Examining the Association between Recent Maternal Incarceration and Adolescents’ Sleep Patterns, Dietary Behaviors, and Physical Activity Involvement

Abstract

Maternal incarceration has become an increasingly common life event among adolescents in the U.S., especially those from disadvantaged backgrounds. Evidence suggests that maternal incarceration is a risk factor for adolescents’ internalizing and externalizing problems. However, its collateral consequences on adolescents’ sleep patterns, dietary behaviors and physical activity involvement remain unclear. This study uses the Fragile Family and Child Wellbeing Study dataset (N = 3262) to examine the prevalence of risk behaviors related to sleep, diet, and physical exercise among adolescents with maternal incarceration histories and assess the relationship between maternal incarceration and these behavioral health risks. Findings suggested that a significantly lower proportion of adolescents with maternal incarceration experiences consumed breakfast for at least four days a week than those without maternal incarceration experiences (41.78% vs. 52.76%, p = 0.009), a higher proportion of them consumed fast food for at least two days a week (57.35% vs. 48.61%, p = 0.035), had at least two sweetened drinks per day (71.92% vs. 62.21%, p = 0.018), and reported more days having problems staying asleep per week (1.70 vs. 1.26, p = 0.008). Regression analyses suggested that adolescents with maternal incarceration experiences were more likely to have problems staying asleep (B = 0.44, p = 0.012) than those without maternal incarceration experiences.

1. Introduction

Maternal incarceration is an increasingly prevalent life event that has been linked to a wide range of adverse health and behavioral outcomes among adolescents (e.g., depression, aggression, and substance use) [1,2,3,4,5]. The association between maternal incarceration and adolescents’ sleep patterns, dietary behaviors, and physical activity involvement, however, remains under-investigated [6,7,8,9,10]. Evidence suggests that sufficient sleep, adequate nutritional intake, and regular exercise are essential for healthy adolescent development [9]. Therefore, knowledge of the prevalence and etiology of related health risk behaviors is necessary for effective disease prevention in adulthood. This study uses the Fragile Family and Child Wellbeing Study (FFCW) dataset to examine the prevalence of risk behaviors related to sleep, diet, and physical exercise among adolescents with maternal incarceration histories. It also determines the role of maternal incarceration on adolescents’ sleep patterns, dietary behaviors, and physical activity involvement.

1.1. The Prevalence of Maternal Incarceration

Women represent the fastest-growing population in the U.S. institutional corrections facilities. In the past four decades, the number of women incarcerated has increased by more than 475%, rising from 26,326 in 1980 to 152,854 in 2020 [11]. Because the majority of imprisoned women are mothers with an average of two to three children [12], a conservative estimate indicates that at least one million American children have experienced maternal incarceration [1], and a substantial portion (42.5% to 50.4%) of them are adolescents [10,13,14]. It is important to note that, primarily due to structural and systematic bias in criminal justice policy and practice, adolescents of color are at significantly higher risk of experiencing maternal incarceration than their non-Hispanic White counterparts [3,4,15]. The broad scope and unequal distribution of maternal incarceration underscore the importance of research and practice to better serve this adolescent population [10].

1.2. Maternal Incarceration and Health Risk Behaviors

Maternal incarceration is a risk factor for a number of internalizing (e.g., depression and withdrawal) and externalizing problems (e.g., substance use and delinquency) among adolescents [2,3,4,5]. Nonetheless, the prevalence of sleep problems, unhealthy dietary behaviors, irregular physical activity involvement among adolescents with maternal incarceration experiences, and the role of maternal incarceration on their potentially elevated risks remain understudied [6,7,8,9,10].

There is little scientific debate that sufficient sleep, adequate nutritional intake, and regular physical exercise are cornerstones for healthy development [9], especially in adolescence—a critical transitional and formative period in the lifespan characterized by significant physical, cognitive, and psychosocial changes [10]. According to the Centers for Disease Control and Prevention, adolescents without adequate sleep are at a higher risk for obesity, diabetes, injuries, poor mental health, and attention and behavior problems [16,17,18]. Unhealthy eating behaviors can lead to a series of problems, including delayed sexual development and slower linear growth, as body changes during adolescence affect nutritional and dietary needs [19]. Moreover, irregular physical exercise can prevent adolescents from improving cardiorespiratory fitness and building bones and muscles, increase symptoms of anxiety and depression, and enhance their risk of health problems [20]. Because of the detrimental effects of risk behaviors related to sleep, diet, and exercise and the modifiable nature of unhealthy behaviors in adolescence, understanding the prevalence and etiology of these health risk behaviors is essential to disease prevention in adulthood.

There is limited research on the prevalence of sleep problems, unhealthy diets, and inadequate physical exercise among adolescents with incarcerated parents. A broad search of the extant literature using PsychINFO and Scopus identified only one study conducted by Hiolski and colleagues using cross-sectional data collected through the 2016 Minnesota Student Survey (N = 119,029) with related data [8]. It finds that youth with a history of parental incarceration differed from their peers with no history of parental incarceration in important ways. Specifically, a higher percentage of those with formerly or currently incarcerated parents were less physically active, consumed less fruit and vegetables but more fast food and sugar-sweetened beverages, and were getting inadequate sleep than their peers without parental incarceration exposure. A limitation of this study is the lack of gender differentiation regarding parental incarceration.

