Research Proposal

Parental Incarceration and Child Sleep and Eating Behaviors Dylan B. Jackson, PhD1, and Michael G. Vaughn, PhD2

Objectives To examine whether parental incarceration is significantly associated with a number of sleep and eating behaviors among offspring during early childhood. Study design Data from the Fragile Families and Child Well-Being Study, an at-risk sample of parents and their offspring, were employed to test this possibility. Both maternal and paternal incarceration history were examined as predictors of whether children manifested high levels of the following 7 health behaviors: sleep problems, short sleep duration, salty snack consumption, starch consumption, sweets consumption, soda consumption, and fast food consumption. Logistic regression techniques were used to carry out the analyses. Results Both maternal and paternal incarceration significantly increased the odds of a number of risky sleep and eating behaviors during childhood. Ancillary analysis also revealed that the predicted probability of exhibiting mul- tiple risky behaviors across the sleep and eating domains was twice as large among children whose parents had both been incarcerated, relative to children whose parents had not been incarcerated. Conclusions Parental incarceration may have important implications for the sleep and eating behaviors of off- spring. Both scholars and practitioners may, therefore, want to consider the possible negative repercussions of pa- rental incarceration for the sleep and eating behaviors of children, and the potential for these high-risk health behaviors to compromise the health and well-being of children as they age. (J Pediatr 2017;185:211-7).

T he relatively large proportion of incarcerated adults in the US, who are also parents, has a profound impact on childrengrowing up in these disrupted homes. An emerging body of literature using multiple data sources has documented manyof the collateral effects of imprisonment on children and families. These include, but are not limited to, reports of aca- demic and socioemotional skills deficits,1-3 antisocial and criminal behavior, internalizing symptoms, and related mental health problems,3-6 illicit drug use,7 and food insecurity among children whose parents have been incarcerated.8,9

Although a few recent investigations have examined health outcomes among adults who have had a parent incarcerated,10-13

surprisingly few studies have systematically examined the link between parental incarceration on health outcomes during childhood.14

In particular, the effects of parental incarceration on key childhood health behaviors such as sleeping and eating patterns have received little empirical attention. This is an important oversight in the literature, as there is little scientific debate that suffi- cient sleep and an adequate diet are cornerstones for healthy development. This is perhaps especially true in the case of sleep and dietary behaviors during early childhood because of the substantial plasticity of the brain during early life stages.15 This plasticity is reflected in a body of literature that has linked sleep and dietary practices among preschool and young school- aged children (eg, 5- to 6-year-olds) to a number of other developmental risk factors. For instance, poor sleep and dietary intake during early childhood have been linked to diminished cognitive functioning,16,17 poor school performance,18,19 and conduct problems.20-22 Consensus messages regarding sleep and diet as important components of healthy development can be readily viewed not only in a number of journal articles,23,24 but also in the health guidelines provided by scientific organizations such as the American Academy of Pediatrics and the Academy of Nutrition and Dietetics.25,26 Given the relatively high rates of im- prisonment, particularly in vulnerable minority communities, examining the links between maternal and/or paternal incar- ceration and childhood sleep and eating behaviors is a worthwhile endeavor, especially because these sleep and eating behaviors are, for the most part, modifiable.

Our objective was to explore whether parental incarceration is significantly associated with child sleep and eating behaviors. Consistent with the aforementioned findings revealing increased family disruption and vulnerability in response to parental incarceration, we hypothesize that children whose parents have been incarcerated will exhibit riskier sleep and eating behaviors.

Methods

Data from the Fragile Families and Child Well-being Study (FFCWS) were used in the current study. Approval for the current study was obtained from the In- stitutional Review Board of the University of Texas at San Antonio. The FFCWS is a longitudinal study of 5000 couples and their offspring born in the US between

FFCWS Fragile Families and Child Well-Being Study

From the 1Department of Criminal Justice, The University of Texas at San Antonio, San Antonio, TX; and 2College for Public Health and Social Justice, Saint Louis University, St. Louis, MO

The authors declare no conflicts of interest.

