Journal of Adolescent Health
Volume 43, Issue 2 , Pages 141-150, August 2008

Moderation and Mediation in the Relationship Between Mothers' or Fathers' Serious Psychological Distress and Adolescent Substance Use: Findings from a National Sample

  • Mindy A. Herman-Stahl, Ph.D.

      Affiliations

    • RTI International, Research Triangle Park, North Carolina
    • ,
  • Olivia Silber Ashley, P.H

      Affiliations

    • RTI International, Research Triangle Park, North Carolina
    • Corresponding Author InformationAddress correspondence to: Olivia Silber Ashley, P.H., RTI International, Risk Behavior and Family Research Program, 3040 Cornwallis Road, Research Triangle Park, NC 27709.
    • ,
  • Michael A. Penne, M.P.H.

      Affiliations

    • RTI International, Research Triangle Park, North Carolina
    • ,
  • Karl E. Bauman, Ph.D.

      Affiliations

    • RTI International, Research Triangle Park, North Carolina
    • ,
  • Jason Williams, Ph.D.

      Affiliations

    • RTI International, Research Triangle Park, North Carolina
    • ,
  • Rebecca P. Sanchez, Ph.D.

      Affiliations

    • Pacific Institute for Research and Evaluation, Chapel Hill, North Carolina
    • ,
  • Kellie M. Loomis, M.Ed

      Affiliations

    • RTI International, Research Triangle Park, North Carolina
    • ,
  • Megan S. Williams, M.S.P.H., M.S.W.

      Affiliations

    • RTI International, Research Triangle Park, North Carolina
    • ,
  • Joseph C. Gfroerer

      Affiliations

    • Office of Applied Studies, Substance Abuse and Mental Health Services Administration, Rockville, Maryland

Received 1 May 2007; accepted 11 January 2008. published online 02 May 2008.

Article Outline

Abstract 

Purpose

This study estimated percentages of adolescents living with a mother or father with serious psychological distress (SPD), and examined moderation and mediation of the relationships between mother or father SPD and adolescent substance use.

Methods

We analyzed data from nationally representative samples of adolescents interviewed with their mothers (n = 4734) and fathers (n = 3176) in the combined 2002 and 2003 National Surveys on Drug Use and Health (NSDUHs).

Results

An estimated 4.1% of adolescents living with their father had a father with SPD during the past year, and 11.5% of adolescents living with their mother had a mother with SPD during this time period. A positive association was found between mothers' SPD and adolescent binge drinking (OR = 1.49, 95% CI = 1.01–2.21), but no association was found between fathers' SPD and adolescent binge drinking. Mothers' SPD was associated with increased risk of binge drinking among adolescents aged 14–15 years (OR = 2.52, 95% CI = 1.38–4.60), and fathers' SPD was associated with lowered risk of binge drinking among black adolescents (OR = .08, 95% CI = .01–.79). A positive association was found between mothers' SPD and adolescent illicit drug use (OR = 1.55, 95% CI = 1.08–2.23), but no association was found between fathers' SPD and adolescent illicit drug use. Mothers' SPD was associated with increased risk of illicit drug use among female adolescents (OR = 2.14, 95% CI = 1.24–3.70) and among adolescents of white ethnicity (OR = 1.78, 95% CI = 1.19, 2.68). Parental involvement partially mediated the relationship between mothers' SPD and daughters' illicit drug use; mothers' SPD was associated with lower levels of parental involvement, which in turn were associated with an increased probability of daughters' illicit drug use.

Conclusions

Overall, parents' SPD is associated differentially with adolescent substance use depending on the gender of parent and adolescent, adolescent age, race/ethnicity, and substance used. Parental involvement appears to be one mechanism through which mothers' SPD influences daughters' illicit drug use. Future research should further consider the interindividual effects of parents' SPD and associated parenting behaviors on adolescent risk behaviors.

Keywords: Parent psychological distress, Adolescent, Binge drinking, Underage drinking, Illicit drug use, Parental involvement

 

Adolescents of psychologically distressed parents are more likely to experience internalizing and externalizing disorders, poor academic achievement, reduced social competence, suicidal behaviors, and cognitive impairments compared with adolescents with parents who do not report psychological symptoms [1], [2], [3], [4]. A small amount of literature has linked parents' psychological symptoms with increased risk of adolescent substance use, early initiation, and abuse [5]. However prior research has primarily focused on the effects of mothers' psychological distress on child outcomes [6], and the impact of fathers' psychological symptoms is largely unexplored. Moreover research typically has focused on clinical disorders among small samples of predominantly white, middle-class parents. The present study increases our knowledge base by examining the association between mothers' or fathers' subclinical levels of psychological distress and adolescent substance use, using a nationally representative sample. Identifying variables that moderate and mediate the association between mothers' or fathers' psychological distress and adolescent substance use in a variety of family situations is essential for further understanding and intervention.