In terms of the association between parental incarceration and children’s sleep patterns, dietary behaviors, and physical activity involvement, three existing studies investigated these relationships. Two examined the impact of maternal and paternal incarceration but did not study adolescents. Specifically, Jackson and Vaughn’s study using the FFCW dataset reveals that maternal incarceration significantly increased the odds of experiencing sleep deprivation and exhibiting risky eating behaviors during childhood but did not significantly increase the odds of having trouble getting to sleep [9]. Heard-Garris and colleagues’ study using the National Longitudinal Survey of Adolescent to Adult Health dataset suggests that young adults with parental incarceration experiences were more likely to engage in several unhealthy behaviors [7]. In the unadjusted analyses, any history of maternal incarceration was associated with taking sugary drinks. However, maternal incarceration history was no longer significantly associated with any general health behaviors in adjusted analyses.

Moreover, the study by Hiolski and colleagues examined the effect of parental incarceration on adolescent children [8]. They find that youth with incarcerated parents had significantly lower odds of getting at least eight hours of sleep, consuming fruit and vegetables, and having significantly higher odds of reporting daily fast food consumption and drinking sugar-sweetened beverages than their peers with no history of parental incarceration. Moreover, youth with formerly incarcerated parents were less likely to report being physically active, whereas youth with currently incarcerated parents were more likely, compared with their counterparts without parental incarceration experiences, even after controlling for economic hardship. This is an important study that specifically focused on the impact of parental incarceration on adolescent health risk behaviors. Nonetheless, as previously mentioned, whether the effect of parental incarceration was due to the imprisonment of the mother, father, or both were not differentiated in this study.

1.3. The Current Study

Previous research suggests that maternal and paternal incarceration may have a distinct impact on adolescents’ health risk behaviors, with maternal incarceration potentially being more detrimental [5]. However, no study has specifically focused on these health risk behaviors among adolescents with maternal incarceration experiences. This study uses the FFCW dataset to understand the prevalence of sleep problems, unhealthy dietary behaviors, and inadequate physical activity involvement among adolescents with maternal incarceration histories. It also examines the role of maternal incarceration on these health risk behaviors. We hypothesize that a higher proportion of adolescents with maternal incarceration experiences will report sleep problems, unhealthy dietary behaviors, and inadequate physical activity involvement than their counterparts without such an experience. Moreover, we hypothesize that maternal incarceration is a risk factor for adolescents’ sleep problems, unhealthy dietary behaviors, and inadequate physical activity involvement. This study contributes to the growing literature on the prevalence and etiology of health risk behaviors of adolescents with maternal incarceration exposures [5].

2. Methods

2.1. Dataset

This study used the FFCW dataset for analysis. The FFCW study is based on a stratified, multistage sample of 4898 children born in 20 large cities throughout the United States between 1998 and 2000 [21]. This study oversampled births to unmarried mothers, which included a high proportion of Black/African American, Hispanic/Latino, and low-income families. Shortly after childbirth, 4897 mothers and 3830 fathers were interviewed in the hospital (wave 1), and follow-up interviews were conducted when children aged 1, 3, 5, 9, and 15 (wave 2–wave 6). These interviews collected data in various domains, including sociodemographic characteristics, interpersonal relationships, attitudes and behaviors, health, neighborhood characteristics, and program participation. This study used wave 6 data, particularly interviews with teens and their primary caregivers (PCGs).

2.2. Sample

At wave 6, interviews were conducted with 3444 teens and their PCGs, including 3027 biological mothers, 245 biological fathers, and 172 non-parental caregivers. Because only the biological mother and father (no other PCGs) were asked about the incarceration histories of the biological mother, the study sample only included adolescents whose PCGs were their biological mother or father at wave 6. As described in the Users’ Guide, if the biological mother or father lived with the teen for “half the time” or more, she or he was considered the primary caregiver [22]. The final sample featured 3262 adolescents, with 3019 biological mothers and 243 biological fathers.

2.3. Measures

2.3.1. Dependent Variables

The dependent variables in this study were three categories of health risk behaviors related to sleep, diet, and physical activity involvement.

Sleep Patterns. In measuring adolescents’ sleep patterns, we assessed self-reported sleep deprivation and insomnia symptoms (i.e., have problems falling asleep; have problems staying asleep). Sleep duration on school nights was calculated as the interval between reported bedtime and wake-up time on school nights ranging from 2 to 23. If the sleep duration was less than 7 h, they were considered sleep deprived [23]. Insomnia symptoms were assessed with two items that were coded independently. Adolescents were asked: “How many nights out of seven in a typical week do you have: (1) problems falling asleep throughout the night? and (2) problems staying asleep throughout the night?” Responses ranged from 0 to 7, with higher scores representing a higher frequency of having problems falling asleep or staying asleep, respectively.