0022-3476/$ - see front matter. © 2017 Elsevier Inc. All rights

reserved.

http://dx.doi.org10.1016/j.jpeds.2017.03.026

THE JOURNAL OF PEDIATRICS • www.jpeds.com ORIGINAL ARTICLES

211

1998 and 2000. Available data follow high-risk families from the child’s birth through late childhood. The sample was se- lected through a 3-stage process in which low-income, unwed couples were oversampled. First, a stratified random sample of 20 cities was obtained from the population of large US cities (ie, those with at least 200 000 residents). Next, a sample of 75 hospitals was taken from these 20 cities, followed by a random sample of both married and unmarried couples who agreed to participate in the study. At the first wave of data col- lection, only 25% of couples were married, and another 36% were unmarried and cohabiting. By the final wave of data used in the current study, 32% of couples were married, and 14% were unmarried and cohabiting. The data are well-suited to the current study, as they include a sufficient number of parents who have been incarcerated, as well as a number of items cov- ering the sleep and eating behaviors of the offspring during childhood.

At the fourth wave of data collection, when focal children were approximately 5 years old, parents were asked a number of questions concerning the child’s involvement in behaviors that can pose risks to their health including sleep problems and short sleep duration. To measure sleep problems, mothers were presented with the following statement concerning their child: “(He/She) has trouble getting to sleep.” Focal children whose mothers reported that the statement was sometimes or often true were assigned a value of 1 (0 for a “not true” response). To measure short sleep duration, mothers were asked, “How many hours of sleep a night does the child usually get?” Ac- cording to the National Sleep Foundation, school-aged chil- dren need between 9 and 11 hours of sleep per night for healthy functioning.27 Children who scored below this amount (≤8 hours) were in the bottom quartile of the distribution and were assigned a value of 1, whereas the remaining number of chil- dren were assigned a value of 0 (mean = 9.4 hours).

The next set of child health behaviors measured in the present study cover eating behaviors such as salty snack con- sumption, starch consumption, sweets consumption, soda con- sumption, and fast food consumption. During the in-home survey at the fourth wave of data collection, primary caregiv- ers were asked, “On a typical day, about how many servings of the following foods does (the child) eat?”: (1) salty snacks (eg, chips, pretzels, etc.); (2) starches (eg, bread, cereal, spa- ghetti, etc.); (3) sweets (eg, candy, cake, etc.); (4) soda (eg, Coke, Pepsi, etc.); (5) fast food (eg, McDonald’s, KFC, etc.).

Response options ranged from none (0) to 5 or more (5). To tap particularly high frequencies of poor eating behav- iors, scores on each of these items were dichotomized. Spe- cifically, for each of the eating behaviors, children who scored within the top quartile (ie, above the 75th percentile) were as- signed a value of 1 and all others a value of 0. Alternative coding strategies did not alter the substantive results of the study. Finally, we assessed multiple risky health behaviors across the sleep and eating domains by assigning a value of 1, if chil- dren met both of the following conditions: (1) they experi- enced short sleep durations and/or sleep problems at wave 4; and (2) they reported poor eating behaviors in 1 or more of the examined eating domains (1 in 5 children in the FFCWS

met these criteria). If they did not meet both of these crite- ria, they were assigned a value of 0.

The FFCWS include a number of parents with a history of criminal justice involvement, including incarceration. Mothers and fathers were classified as having a history of incarcera- tion if the mother, the father, or both reported that the mother/ father had spent time in jail or prison at or before the fourth wave of data collection (46% of fathers and 12% of mothers). Finally, an index of parental incarceration was also created. To create the index, children of families in which neither parent had been incarcerated by wave 4 were assigned a value of 0 (52%), children of families in which only 1 parent had been incarcerated by wave 4 were assigned a value of 1 (42%), and children of families in which both parents had been incarcer- ated by wave 4 were assigned a value of 2 (6%).

In addition to the key independent and dependent vari- ables of interest, the child’s age (in months), race (non- white = 1), and sex (male = 1) were included in the analysis as covariates as well as the following parental characteristics: maternal education, household income, maternal depres- sion, and maternal involvement. With respect to maternal edu- cation, response options ranged from no formal schooling (1) to graduate or professional school (9). A measure of quintile- based household income was also used in the present study. The FFCWS data also include an indicator of whether the child’s mother met the clinical criteria for depression at the third wave of data collection. The measure uses the 12 month Diagnos- tic and Statistical Manual for Mental Disorders diagnosis from the Composite International Diagnostic Interview–Short Form, version 1.0. The measure is a count of the number of depres- sive symptoms (eg, restlessness, sadness, loneliness, poor ap- petite, trouble concentrating, and poor sleep) ranging from 0 to 7, with subjects receiving a score of 3 or more being des- ignated as meeting the clinical criteria for depression.