Back to Article Outline

Potential moderators 

Our research is guided by Goodman and Gotlib's Integrative Model for Transmission of Risk [1]. The model developers explored whether some children are more vulnerable than others to the interpersonal context or the dysfunctional neuroregulatory mechanism associated with psychological disorder. A few studies have found that adolescent girls are more susceptible to mothers' depression, possibly because of girls' sensitivity to resulting family discord [7]. Some researchers attest that parents' psychological symptoms are more disruptive to younger than older children because younger children are more dependent on their mothers [4], whereas other investigators assert that the relationship between mothers' psychological symptoms and youth behavioral problems becomes stronger as children get older. Low family income appears to increase risk for negative outcomes by increasing family stress and lowering fiscal resources [8].

To date, no study has examined race/ethnicity as a moderator of the relation between mothers' psychological problems and child outcomes. Racial minorities are sometimes under-represented in mental health care [9], and barriers to obtaining services combined with stress associated with minority status [10] may increase the likelihood of substance use among minority children of parents with mental illness. Furthermore, although parent education has not been explored as a potential moderator, stress associated with low levels of education, barriers that a lack of education may pose on service access, and poorer parent–child interaction stemming from parent psychological distress combined with low socioeconomic status [11], increase the likelihood of negative outcomes for adolescents. Few studies have addressed differences in outcomes among subgroups of children exposed to fathers' psychological problems, relying on small samples of fathers from two-parent households.

Parental involvement as a potential mediator 

The Integrative Model for Transmission of Risk posits that parental psychological symptoms may influence child development vis-à-vis parenting behaviors [1]; parents experiencing psychological symptoms display more sad and irritable affect, are less positive and affectionate, withdraw from their children more often, and vacillate between punitive discipline and permissive undercontrol more than do parents without psychological symptoms [4]. Parental involvement might therefore mediate the relationship between maternal or paternal psychological distress and adolescent substance use.

In this study we examined (a) the percentage of adolescents living with their father who had a father with serious psychological distress (SPD) and the percentage of adolescents living with their mother who had a mother with SPD; (b) the associations between mothers' or fathers' SPD and adolescent substance use; (c) sociodemographic moderators of these associations, and (d) parental involvement as a mediator of these associations. As depicted in Figure 1, research about parents' psychological distress and other adolescent outcomes informed the following hypotheses: (a) Mothers' or fathers' SPD will be associated with increased risk of adolescent substance use; (b) Adolescent gender, age, and race/ethnicity, mothers' or fathers' education, and family income will moderate the relations between mothers' or fathers' SPD and adolescent substance use; and (c) Mothers' or fathers' SPD will be related to decreased parental involvement, which in turn will be linked with a higher probability of adolescent substance use. The current study extends previous work by analyzing a nonclinical, nationally representative sample of mother–child pairs and father–child pairs, examining a variety of family forms, and including important moderators and mediators.

Back to Article Outline

Methods 

Data for this study were from the combined 2002 and 2003 National Surveys on Drug Use and Health (NSDUH); the NSDUHs provide annual estimates of SPD and substance use in the noninstitutionalized, civilian U.S. population [12], [13]. Combining 2 years of data was necessary to obtain adequate sample sizes of fathers with SPD to conduct moderation analyses. The original survey procedures were approved by RTI International's Committee for the Protection of Human Subjects and the Office of Management and Budget (OMB). Active parental consent and adolescent assent were obtained for adolescent participation in the study. The face-to-face household interview took approximately 1 hour. Computer-assisted interviewing and audio computer-assisted self-interview were used. In the 2002 and 2003 surveys, a $30 incentive payment was offered. The overall response rates were 78% (n = 68,126) in 2002 [12] and 77% (n = 67,784) in 2003 [13]. Other details of the survey design and data collection procedures have been reported elsewhere [12], [13].