Dietary Behaviors. In measuring risky dietary behaviors, this study specifically focused on their consumption of breakfast, fruit and vegetables, fast food, and sweetened drinks.

Breakfast consumption was assessed with the question, “How many days in a typical school week do you eat breakfast? Do not count the weekend.” The response categories ranged from 0 to 5. Literature indicates that eating breakfast 71% of the week is associated with a reduced risk of being overweight in adolescents [24]. Because the FFCW only surveyed breakfast consumption during the 5-day school week rather than all seven days, this variable was dichotomized to model the odds of consuming breakfast into ≥4 days (71% of 5 is 3.55, which rounds to 4 days or at least 80% of the time) during the school week and <4 days.

Consumption of fruit and vegetables was assessed with the question, “In a typical week, how many days do you eat at least some green vegetables or fruit?” The response categories ranged from 0 to 7. Based on the recommendations for fruit and vegetable consumption from the United States Department of Agriculture [25], the consumption of fruit and vegetables was dichotomized to model the odds of consumption all seven days of the week or <7 days.

Fast food consumption was assessed with the question, “How many days in a typical week do you eat food from a fast-food restaurant, such as McDonald’s, Burger King, Wendy’s, Arby’s, Pizza Hut, Taco Bell, or Kentucky Fried Chicken or a local fast-food restaurant?” Responses to this question ranged from 0 to 7. As most participants had consumed fast food at least one day of the week (85.00%), fast food consumption was dichotomized to model the odds of consumption ≥2 days during the week and <2 days [26].

Sweetened drink consumption was assessed with the question, “In a typical day, how many regular, non-diet sweetened drinks do you have? Include regular soda, juice drinks, sweetened tea or coffee, energy drinks, flavored water, or other sweetened drinks.” Based on literature indicating that ≥2 sweetened drinks per day were associated with weight gain [27], sweetened drink consumption was dichotomized to model the odds of consuming ≥2 sweetened drinks daily versus <2 drinks daily.

Physical Activity Involvement. Physical activity involvement was measured using two items. To measure moderate physical activity, adolescents were asked to report how many days in the past week they were physically active for a total of at least 60 min per day. Eight response options ranged from 0 to 7. The World Health Organization recommends adolescents engage in physical activity seven days per week [28]. Therefore, those who were physically active for seven days were coded “1”, and all others were coded “0”. To measure vigorous physical activity, adolescents were asked to report how many days in a typical week they participated in vigorous physical activities, with responses ranging from 0 to 7. According to the World Health Organization, adolescents should incorporate vigorous-intensity physical activities at least three days a week [28]. Adolescents who participated in vigorous physical activities for at least three days a week were coded as “1”, and others were coded as “0”.

2.3.2. Independent Variable

Maternal Incarceration. Maternal incarceration was reported by PCGs who were either the biological mother or father. PCGs were asked if the biological mother of the focal adolescent had spent time in jail since the last interview (i.e., Year 9). If the mother ever spent time in jail between the Year 9 and Year 15 interviews (i.e., when the focal child was 9 to 15 years old), the adolescent was considered to have experienced recent maternal incarceration (0 = without recent maternal incarceration experience, 1 = with recent maternal incarceration experience).

2.3.3. Control Variables

Several sociodemographic characteristics that might be associated with adolescent sleep patterns, dietary behaviors, and physical activity involvement were included in the analyses as control variables. Specifically, categorical variables included the adolescent’s gender (1 = boy, 0 = girl), race/ethnicity (i.e., non-Hispanic White, Black/African American, Hispanic/Latino, and Multiracial or Other), family structure (i.e., living with both biological parents, living with biological mother only, living with biological mother and new partner, living with biological father only, living with biological father and new partner, and living with other primary caregivers), housing stability (1 = yes, 0 = no), paternal incarceration history (1 = yes, 0 = no), and their mothers’ educational background (i.e., less than high school, high school or equivalent, some college, college or graduate school). Moreover, household income was included as a continuous variable.

2.4. Data Analysis

Data analyses proceeded in three steps using Stata 15 [29]. First, univariate analysis was conducted to develop a profile of adolescents aged 15 at wave 6 in the FFCW study. Second, we employed t-tests (for continuous variables) and chi-square tests (for categorical variables) to compare adolescents with and without recent maternal incarceration experiences in terms of sleep patterns, dietary behaviors, physical activity involvement, and sociodemographic characteristics. Lastly, controlling for all sociodemographic covariates, multivariate logistic regressions and multiple linear regressions were conducted to determine the association between maternal incarceration and adolescent health risk behaviors (i.e., sleep patterns, dietary behaviors, and physical activity involvement). Specifically, for sleep patterns, we conducted logistic regression on sleep deprivation and linear regressions on having problems falling asleep and having problems staying asleep, respectively. For dietary behaviors, we performed logistic regressions on breakfast consumption, fruit and vegetable consumption, fast food consumption, and sweetened drink consumption, respectively. For physical activity involvement, we conducted logistic regressions on moderate physical activity involvement and vigorous physical activity involvement, respectively. The predictor for all the models was maternal incarceration, with control variables incorporated in each model. There were limited missingness on maternal incarceration (3.43%), race/ethnicity (5.27%), and other variables (<1.60%). Stata only included observations with complete data in regression analyses.