Finally, a measure of maternal involvement with the child was also included in the study as well as a measure of cognitively stimulating activities at wave 2 of data collection. At this wave, mothers reported on how many days a week they played peek- a-boo with the child, sang songs, read stories, told stories, played inside games (eg, blocks), visit relatives, hug/show affection to the child, and put the child to bed. A scale reflecting the number of activities mothers engaged in each day of the week with their child was created and included in the analysis (alpha = .94).

Statistical Analyses The present analysis explored the role of parental incarcera- tion in the sleep and dietary patterns of young children. The analysis proceeded as follows. First, we calculated and exam- ined the descriptive statistics of the variables included in the analyses (eg, means, SDs). Next, we employed logistic regres- sion techniques to assess the extent to which paternal, mater- nal, and combined parental incarceration increase the odds of poor sleep and dietary patterns across the following items: sleep problems, short sleep duration, salty snack consump- tion, starch consumption, sweets consumption, soda con- sumption, and fast food consumption. Ancillary analyses were also conducted to explore the increase in the predicted prob-

THE JOURNAL OF PEDIATRICS • www.jpeds.com Volume 185

212 Jackson and Vaughn

ability of exhibiting 1 or more risks across the sleep and eating domains by degree of parental incarceration. Final sample sizes ranged from 2100 to 2388, contingent on the model being ex- amined. Despite the level of attrition from wave 1, analyses re- vealed that those who remained in the study through wave 4 did not significantly differ from those had who dropped out of the study by wave 4 on key covariates (eg, race, income, education).

Results

As shown in Table I, 46% of fathers had experienced incar- ceration by wave 4, whereas 12% of mothers had experi- enced incarceration by wave 4. Descriptive analyses also revealed that, in 52% of households, neither parent had been incar- cerated, in 42% of households only 1 parent had been incar- cerated, and in 6% of households both parents had been incarcerated. Importantly, the proportion of parents who have been incarcerated in the present sample is much higher than

the proportion found in nationally representative samples, because of the previously mentioned high-risk features of the FFCWS.4 As illustrated in Table I, the sample of children is 52% male and 69% nonwhite. At the fourth wave of data collec- tion (ie, the wave during which the sleep and eating behav- iors were reported), children were between 57 and 72 months of age (ie, 4 years, 9 months-6 years), with an average age of approximately 62 months (ie, 5 years and 2 months). The average score on maternal education, furthermore, was 4.66 (which approximates the receipt of a high school diploma) and 12% of mothers were categorized as demonstrating clinical levels of depressive symptomatology at wave 3.

In general, multivariate, logistic regression analyses re- vealed that the covariates included in the present study are indeed predictive of 1 or more of the child health behaviors examined. For instance, in models including parental incar- ceration measures, race predicted both of the sleep measures (eg, nonwhite subjects have significantly higher odds of short sleep durations; OR = 1.97). Comparatively, race was not sig- nificantly predictive of any of the eating behaviors. Maternal education, however, was significantly and negatively predic- tive of both sleep measures (eg, sleep problems; OR = .93), as well as most eating behaviors, including sweets, soda, and salty snack consumption (ORs range from .84 to .87; the results for household income were largely the same). Both maternal in- volvement and maternal depression were associated with sleep problems but were not associated with sleep duration (ma- ternal involvement was negatively associated with sleep prob- lems [OR = .51], whereas maternal depression was positively associated with sleep problems [OR = 1.27]). Maternal in- volvement was also found to be significantly and negatively as- sociated with sweets (OR = .58) and soda consumption (OR = .63). Finally, child sex (male = 1) was significantly as- sociated with fast food consumption (OR = 1.42), whereas child age was significantly associated with sleep problems (OR = .97).

The association between maternal and paternal incarcera- tion and child sleep and eating behaviors, net of covariates, are displayed in Table II. The first 2 columns pertain to sleep behaviors, whereas the final 5 columns pertain to each of the measured eating behaviors. The results reveal that paternal in- carceration was associated with all outcomes, except fast food; whereas maternal incarceration was associated with sleep du- ration and all eating outcomes, except starchy foods. Overall,

Table I. Descriptive statistics*

Variables Mean SD Range

Child sleep behaviors Sleep problems .25 .43 0-1 Short sleep duration .25 .44 0-1

Child eating behaviors Salty snack consumption .18 .38 0-1 Starch consumption .20 .40 0-1 Sweets consumption .14 .35 0-1 Soda consumption .18 .39 0-1 Fast food consumption .11 .31 0-1

