10.1016/j.jpeds.2008.07.003. [DOI] [PubMed] [Google Scholar]
- 14.Micali N, Simonoff E, Stahl D, Treasure J. Maternal eating disorders and infant feeding difficulties: maternal and child mediators in a longitudinal general population study. J Child Psychol Psychiatry. 2010 doi: 10.1111/j.1469-7610.2010.02341.x. [DOI] [PubMed] [Google Scholar]
- 15.Arenz S, Ruckerl R, Koletzko B, von Kries R. Breast-feeding and childhood obesity--a systematic review. International Journal of Obesity and Related Metabolic Disorders. 2004;28:1247–1256. doi: 10.1038/sj.ijo.0802758. [DOI] [PubMed] [Google Scholar]
- 16.Strober M, Freeman R, Lampert C, Diamond J, Kaye W. Controlled family study of anorexia nervosa and bulimia nervosa: evidence of shared liability and transmission of partial syndromes. Am J Psychiatry. 2000;157:393–401. doi: 10.1176/appi.ajp.157.3.393. [DOI] [PubMed] [Google Scholar]
- 17.Bulik CM, Sullivan PF, Wade TD, Kendler KS. Twin studies of eating disorders: a review. Int J Eat Disord. 2000;27:1–20. doi: 10.1002/(sici)1098-108x(200001)27:1<1::aid-eat1>3.0.co;2-q. [DOI] [PubMed] [Google Scholar]
- 18.Magnus P, Irgens LM, Haug K, Nystad W, Skjaerven R, Stoltenberg C. Cohort profile: the Norwegian Mother and Child Cohort Study (MoBa). Int J Epidemiol. 2006;35:1146–1150. doi: 10.1093/ije/dyl170. [DOI] [PubMed] [Google Scholar]
- 19.WHO Multicentre Growth Reference Study Group . WHO Child Growth Standards: Growth velocity based on weight, length and head circumference: Methods and development. World Health Organization; Geneva: 2009. [Google Scholar]
- 20.Organization WH. WHO Child Growth Standards: SAS igrowup Package. 2011 [Google Scholar]
- 21.Bulik CM, Von Holle A, Hamer R, Knoph Berg C, Torgersen L, Magnus P, et al. Patterns of remission, continuation and incidence of broadly defined eating disorders during early pregnancy in the Norwegian Mother and Child Cohort Study (MoBa). Psychol Med. 2007;37:1109–1118. doi: 10.1017/S0033291707000724. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Bulik CM, Von Holle A, Siega-Riz AM, Torgersen L, Lie KK, Hamer RM, et al. Birth outcomes in women with eating disorders in the Norwegian Mother and Child cohort study (MoBa). Int J Eat Disord. 2009;42:9–18. doi: 10.1002/eat.20578. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Benjamini Y, Hochberg Y. Controlling the False Discovery Rate - a Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society Series B-Methodological. 1995;57:289–300. [Google Scholar]
- 24.Beath KJ. Infant growth modelling using a shape invariant model with random effects. Statistics in Medicine. 2007;26:2547–2564. doi: 10.1002/sim.2718. [DOI] [PubMed] [Google Scholar]
- 25.Cole TJ, Donaldson MD, Ben-Shlomo Y. SITAR--a useful instrument for growth curve analysis. International Journal of Epidemiology. 2010;39:1558–1566. doi: 10.1093/ije/dyq115. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.R Development Core Team . R: A language and environment for statistical computing. Vienna, Austria: 2012. [Google Scholar]
- 27.Pinheiro J, Bates D, DebRoy S, Sarkar D, R Development Core Team nlme: Linear and Nonlinear Mixed Effects Models, R package version 3.1-103. 2012.
- 28.Zerwas S, Von Holle A, Torgersen L, Reichborn-Kjennerud T, Stoltenberg C, Bulik CM. Maternal Eating Disorders and Infant Temperament: Findings from the Norwegian Mother and Child Cohort Study (MoBa) Under review. [DOI] [PMC free article] [PubMed]
- 29.Waugh E, Bulik CM. Offspring of women with eating disorders. Int J Eat Disord. 1999;25:123–133. doi: 10.1002/(sici)1098-108x(199903)25:2<123::aid-eat1>3.0.co;2-b. [DOI] [PubMed] [Google Scholar]
- 30.Beresford HJ. The success of breast feeding. IPPF Med Bull. 1984;18:3–4. [PubMed] [Google Scholar]
- 31.Jelliffe DB, Jelliffe EF. The volume and composition of human milk in poorly nourished communities. A review. Am J Clin Nutr. 1978;31:492–515. doi: 10.1093/ajcn/31.3.492. [DOI] [PubMed] [Google Scholar]
- 32.Zerwas SC, Von Holle A, Perrin E, Skinner AC, Reba-Harrelson L, Stoltenberg C, et al. Patterns of Gestational and Postpartum Weight Change for Mothers with Eating Disorders in the Norwegian Mother and Child Cohort Study (MoBa) In progress.
- 33.Easter A, Naumann U, Northstone K, Schmidt U, Treasure J, Micali N. A Longitudinal Investigation of Nutrition and Dietary Patterns in Children of Mothers with Eating Disorders. J Pediatr. 2013 doi: 10.1016/j.jpeds.2012.11.092. [DOI] [PubMed] [Google Scholar]
- 34.Reichborn-Kjennerud T, Bulik CM, Sullivan PF, Tambs K, Harris JR. Psychiatric and medical symptoms in binge eating in the absence of compensatory behaviors. Obes Res. 2004;12:1445–1454. doi: 10.1038/oby.2004.181. [DOI] [PubMed] [Google Scholar]
- 35.Goldfein JA, Devlin MJ, Kamenetz C. Eating Disorder Examination-Questionnaire with and without instruction to assess binge eating in patients with binge eating disorder. Int J Eat Disord. 2005;37:107–111. doi: 10.1002/eat.20075. [DOI] [PubMed] [Google Scholar]
- 36.Favaro A, Ferrara S, Santonastaso P. The spectrum of eating disorders in young women: a prevalence study in a general population sample. Psychosom Med. 2003;65:701–708. doi: 10.1097/01.psy.0000073871.67679.d8. [DOI] [PubMed] [Google Scholar]
- 37.Hoek HW. Incidence, prevalence and mortality of anorexia nervosa and other eating disorders. Curr Opin Psychiatry. 2006;19:389–394. doi: 10.1097/01.yco.0000228759.95237.78. [DOI] [PubMed] [Google Scholar]