3. Results

3.1. Descriptive Statistics

Table 1. Descriptive statistics of adolescents in the study sample, FFCW (N = 3262).

	N	M/%	SD	Range
Maternal Incarceration History	3150			0–1
Yes	147	4.67%
No	3003	95.33%
Sleep Patterns
Sleep Deprivation	3262			0–1
Yes	446	13.67%
No	2816	86.33%
Trouble with Falling Asleep	3241	1.75	2.04	0–7
Trouble with Staying Asleep	3241	1.28	1.96	0–7
Dietary Behaviors
Breakfast Consumption ≥ 4 Days per Week	3241			0–1
Yes	1699	52.42%
No	1542	47.58%
Fruit or Vegetables Consumption 7 Days per Week	3229			0–1
Yes	998	30.91%
No	2231	69.09%
Fast Food Consumption ≥ 2 Days per Week	3240			0–1
Yes	1581	48.80%
No	1659	51.20%
Sugar Consumption ≥ 2 per Day	3220			0–1
Yes	2007	62.33%
No	1213	37.67%
Physical Activity Involvement
Moderate Physical Activity 7 Days per Week	3210			0–1
Yes	593	18.47%
No	2617	81.53%
Vigorous Physical Activity ≥ 3 Days per Week	3210			0–1
Yes	1905	59.35%
No	1305	40.65%
Sociodemographic Characteristics
Gender	3262			0–1
Boy	1674	51.32%
Girl	1588	48.68%
Race/Ethnicity	3090
Non-Hispanic White	561	18.16%
Black/African American	1514	49.00%
Hispanic/Latino	769	24.89%
Multiracial or Other	246	7.96%
Mothers’ Education Level	3241
Less than High School	567	17.49%
High School/Equivalent	611	18.85%
Some College	1436	44.31%
College/Graduate School	627	19.35%
Household Income	3257	61,228.32	62,695.60	0–800,000
Family Structure	3257
Both Parents	964	29.60%
Mother Only	1313	40.31%
Mother and New Partner	757	23.24%
Father Only	105	3.22%
Father and New Partner	106	3.25%
Other PCGs	12	0.37%
Stable Housing	3255
Yes	3190	98.00%
No	65	2.00%
Paternal Incarceration	3214
Yes	376	11.70%
No	2838	88.30%

demonstrates the profile of adolescents in the FFCW study whose PCG was the biological mother or father. There were 147 (4.67%) adolescents who experienced maternal incarceration since the last wave of data collection (wave 5). Over half of the adolescents had breakfast at least four days a week (52.42%), and only 30.91% consumed fruit or vegetables seven days a week. Half of the sample consumed fast food more than two days a week (48.80%). Most participants consumed at least two sweetened drinks daily (62.33%). Only 18.47% of the adolescents were physically active seven days a week. Approximately 60% of the sample incorporated vigorous physical activities at least three days a week. The mean sleep hour during school days was 8.00 (SD = 1.33) among adolescents, with 13.67% reporting sleep deprivation. On average, adolescents reported having problems falling asleep for 1.75 days per week (SD = 2.04) and having problems staying asleep for 1.28 days per week (SD = 1.96). A majority of the adolescents were Black/African American (49.00%), followed by Hispanic/Latino (24.89%) and non-Hispanic White (18.16%). Approximately 80% of the biological mothers received a high school education. Most of the adolescents lived with their biological mother only (40.31%), followed by living with both parents (29.60%) and living with their biological mother and new partner (23.24%). The majority of adolescents reported having stable housing (98.00%). Approximately 12% of them experienced recent paternal incarceration.

3.2. Comparative Statistics

Table 2. Comparative analysis of adolescents with and without maternal incarceration histories in health risk behaviors and sociodemographic characteristics, FFCW (N = 3150).

	With Maternal Incarceration		Without Maternal Incarceration
	(N = 147)		(N = 3003)
	n	M (SD)/%	n	M (SD)/%	χ2/t	p
Sleep Patterns
Sleep Deprivation	25	17.01%	411	13.69%	1.30	0.255
Problems Falling Asleep	146	1.96 (2.24)	2984	1.73 (2.02)	−1.39	0.164
Problems Staying Asleep	145	1.70 (2.26)	2986	1.26 (1.95)	−2.65	0.008
Dietary Behaviors
Breakfast Consumption	61	41.78%	1575	52.76%	6.73	0.009
Fruit or Vegetables Consumption	48	33.10%	904	30.40%	0.48	0.489
Fast Food Consumption	84	57.53%	1451	48.61%	4.44	0.035
Sweetened Drink Consumption	105	71.92%	1844	62.21%	5.60	0.018
Physical Activity Involvement
Moderate Physical Activity	26	18.06%	540	18.26%	0.00	0.952
Vigorous Physical Activity	83	57.64%	1747	59.06%	0.11	0.735
Sociodemographic Characteristics
Gender (ref: Girl)	81	55.10%	1534	51.08%	0.91	0.341
Race/Ethnicity					2.55	0.466
Non-Hispanic White	25	17.73%	509	17.85%
Black/African American	72	51.06%	1403	49.21%
Hispanic/Latino	29	20.57%	716	25.11%
Multiracial or Other	15	10.64%	223	7.82%
Mothers’ Education Level					18.17	<0.001
Less than High School	29	20.28%	518	17.33%
High School/Equivalent	29	20.28%	558	18.67%
Some College	77	53.85%	1320	44.16%
College/Graduate School	8	5.59%	593	19.84%
Household Income	147	40,262.81 (34,337.94)	2998	61,255.38 (62,654.37)	4.03	<0.001
Family Structure					137.32	<0.001
Both Parents	12	8.16%	922	30.73%
Mother Only	72	48.98%	1237	41.23%
Mother and New Partner	30	20.41%	725	24.17%
Father Only	16	10.88%	61	2.03%
Father and New Partner	16	10.88%	44	1.47%
Other Primary Caregivers	1	0.68%	11	0.37%
Stable Housing	143	97.28%	2939	97.97%	0.33	0.567
Paternal Incarceration	41	28.47%	335	11.25%	38.46	<0.001