Parental incarceration Paternal incarceration .46 .50 0-1 Maternal incarceration .12 .32 0-1 Incarceration index .53 .60 0-2

Covariates Age (mo) 61.85 2.84 57-72 Race (nonwhite = 1) .69 .46 0-1 Sex (male = 1) .52 .50 0-1 Maternal education 4.66 1.82 1-9 Household income 3.07 1.41 1-5 Maternal depression .12 .33 0-1 Maternal involvement .53 .28 0-1

*In the case of maternal education, 1 indicates no formal education whereas 9 indicates gradu- ate or professional school (4.66 approximates a high school diploma). Household income is coded in quintiles, with 1 representing the bottom 20% of the income distribution of the sample, and 5 representing the top 20% of the income distribution.

Table II. Does paternal and maternal incarceration influence child sleep and eating behaviors?*

Child sleep and eating behaviors

Sleep problems

Short sleep duration

Salty snack consumption

Starch consumption

Sweets consumption

Soda consumption

Fast food consumption

Parental incarceration OR (CI)

OR (CI)

OR (CI)

OR (CI)

OR (CI)

OR (CI)

OR (CI)

Paternal incarceration 1.47 (1.20-1.80)

1.28 (1.04-1.57)

1.41 (1.11-1.79)

1.41 (1.12-1.77)

1.50 (1.16-1.95)

1.31 (1.04-1.66)

1.02 (.76-1.37)

Maternal incarceration 1.16 (.81-1.66)

1.49 (1.04-2.14)

1.67 (1.14-2.46)

1.26 (.84-1.87)

2.05 (1.38-3.06)

1.80 (1.23-2.64)

2.09 (1.32-3.30)

*ORs and CIs in bold are statistically significant at P < .05 or lower. Controls for age, race, sex, maternal education, household income, maternal depression, and maternal involvement are in- cluded in each of the models.

June 2017 ORIGINAL ARTICLES

213Parental Incarceration and Child Sleep and Eating Behaviors

the results reveal mostly significant associations between ma- ternal and paternal incarceration and child behaviors that pose risks to their health. For instance, the odds of sleep problems were predicted to be 47% higher among children whose fathers have been incarcerated relative to children whose fathers have no previous history of incarceration. Maternal incarceration, however, did not significantly increase the odds of sleep prob- lems. The odds of short sleep duration, moreover, were 28% higher among children whose fathers have been incarcerated and 49% higher among children whose mothers have been in- carcerated. Similar results emerged when examining eating be- haviors. Relative to children whose fathers have no history of incarceration, children whose fathers have been incarcerated incurred a 41% increase in the odds of salty snack consump- tion, a 41% increase in the odds of starch consumption, a 50% increase in the odds of sweets consumption, and a 31% in- crease in the odds of soda consumption. The odds of fast food consumption, however, were not significantly higher among children whose fathers had been incarcerated. Maternal in- carceration was also associated with a number of risky eating behaviors. Specifically, children of mothers who have been in- carcerated incurred a 67% increase in the odds of salty snack consumption, a 105% increase in the odds of sweets con- sumption, an 80% increase in the odds of soda consump- tion, and a 109% increase in the odds of fast food consumption. Maternal incarceration did not, however, significantly in- crease the odds of starch consumption, although the OR was in the expected direction (OR = 1.54). Although findings per- taining to consumptive behaviors are consistent with prior re- search demonstrating that lower income households consume diets higher in fat, sodium, sugar, and refined carbohydrates,28

there is little research on these behaviors in families experi- encing a parent who is incarcerated. Despite the overall pattern of maternal and paternal incarceration being associated with significant increases in the odds of risky sleep and eating be- haviors, statistically significant differences between the ORs per- taining to paternal incarceration and maternal incarceration did not emerge.