Samples 

At each selected address, sample persons were randomly selected using an automated screening procedure programmed in a hand-held computer carried by NSDUH interviewers. If screening was not completed during the first visit, at least four callbacks were made at different hours on different days of the week. To provide the necessary sample sizes for the specified population age groupings used in NSDUH, the screening procedure involved listing all household members along with their demographic data. Using a preprogrammed algorithm selection, the computer selects no, one, or two sample persons, depending on the age composition of all eligible household members [14]. Adolescents aged 12–17 years had the highest probability of selection. If a selected respondent was unavailable, at least four additional visits were made at different times of the day on different days of the week. Selected respondents who refused participation or who were unavailable for interview were not replaced with another member of the household. Analyses for the present study were conducted on the subsamples of adolescents aged 12–17 years who were selected and whose mother (n = 4734) or father (n = 3176) living in the home was also selected. (The sampling algorithm of selecting a maximum of two persons from any household precluded the possibility of obtaining information from an adolescent and both parents.) The terms “mother” and “father” in this study refer to biological, step-, adoptive, and foster parents.

Measures 

NSDUH uses the K6 Scale of Non-Specific Psychological Distress for adults aged 18 years or more [15], [16], originally designed for use in the core of the redesigned U.S. National Health Interview Survey to measure frequency of psychological distress symptoms [17]. It measures cognitive, behavioral, emotional, and psycho-physiological symptoms commonly elevated among adults with a wide range of mental disorders [16]. Selected mothers or fathers rated the frequency of six symptoms during the 1 month of the past 12 months when they felt the most depressed, anxious, or emotionally stressed. The symptoms were as follows: feeling nervous; hopeless; restless or fidgety; so sad or depressed that nothing could cheer you up; everything was an effort; and down on yourself, no good, or worthless. The response categories ranged from none of the time (0) to most of the time (4). Reliability (Cronbach's α) was .89. Responses were combined to produce a score ranging from 0–24, with a score of 13 identified as the optimal cut-off for discriminating serious mental illness cases, based on previous research [16], [18]. The NSDUH refers to the K6 measure as “serious psychological distress (SPD)” [19]. Dichotomous measures of SPD were chosen for our analyses instead of continuous measures because the data were skewed. SPD was positively correlated with receiving mental health treatment (defined as inpatient treatment, outpatient treatment, or use of prescription medications for treatment during the past year) among mothers (tetrachoric correlation coefficient = .67) and fathers (.55).

Adolescent binge drinking and past-month illicit drug use were measured by creating two dichotomous measures from several items that asked adolescents to indicate use of alcohol or any illicit drug during the past 30 days. Illicit drugs included marijuana or hashish, cocaine, heroin, hallucinogens, inhalants, or psychotherapeutic drugs (pain relievers, tranquilizers, stimulants, or sedatives) used nonmedically. Adolescents were regarded as past-month illicit drug users if they reported use of any illicit drug within the past 30 days. Adolescents who responded that they had had five or more drinks on the same occasion on at least 1 day within the past 30 days were considered binge drinkers.

Social and demographic characteristics assessed by NSDUH included adolescent age, adolescent gender, total annual household income, and mother or father education. Adolescent age was categorized as 12–13, 14–15, and 16–17 years, corresponding to early, middle, and late adolescence and similar to categories defined in past research [20]. We grouped adolescent race/ethnicity into three categories: non-Hispanic white, non-Hispanic black, or Hispanic. Individuals from other racial/ethnic groups were excluded from analyses of moderation by adolescent race/ethnicity because of small sample sizes.

In this study, parental involvement is defined as the degree to which parents (1) spend time relating with their children concerning schoolwork, and (2) show acceptance and warmth. Involvement, defined using these two dimensions, has been negatively correlated with adolescent problem behavior [21]. A four-item scale, including two items from each dimension, was used (Cronbach's α = .75). Adolescents were asked to report the frequency (1 = always, 4 = never) of parents in the home helping with homework, checking homework, expressing pride in the adolescent, and praising the adolescent. After reverse coding, items were averaged. This measure is similar to one used in a recent study of parental involvement and adolescent substance use, using Monitoring the Future data [22].

Because parents' substance use may affect adolescent substance use, and so as not to incorrectly attribute the effects of parents' substance use to the effects of parents' SPD, mothers' or fathers' substance use was controlled for in all analyses. Adults' SPD and substance use are correlated [23], [24], [25], and we identified marginal tetrachoric correlations among mothers (.24) and fathers (.06). Because distributions were skewed, we created a dichotomous indicator based on mothers' or fathers' self-reports of binge alcohol use (based on consumption of five or more drinks on the same occasion at least once in the past 30 days) or any illicit drug use during the past 30 days. Past-month mothers' or fathers' substance use was indicated if the parent reported any of the above. Because adolescents living with their father were more likely to be in a two-parent household (95%) than adolescents living with their mother (77%), we controlled for family structure in all analyses.

Item response rates were above 96% for study variables. For all variables except parent SPD and parental involvement that had missing or ambiguous values after editing, statistical imputation was used to replace these values with appropriate response codes [12], [13].