compares adolescents with and without maternal incarceration experiences in terms of sleep patterns, dietary behaviors, physical activity involvement, and sociodemographic characteristics. Bivariate analyses showed that a lower proportion of adolescents with recent maternal incarceration experiences consumed breakfast for at least four days a week than those without maternal incarceration experiences (41.78% vs. 52.76%, χ² = 6.73, p = 0.009), a higher proportion of them consumed fast food for at least two days a week (57.35% vs. 48.61%, χ² = 4.44, p = 0.035), and had at least two sweetened drinks per day (71.92% vs. 62.21%, χ² = 5.60, p = 0.018). Adolescents with and without maternal incarceration experiences were not significantly different in moderate or vigorous physical activity involvement. However, those with maternal incarceration experiences reported more days having problems staying asleep per week than those without maternal incarceration experiences (1.70 vs. 1.26, t = −2.65, p = 0.008).

Adolescents with and without maternal incarceration exposures also differed significantly in sociodemographic characteristics. Specifically, a higher proportion of mothers without incarceration experiences had at least a high school education (χ² = 18.17, p < 0.001). Adolescents with maternal incarceration experiences had lower household income than those without maternal incarceration experiences (t = 4.03, p < 0.001). A lower proportion of adolescents with maternal incarceration experiences lived with both parents than those without maternal incarceration experiences (8.16% vs. 30.73%, χ² = 137.32, p < 0.001), and a higher proportion of them reported recent paternal incarceration (28.47% vs. 11.25%, χ² = 38.46, p < 0.001).

3.3. Regression Analyses

As shown in

Table 3. Logistic and linear regression models on the association between maternal incarceration and adolescents’ sleep patterns, FFCW.

	Sleep Deprivation (N = 2941)			Have Problems Falling Asleep (N = 2941)		Have Problems Staying Asleep (N = 2929)
	OR	p	95% CI	B	p	B	p
Maternal Incarceration	1.29	0.290	0.80–2.08	0.32 +	0.073	0.44 *	0.012
Sociodemographic Characteristics
Gender a	0.92	0.433	0.74–1.13	−0.45 ***	<0.001	−0.35 ***	<0.001
Race/Ethnicity b
Black/African American	1.73 **	0.002	1.22–2.43	−0.48 ***	<0.001	−0.16	0.151
Hispanic/Latino	1.11	0.602	0.75–1.63	−0.39 **	0.001	−0.29 *	0.016
Multiracial or Other	1.48	0.101	0.93–2.36	−0.19	0.246	−0.12	0.455
Household Income	1.00	0.781	1.00–1.00	−0.00 *	0.016	−0.00 *	0.025
Mothers’ Education Level c
High School/Equivalent	1.10	0.641	0.75–1.60	−0.17	0.180	−0.11	0.352
Some College	1.35 +	0.072	0.97–1.86	−0.12	0.284	−0.07	0.523
College/Graduate School	1.42 +	0.092	0.94–2.12	−0.02	0.916	−0.26 +	0.052
Family Structure d
Mother Only	1.26	0.121	0.94–1.70	0.21 *	0.037	0.07	0.454
Mother and New Partner	1.36 +	0.051	1.00–1.86	0.36 **	0.001	0.37 ***	<0.001
Father Only	1.17	0.671	0.57–2.40	−0.46 +	0.070	−0.21	0.400
Father and New Partner	1.07	0.884	0.44–2.61	0.32	0.278	0.37	0.195
Other PCGs	0.78	0.817	0.10–6.25	−0.11	0.853	0.12	0.837
Housing Status e	1.33	0.518	0.56–3.13	0.20	0.462	−0.44 +	0.096
Paternal Incarceration	0.84	0.292	0.60–1.17	0.14	0.236	0.03	0.825
F Statistic				5.55	<0.001	4.76	<0.001

Note. ^a. Reference group is “Girl”; ^b. Reference group is “Non-Hispanic White”; ^c. Reference group is “Less than High School”; ^d. Reference group is “Living with Both Parents”; ^e. Reference group is “Unstable Housing”. ⁺ p < 0.10; * p < 0.05; ** p < 0.01; *** p < 0.001.