The influence of the number of parents incarcerated on child sleep and eating behaviors are presented in Table III. In general, the odds of sleep and eating problems increased as the number of parents with incarceration experiences increased. Despite the general increase in the ORs as the number of parents who

are incarcerated increases, the incarceration of both parents does not appear to significantly increase the odds of risky sleep and eating behaviors relative to children who experienced the incarceration of 1 parent. Relative to children whose parents have no history of incarceration, however, the results suggest that children of parents who have both been incarcerated in- curred a 50% increase in the odds of sleep problems, a 61% increase in the odds of short sleep duration, and 67% in- crease in the odds of salty snack consumption, a 186% in- crease in the odds of sweets consumption, and an 83% increase in the odds of soda consumption. Comparatively, children with 1 parent who has been incarcerated incurred a 50% increase in the odds of sleep problems, a 29% increase in the odds of short sleep duration, and 41% increase in the odds of salty snack consumption, a 40% increase in the odds of starch consump- tion, a 40% increase in the odds of sweets consumption, and an 40% increase in the odds of soda consumption. Statisti- cally significant results remained unchanged in ancillary analy- ses accounting for household food insecurity. Moreover, additional ancillary analyses revealed that household struc- ture (eg, marital status, separation) did not significantly alter the associations between parental incarceration and the seven health outcomes.

Overall, the findings generally indicate that, net of covariates, incarceration of both mothers and fathers has the potential to increase the odds of a number of risky sleep and eating be- haviors (Figure). The Figure displays the predicted probabil- ity of exhibiting multiple risky health behaviors across the sleep and eating domains by degree of parental incarceration. Al- though the predicted probability of poor sleep and eating be- haviors was .17 among children whose parents had never been incarcerated, it was twice as high (.34) among children whose parents had both experienced incarceration.

Discussion

Incarceration of a family member is a disruptive force that can curtail healthy development and interfere with parent-child in- teractions and, thus, may be associated with inefficient and/ or disordered sleep and poorer diets.29-31 The findings of the present study are generally supportive of our hypothesis that parental incarceration would be significantly associated with these behaviors. In sum, a number of sleep and eating behav-

Table III. Does the number of parents incarcerated influence child sleep and eating behaviors?*

Child sleep and eating behaviors

Sleep problems

Short sleep duration

Salty snack consumption

Starch consumption

Sweets consumption

Soda consumption

Fast food consumption

Incarceration index OR (CI)

OR (CI)

OR (CI)

OR (CI)

OR (CI)

OR (CI)

OR (CI)

One parent incarcerated 1.50 (1.23-1.84)

1.29 (1.04-1.59)

1.41 (1.11-1.79)

1.40 (1.12-1.77)

1.40 (1.16-1.95)

1.38 (1.08-1.76)

1.15 (.85-1.57)

Both parents incarcerated 1.50 (1.01-2.29)

1.61 (1.05-2.47)

1.67 (1.14-2.46)

1.54 (.97-2.46)

2.86 (1.81-4.53)

1.83 (1.16-2.89)

1.51 (.85-2.66)

*ORs and CIs in bold are statistically significant at P < .05 or lower. The reference category is 0 (or children whose mother and father have never been incarcerated). Controls for age, race, sex, maternal education, household income, maternal depression, and maternal involvement are included in each of the models.

THE JOURNAL OF PEDIATRICS • www.jpeds.com Volume 185

214 Jackson and Vaughn

iors that can compromise child health were found to be sig- nificantly more common among children whose parent or parents experienced incarceration. Some differences were found in the effect sizes of maternal and paternal incarceration on the child health behaviors examined, with maternal incarcera- tion having a larger predicted effect on several dietary com- ponents, including sweets, soda, and fast food consumption. Even so, it should be noted that the differences in the ORs across maternal and paternal incarceration were not statistically sig- nificant. Overall, the risk of poor sleep and diet was more pro- nounced among children whose parents had both been incarcerated, although having either a mother or father who had been incarcerated also influenced a number of risky health behaviors across the sleep and dietary domains. Relative to chil- dren whose parents had not been incarcerated, having both parents incarcerated was associated with significantly greater odds of several high-risk health behaviors, most notably sweets and soda consumption, followed by short sleep duration and salty snack consumption. The predicted probability of engag- ing in poor sleep and poor eating behaviors is twice as large for children who have had both parents incarcerated, relative to those who have had neither parent incarcerated. However, having both parents incarcerated did not significantly in- crease the odds of poor sleep and eating behaviors, relative to having 1 parent incarcerated.

Recent studies have revealed significant associations between parental incarceration and several health risks among off- spring at various stages of the life course.10-13,32,33 Although the specific mechanisms are unresolved, the theoretical ratio- nale for these associations is that incarceration represents a major stressor in a household that is one of many develop- mental risks incrementing the probability of poor health and

well-being. As previously mentioned, these developmental risks are wide ranging and include academic and socioemotional skills deficits,1-3 mental health problems and illicit drug use,3-7

and food insecurity.8,9 In general, the notion that incarcera- tion may influence aspects of offspring health and well- being across life stages is well-documented and, for the most part, empirically supported. For instance, researchers have found that parental incarceration during childhood can elevate the risk of posttraumatic stress symptomatology among chil- dren and increased body mass index among female progeny.32,33

Such findings may, to a certain extent, inform the link between parental incarceration, disrupted sleep, and unhealthy eating patterns among children.34,35 Other health conditions, both physical and mental, also appear to be more likely among ado- lescents and adults whose parents have been incarcerated. In addition to diminished education and earnings, depressive symptomatology and drunkenness were more likely to emerge among children of incarcerated parents during late young adulthood.11 Another group of researchers examined a large, nationally representative sample of youth and found that those who experienced parental incarceration during childhood were more likely to be seriously injured and have addiction- related health problems as adults.12 A related study employ- ing the same sample further expanded the health outcomes examined and found that a number of risky mental and physi- cal health conditions were more common among young adults who reported having an incarcerated parent during their child- hood, including anxiety, depression, posttraumatic stress dis- order, migraines, asthma, and high cholesterol.10 The incarceration of a family member during childhood also in- creases the risk of a heart attack later in life.13 In light of these findings and the results of the present investigation, it is

Figure. The predicted probability of poor sleep and eating behaviors in children by number of parents incarcerated. Children were considered to have poor sleep and eating behaviors if they met both of the following conditions: (1) they experienced short sleep durations and/or sleep problems at wave 4; and (2) they reported poor eating behaviors in 1 or more of the exam- ined eating domains.

June 2017 ORIGINAL ARTICLES

215Parental Incarceration and Child Sleep and Eating Behaviors

possible that poor sleep and diet during childhood may at least partially explain the associations detected between pa- rental incarceration and adolescent/adult health. Future re- search employing data from childhood to adulthood may want to empirically consider this possibility. Scholars may also want to explore whether poor sleeping and eating behaviors are an underlying mechanism for the relationship between parental incarceration and antisocial behavior, given the multiple in- dependent lines of research examining the associations between these factors.21,36-38

Ultimately, the precise process by which parental incarcera- tion might lead to poor sleep and diet in children is not cur- rently known, although a number of possible mechanisms could be at work, including disrupted/inconsistent routines, emo- tional stress/strain, childhood behavioral problems associ- ated with poor diet and sleep, parental modeling of high-risk sleep and eating behaviors, household chaos and/or health hazards, and/or a number of high-risk traits (including, in the case of biological parents, genetically influenced traits) that might be shared by parents and children. Regardless of the exact mechanisms at play, which are likely numerous, the results suggest that the negative sequelae of parental incarceration should be expanded to include a number of child behaviors that pose risks to their physical health.

Our study has limitations. First, the available measures per- taining to sleep and eating behaviors of children were some- what limited in the FFCWS. In particular, it would have been preferable to have additional details regarding the extent and nature of the reported sleep problems because parents were only asked about the child’s difficulty getting to sleep. Spe- cific questions about sleep disruptions, waking during the night, night terrors, and other sleep issues would have been useful to more fully explore the link between parental incarceration and childhood sleep behaviors. It also would have been pref- erable to have a larger number of items covering a broader array of specific foods that comprised the children’s diet. Simi- larly, more comprehensive measures of parental involvement (including paternal involvement) would have also been pre- ferred as covariates, including details surrounding the paren- tal monitoring of and involvement in child sleep and eating behaviors. We were unable to examine a number of other, po- tentially relevant child health behaviors, such as physical ex- ercise and sedentary behaviors. Second, the data did not permit an examination of whether or not the influence of parental incarceration on sleep and eating behaviors persisted beyond early childhood. The extent to which child sleep and eating be- haviors might explain long-term health outcomes will depend, at least in part, on the persistence of the risky sleep and eating behaviors over time. Future research in this area would benefit from a more comprehensive analysis of the stability of poor sleep and eating behaviors among children of incarcerated parents as they age into adolescence and adulthood. Third, we were unable to determine whether other family members in the home had also been incarcerated. Some research suggests that, although parental incarceration has important implica- tions for child health, the incarceration of any family members living with the child can also impact health outcomes later in

life.13 The negative effects of familial incarceration on child sleep and eating behaviors may not be fully accounted for in the current study if other family members have been incarcer- ated and such incarceration interferes with healthy eating and sleep habits. Finally, the generalizability of the results is not equivalent to prior studies using nationally representative samples because the present study employs the FFCWS, which oversampled high-risk families that were more likely to have experienced incarceration.