, multiple linear regression revealed that adolescents with maternal incarceration experiences were more likely to have problems staying asleep (B = 0.44, p = 0.012) after controlling for sociodemographic characteristics. However, maternal incarceration did not significantly predict adolescent sleep deprivation or trouble falling asleep.

Multivariate logistic regressions revealed that maternal incarceration was not significantly associated with healthy consumption of breakfast, fruit and vegetables, fast food, and sweetened drinks among adolescents after controlling for sociodemographic characteristics (see

Table 4. Logistic regression models on the association between maternal incarceration and adolescents’ dietary behaviors (FFCW).

	Breakfast Consumption (N = 2926)			Fruit and Vegetables Consumption (N = 2916)			Fast food Consumption (N = 2926)			Sweetened Drink Consumption (N = 2907)
	OR	p	95% CI	OR	p	95% CI	OR	p	95% CI	OR	p	95% CI
Maternal Incarceration	0.73 +	0.085	0.50–1.05	1.23	0.297	0.84–1.79	1.26	0.216	0.88–1.81	1.15	0.493	0.77–1.71
Sociodemographic Characteristics
Gender a	1.73 ***	<0.001	1.49–2.00	0.77 **	0.001	0.65–0.90	1.09	0.244	0.94–1.27	1.33 ***	<0.001	1.14–1.55
Race/Ethnicity b
Black/African American	0.90	0.337	0.71–1.12	0.74 *	0.010	0.58–0.93	1.96 ***	<0.001	1.56–2.46	1.81 ***	<0.001	1.44–2.28
Hispanic/Latino	1.01	0.950	0.79–1.29	0.56 ***	<0.001	0.43–0.73	1.66 ***	<0.001	1.30–2.13	1.24 +	0.086	0.97–1.60
Multiracial or Other	0.98	0.884	0.71–1.35	0.77	0.131	0.55–1.08	1.294	0.117	0.94–1.79	1.18	0.322	0.85–1.63
Household Income	1.00 +	0.083	1.00–1.00	1.00	0.180	1.00–1.00	1.00	0.600	1.00–1.00	1.00 **	0.002	1.00–1.00
Mothers’ Education Level c
High School/Equivalent	1.01	0.951	0.78–1.29	1.15	0.341	0.86–1.54	0.86	0.252	0.67–1.11	1.15	0.322	0.87–1.50
Some College	0.85	0.133	0.68–1.05	1.37 *	0.014	1.06–1.75	0.75 *	0.010	0.60–0.93	0.86	0.205	0.68–1.09
College/Graduate School	1.09	0.547	0.83–1.44	1.83 ***	<0.001	1.35–2.47	0.69 **	0.007	0.52–0.90	0.63 **	0.002	0.47–0.85
Family Structure d
Mother Only	0.79 *	0.020	0.65–0.96	0.97	0.764	0.78–1.20	1.37 **	0.002	1.12–1.67	1.28 *	0.021	1.04–1.57
Mother and New Partner	0.73 **	0.004	0.59–0.91	0.91	0.437	0.72–1.15	1.46 ***	<0.001	1.18–1.81	1.31 *	0.017	1.05–1.63
Father Only	0.67	0.123	0.41–1.11	0.75	0.316	0.42–1.32	1.01	0.963	0.61–1.68	1.93 *	0.022	1.10–3.39
Father and New Partner	1.15	0.644	0.63–2.10	1.20	0.565	0.65–2.22	1.55	0.143	0.86–2.78	2.02 *	0.038	1.04–3.94
Other PCGs	0.51	0.294	0.15–1.78	1.35	0.641	0.38–4.74	1.42	0.574	0.42–4.77	1.20	0.781	0.34–4.24
Housing Status e	1.15	0.606	0.67–1.96	0.85	0.582	0.48–1.51	0.93	0.796	0.55–1.59	0.71	0.262	0.39–1.29
Paternal Incarceration	0.92	0.487	0.73–1.16	1.15	0.274	0.89–1.48	0.99	0.927	0.78–1.25	1.34 *	0.029	1.03–1.75

Note. ^a. Reference group is “Girl”; ^b. Reference group is “Non-Hispanic White”; ^c. Reference group is “Less than High School”; ^d. “Living with Both Parents”; ^e. Reference group is “Unstable Housing”. ⁺ p < 0.10; * p < 0.05; ** p < 0.01; *** p < 0.001.

).

Additionally, as illustrated in

Table 5. Logistic regression models on the association between maternal incarceration and adolescents’ physical activity involvement (FFCW, N = 2904).