Maternal and paternal incarceration elevate the risk of poor sleep and dietary habits during childhood. Our findings un- derscore the importance of considering the unintended health consequences of incarceration for young children and ways to minimize their occurrence and persistence over time. Given that eating and sleeping patterns are potentially modifiable, additional efforts should be made to reach vulnerable popu- lations, such as those who have or have had an incarcerated parent. Although concerns about sleeping and eating are es- sential components of a pediatricians’ repertoire, queries re- garding household disruptions may include any parental contact with the criminal justice system as a means to identify chil- dren who may require special attention with respect to key health behaviors such as sleeping and eating patterns. On a policy level, schools can play an important role in childhood health behaviors, and advocating for healthier options at schools is one way to positively impact a large number of children. It may be worthwhile to consider the implementation of poli- cies that encourage and reward school systems for adopting healthier guidelines for meals and snacks, especially those in disadvantaged communities where there is a larger propor- tion of households affected by incarceration. ■

Submitted for publication Oct 27, 2016; last revision received Feb 27, 2017; accepted Mar 8, 2017

Reprint requests: Dylan B. Jackson, PhD, Department of Criminal Justice, The University of Texas at San Antonio, 501 W César E Chàvez Blvd, San Antonio, TX 78207. E-mail: [email protected]

References 1. Cho RM. Understanding the mechanism behind maternal imprison-

ment and adolescent school dropout. Fam Relat 2011;60:272-89. 2. Foster H, Hagan J. The mass incarceration of parents in America: issues

of race/ethnicity, collateral damage to children, and prisoner reentry. Ann Am Acad Pol Soc Sci 2009;623:179-94.

3. Haskins AR. Paternal incarceration and child reported behavioral func- tioning at age 9. Soc Sci Res 2015;52:18-33.

4. Murray J, Farrington DP. Parental imprisonment: effects on boys’ anti- social behaviour and delinquency through the life-course. J Child Psychol Psychiatry 2005;46:1269-78.

5. Murray J, Farrington DP. Parental imprisonment: long-lasting effects on boys’ internalizing problem through the life course. Dev Psychopathol 2008;20:273-90.

6. Wildeman C. Paternal incarceration and children’s physically aggressive behaviors: evidence from the fragile families and child wellbeing study. Soc Forces 2010;89:285-310.

7. Roettger ME, Swisher RR, Kuhl DC, Chavez J. Parental incarceration and trajectories of drug use from adolescence to young adulthood. Addic- tion 2011;106:121-32.

8. Cox R, Wallace S. Identifying the link between food insecurity and in- carceration. South Econ J 2016;82:1062-77.

THE JOURNAL OF PEDIATRICS • www.jpeds.com Volume 185

216 Jackson and Vaughn

9. Turney K. Paternal incarceration and children’s food insecurity: a con- sideration of variation and mechanisms. Soc Serv Rev 2015;89:335-67.

10. Lee RD, Xiangming F, Luo F. The impact of parental incarceration on the physical and mental health of young adults. Pediatrics 2013;131:e1188- 95.

11. Mears DP, Siennick SE. Young adult outcomes and the life course pen- alties of parental incarceration. J Res Crime Delinq 2016;53:3-35.

12. Miller HV, Barnes JC. The association between parental incarceration and health, education, and economic outcomes in young adulthood. Am J Crim Justice 2015;40:765-84.

13. White BA, Cordie-Garcia L, Fuller-Thompson E. Incarceration of a family member during childhood is associated with later heart attack: findings from two large, population-based surveys. J Crim Justice 2016;44:89-98.

14. Turney K. Stress proliferation across generations? Examining the rela- tionship between parental incarceration and childhood health. J Health Soc Behav 2014;55:302-19.

15. Georgieff MK, Brunette KE, Tran PV. Early life nutrition and neural plas- ticity. Dev Psychopathol 2015;27:411-23.

16. Liu J, Zhou G, Wang Y, Ai Y, Pinto-Martin J, Liu X. Sleep problems, fatigue, and cognitive performance in Chinese kindergarten children. J Pediatr 2012;161:520-5.

17. Von Stumm S. You are what you eat? Meal type, socio-economic status and cognitive ability in childhood. Intelligence 2012;40:576-83.