	Moderate Physical Activity Involvement			Vigorous Physical Activity Incorporation
	OR	p	95% CI	OR	p	95% CI
Maternal Incarceration	0.83	0.463	0.51–1.36	1.04	0.856	0.71–1.51
Sociodemographic Characteristics
Gender a	2.64 ***	<0.001	2.15–3.24	2.73 ***	<0.001	2.33–3.19
Race/Ethnicity b
Black/African American	1.00	0.983	0.75–1.34	1.14	0.279	0.90–1.44
Hispanic/Latino	0.90	0.517	0.65–1.24	1.31 *	0.042	1.01–1.69
Multiracial or Other	1.45 +	0.055	0.99–2.13	1.49 *	0.023	1.06–2.11
Household Income	1.00	0.218	1.00–1.00	1.00 ***	<0.001	1.00–1.00
Mothers’ Education Level c
High School/Equivalent	1.22	0.255	0.87–1.72	1.41 **	0.009	1.09–1.83
Some College	1.10	0.548	0.81–1.49	1.36 **	0.007	1.09–1.71
College/Graduate School	1.41 +	0.062	0.98–2.03	1.38 *	0.032	1.03–1.84
Family Structure d
Mother Only	0.79 +	0.075	0.61–1.02	0.90	0.311	0.73–1.11
Mother and New Partner	0.83	0.179	0.63–1.09	0.88	0.282	0.71–1.11
Father Only	1.54	0.145	0.86–2.75	0.90	0.704	0.53–1.53
Father and New Partner	1.56	0.177	0.82–3.00	0.69	0.230	0.37–1.27
Other PCGs	0.63	0.663	0.08–5.05	0.54	0.344	0.15–1.92
Housing Status e	0.80	0.496	0.41–1.54	1.03	0.920	0.59–1.79
Paternal Incarceration	1.09	0.601	0.80–1.48	0.99	0.934	0.77–1.27

Note. ^a. Reference group is “Girl”; ^b. Reference group is “Non-Hispanic White”; ^c. Reference group is “Less than High School”; ^d. Reference group is “Living with Both Parents”; ^e. Reference group is “Unstable Housing”. ⁺ p < 0.10; * p < 0.05; ** p < 0.01; *** p < 0.001.

, multivariate logistic regressions revealed that maternal incarceration was not significantly associated with moderate or vigorous physical activity involvement among adolescents after controlling for sociodemographic characteristics.

4. Discussion

This study is among the first to examine the prevalence of sleep problems, unhealthy dietary behaviors, and inadequate physical activity involvement among adolescents with maternal incarceration experiences and to determine the association between maternal incarceration experiences and these health risk behaviors. Comparative analyses revealed that a lower proportion of adolescents with experiences of maternal incarceration consumed breakfast for at least four days a week, and a higher proportion consumed fast food for at least two days a week, had at least two sugar-sweetened drinks per day, and had trouble staying asleep when compared to adolescents with no such experience. This finding is broadly consistent with our first hypothesis. After controlling for sociodemographic characteristics, regression analyses suggested that maternal incarceration was a risk factor for having trouble staying asleep, which partially supports our second hypothesis. Findings from this study contributed to the growing literature on the consequences of maternal incarceration on adolescent health risk behaviors and may inform interventions to change their risky behaviors and improve population health.

The current study only identified maternal incarceration as a risk factor for one health risk behavior (i.e., having trouble staying asleep) among adolescents after controlling for sociodemographic characteristics such as gender, race/ethnicity, family structure, and income. Maternal incarceration—due to the increased likelihood of separation from biological parents [30,31]—may lead to new household contexts (e.g., living with new household members or a foster home) that better support healthy eating and exercise. On the other hand, having trouble staying asleep at night may reflect the difficulty of adjusting to maternal incarceration. Given that the majority of incarcerated women serve as the solo caregiver of children before imprisonment, children may experience emotional distress (e.g., ambivalence, fear, anger, and worrisome), trauma due to witnessing the often violent arrest of the mother, residential instability, and financial strain immediately following the occurrence of maternal incarceration [32]. As a traumatic event, maternal incarceration may set in motion chains of risks over time, including disrupted mother-child bonds, social stigmatization, social network change, and negative interaction with other systems (e.g., child welfare, school, and juvenile justice) [33,34,35]. These stressors may collectively contribute to the heightened risk of insomnia among adolescents. While there are effective behavioral intervention programs targeting insomnia among adolescents [36], there is no evidence-based intervention targeting the health risk behaviors of adolescents with incarcerated mothers [37]. To meet the unique needs of this adolescent population, existing effective interventions such as cognitive behavioral therapy may need to incorporate psychological treatment components to address stress and anxiety stemming from maternal incarceration [38]. Moreover, to alleviate the negative consequences of sleep problems and improve health equity among adolescents, future research that investigates the mechanism linking maternal incarceration and adolescent sleep problems is necessary.

Similar to what Heard-Garris and colleagues found [7], once sociodemographic characteristics were controlled, regression analyses did not show any other impact. While a strength in this analysis is the isolation of the effect of maternal incarceration without sociodemographic characteristics, these variables and their interactions may be necessary to better understand the structural issues and health beliefs and practices that may lead to such risk. For example, Curtin and colleagues analyzed health prevention beliefs and approaches by social determinants of health, such as gender, education, and income [39]. They found significant differences that may impact health behavior. In this case, when considering maternal incarceration, targeting specific social determinants of health, as independently associated with maternal incarceration, may be particularly important to achieve more significant health equity. It is also important to note that maternal incarceration remains prevalent among racial/ethnic minorities due to structural racism and other factors, such as poverty [3,40]. It is, therefore, difficult to disentangle a variable such as incarceration from income, race, and other structural social determinants of health that impact who is more vulnerable to the consequences of the criminal justice system and the effects of maternal incarceration. Moreover, adolescence is an important developmental stage characterized by not only physical changes but also mental and social changes that make adolescents’ health behaviors susceptible to the impact of contextual factors such as family climate, peer relationships, and school experiences [10]. Future research is needed to explore these interactive relationships and better understand the health prevention needs of adolescents with maternal incarceration exposures.

Despite the results of regression analyses, bivariate findings still indicate that more adolescents with experiences of maternal incarceration reported unhealthy dietary behaviors and poor sleeping quality compared to peers without such experience. These findings highlight the basic needs of adolescents with recent maternal incarceration experiences, especially in a critical developmental stage. Family stress theory may offer a valuable lens for a better understanding of why a higher proportion of adolescents with maternal incarceration exposures are practicing unhealthy dietary behaviors and experiencing insomnia. After maternal incarceration, stress endured by the family or the caregiver may diminish its capacity to adequately respond to the needs of its members, including adolescent children [1,41]. For instance, these adolescent children may experience diminished parental supervision and financial strain, which may contribute to potential neglect and abuse [34]. Other research has suggested maternal incarceration represents a unique risk marker of common other family and youth risk experiences that are also present in this sample, such as lower household income, lower maternal education, and a higher prevalence of paternal incarceration [42]. The overlap of these vulnerabilities with food insecurity is well documented [43]. Moreover, the often separation from both biological parents after maternal incarceration [30,31] may lead to greater life disruption that then changes health behavior patterns (e.g., limited access to food and opportunities for exercise) for some family units, and disrupts typical sleep patterns due to changes in mood, anxiety, or other responses to this stressful experience.

Regardless of the ways maternal incarceration may or may not serve as a unique risk mechanism, it still serves as an important risk marker of family poverty which can limit opportunities to engage in healthy eating, physical activity, and sleep. Future research should further examine the association between maternal incarceration and the health behaviors of adolescents with such experience, explicitly considering the nature and characteristics of maternal incarceration. For instance, recent research suggests that the type of correctional facilities confined (e.g., jail or prison), facility security level, the length of incarceration, the number of episodes of incarceration, and maternal incarceration trajectories, can all make a difference in its impact on adolescent behaviors [44]. For these reasons, future studies should expand on how maternal incarceration is defined and measured and use comprehensive measures that capture the full picture of maternal incarceration experiences to distinguish subgroup differences. In the current study, maternal incarceration is measured as a dichotomous variable—any time since the last wave of interviews (i.e., in the last six years). There is room for great variability in this experience, from a brief arrest and detainment to long-term incarceration. It is also essential to understand the impact of detention on family status—such as entering the child welfare system or moving with a supportive family member—which may allow researchers to better understand these children’s varied experiences. Additionally, some other individual characteristics related to adolescent health behaviors (e.g., ADHD) were not incorporated into the analysis. Future research that aims to understand and disentangle the interaction among individual characteristics, contextual factors and maternal incarceration on healthy adolescent development is warranted. It should also be noted that the data relied solely on caregiver self-reported data, which may limit the accuracy of all responses. Finally, this study is based on children born in 20 large cities, and the sample overrepresents African American, Hispanic/Latino, and low-income families. Therefore, caution is needed in generalizing the findings beyond this sampling frame.

Moreover, it is essential to explore policies and programs that can reduce the impact of structural and systemic factors on adolescents with incarcerated mothers, thereby improving healthy youth development. While some prior programming has been developed to offer support groups and family skills training for these adolescents and their caregivers, there continues to be a dearth of programming focused on diet, exercise, and sleep [45]. Supports to these families might come in the form of nutrition education, access to healthy food through existing school programs (including breakfast) during the school year and summer months, case management that links these families to counseling services, food banks, or other local nutrition programs, and additional accessible opportunities to become involved in sports or other physical activities. It is important that researchers, scholars, and community members advocate for changes in policy to fund such programs, given the increasing prevalence of maternal incarceration among adolescents in the U.S. Broader changes in policy that focus on reducing child poverty in the U.S. are also likely to benefit this population.

5. Conclusions

This study is among the first to use a large national dataset to explore health risk behaviors related to sleep patterns, dietary behaviors, and physical activity involvement among adolescents with maternal incarceration experiences. Findings suggest that a lower proportion of adolescents with maternal incarceration experiences regularly consumed breakfast, and a higher proportion of them frequently consumed fast food and sugar-sweetened drinks and had trouble staying asleep when compared with adolescents without such experience. Moreover, maternal incarceration was a risk factor for poor sleep health, specifically having trouble staying asleep. Findings highlight the importance of enhancing and developing programs and policies to best meet this adolescent population’s physical and mental health needs. Future research examining the relationship between maternal incarceration and intersecting determinants of health among adolescents is warranted.