18. Ravid S, Afek I, Suraiya S, Shahar E, Pillar G. Kindergarten children’s failure to qualify for first grade could result from sleep disturbances. J Child Neurol 2009;24:816-22.

19. Feinstein L, Sabates R, Sorhaindo A, Rogers I, Herrick D, Northstone K, et al. Dietary patterns related to attainment in school: the importance of early eating patterns. J Epidemiol Commun H 2008;62:734-9.

20. Lavigne JV, Arend R, Rosenbaum D, Smith A, Weissbluth M, Binns HJ, et al. Sleep and behavior problems among preschoolers. J Dev Behav Pediatr 1999;20:164-9.

21. Jackson DB. The link between poor quality nutrition and childhood an- tisocial behavior: a genetically informative analysis. J Crim Justice 2016;44:13-20.

22. Suglia SF, Solnick S, Hemenway D. Soft drinks consumption is associ- ated with behavior problems in 5-year-olds. J Pediatr 2013;163:1323-8.

23. Astill RG, Van der Heijden KB, Van IJzendoorn MH, Van Someren EJ. Sleep, cognition, and behavioral problems in school-age children: a century of research meta-analyzed. Psychol Bull 2012;138:1109-38.

24. Malik VS, Pan A, Willett WC, Hu FB. Sugar-sweetened beverages and weight gain in children and adults: a systematic review and meta-analysis. Am J Clin Nutr 2013;98:1084-102.

25. American Academy of Pediatrics. American Academy of Pediatrics Sup- ports Childhood Sleep Guidelines. 06/13/2016; 2016.

26. Ogata BN, Hayes D. Position of the Academy of Nutrition and Dietet- ics: nutrition guidance for healthy children ages 2 to 11 years. J Acad Nutr Diet 2014;114:1257-76.

27. Hirshkowitz M, Whiton K, Albert SM, Alessi C, Bruni O, DonCarlos L, et al. National Sleep Foundation’s sleep time duration recommenda- tions: methodology and results summary. Sleep Health 2015;1:40-3.

28. Drewnowski A, Specter SE. Poverty and obesity: the role of energy density and energy costs. Am J Clin Nutr 2004;79:6-16.

29. Bell BG, Belsky J. Parents, parenting, and children’s sleep problems: ex- ploring reciprocal effects. Brit J Dev Psychol 2008;26:579-93.

30. Hingle MD, O’Connor TM, Dave JM, Baranowski T. Parental involve- ment in interventions to improve child dietary intake: a systematic review. Prev Med 2010;51:103-11.

31. Geller A, Garfinkel I, Western B. Paternal incarceration and support for children in fragile families. Demography 2011;48:25-47.

32. Bocknek EL, Sanderson J, Britner PA. Ambiguous loss and posttrau- matic stress in school-age children of prisoners. J Child Fam Stud 2009;18:323-33.

33. Roettger ME, Boardman JD. Parental incarceration and gender-based risks for increased body mass index: evidence from the National Longitudi- nal Study of Adolescent Health in the United States. Am J Epidemiol 2012;175:636-44.

34. Cespedes EM, Hu FB, Redline S, Rosner B, Gillman MW, Rifas-Shiman SL, et al. Chronic insufficient sleep and diet quality: contributors to child- hood obesity. Obesity (Silver Spring) 2016;24:184-90.

35. Gavrieli A, Farr OM, Davis CR, Crowell JA, Mantzoros CS. Early life ad- versity and/or posttraumatic stress disorder severity are associated with poor diet quality, including consumption of trans fatty acids, and fewer hours of resting or sleeping in a US middle-aged population: a cross-sectional and prospective study. Metabolism 2015;64:1597- 610.

36. Murray J, Farrington DP, Sekol I. Children’s antisocial behavior, mental health, drug use, and educational performance after parental incarcera- tion: a systematic review and meta-analysis. Psychol Bull 2012;138: 175.

37. Oddy WH, Robinson M, Ambrosini GL, Therese AO, de Klerk NH, Beilin LJ, et al. The association between dietary patterns and mental health in early adolescence. Prev Med 2009;49:39-44.

38. Simola P, Liukkonen K, Pitkäranta A, Pirinen T, Aronen ET. Psychoso- cial and somatic outcomes of sleep problems in children: a 4-year follow- up study. Child Care Health Dev 2014;40:60-7.

June 2017 ORIGINAL ARTICLES

217Parental Incarceration and Child Sleep and Eating Behaviors