Data analysis 

The data were analyzed using SUDAAN software [26], which applies a Taylor series linearization method to account for the complex design features of the NSDUH. Parent–child pair weights were constructed to represent the population of eligible person pairs within NSDUH dwelling units [14]. The weights were adjusted for pair nonresponse and calibrated to match group household roster information from a larger sample of dwelling units. The weighted estimates are representative of the national populations of (1) adolescents aged 12–17 years living with their mothers and (2) adolescents aged 12–17 living with their fathers (regardless of whether a second parent was also in the home). Separate analyses were conducted for mother–child and father–child pairs. First, we calculated the percentages of adolescents living with a father or mother with SPD. We used odds ratios and 95% confidence intervals to examine the relationship between mothers' or fathers' SPD and each dichotomous adolescent substance use outcome (Hypothesis A). We augmented the basic logistic regression model with an interaction term for mothers' or fathers' SPD and each sociodemographic variable (Hypothesis B). When the interaction term was statistically significant, the sample was stratified by the categories of the relevant sociodemographic variable to examine the nature of associations between mothers' or fathers' SPD and adolescent substance use at various levels of the moderating variable. Mediation by parental involvement was evaluated using the standard z-test based on the multivariate delta standard error for the estimate of the mediated effect (Hypothesis C) [27], [28]. The product of the coefficients relating mothers' or fathers' SPD to the mediator and the mediator to adolescent substance use formed the point estimate of the mediated effect. This was then divided by an estimate of its standard error, and the resulting ratio was compared with the standard normal distribution for significance testing. To scale effects comparably, all effects were standardized based on the unobserved variance in the binary outcome [29].

To consider length of time living with the parent, we tested interactions between parent status (biological vs. foster vs. step- or adoptive parent) and parents' SPD. There were no significant interactions between parent status and parents' SPD and when predicting adolescent substance use. Therefore parent status × parents' SPD interaction terms were dropped.

Back to Article Outline

Results 

As shown in Table 1, approximately 11.5% of adolescents living with their mother had a mother with SPD, and 4.1% of adolescents living with their father had a father with SPD.

Table 1. Characteristics of study population
VariableAdolescents living with mothersAdolescents living with fathers
na%MeanSDna%MeanSD
Parent's SPD
Yes62811.5 1574.1
No410688.5 301995.9
Adolescent binge drinking
Yes4889.2 3009.4
No424690.8 287690.6
Adolescent past-month illicit drug use
Yes53410.5 3049.0
No420089.5 287291.0
Adolescent gender
Male237651.1 164252.2
Female235848.9 153447.8
Adolescent age (y)
12–13178335.2 126035.4
14–15158934.0 105132.7
16–17136230.8 86531.9
Adolescent race/ethnicity
Non-Hispanic white321562.1 228668.9
Non-Hispanic black60214.9 2669.3
Hispanic62116.4 42215.3
Asian1084.4 734.0
American Indian/Alaska Native61.8 461.0
Two or more ethnicities1271.4 831.6
Education of the parent
High school graduate or less226148.2 158346.3
Some college or more247351.8 159353.7
Total annual family income
<$20,00079015.7 2538.1
$20,000–49,000175434.4 109231.2
$50,000–75,00098519.7 76722.6
>$75,000120530.1 106438.0
Parental involvement4734 3.3.683176 3.4.66
Substance use by parentb
Yes94617.8 114633.7
No377082.2 203066.3
Family structure
Two parents353177.2 298294.9
One parent120322.8 1945.1

SD = standard deviation; SPD = serious psychological distress.

Percentages for each variable may not total 100% because of rounding.

aTotal n represents the study sample size.

bIncludes binge drinking or past-month use of any illicit drug.

Mothers' SPD was positively associated with adolescent binge drinking (OR = 1.49, 95% CI = 1.01–2.21) after controlling for mothers' substance use and family structure. However fathers' SPD was not associated with adolescent binge drinking. The association between mothers' SPD and adolescent binge drinking was moderated by adolescent age, such that mothers' SPD increased risk of binge drinking among adolescents aged 14–15 years (OR = 2.52, 95% CI = 1.38–4.60) but not among adolescents in other age groups (Table 2). The association between fathers' SPD and adolescent binge drinking was moderated by adolescent race/ethnicity: Among adolescents of black ethnicity, the odds of binge drinking were lower when living with a father with SPD than when living with a father without SPD (OR = .08, 95% CI = .01–.79), but fathers' SPD was not associated with adolescent binge drinking among adolescents of white or Hispanic ethnicity.

Table 2. Summary of logistic regression moderation models for interactions between mothers' or fathers' serious psychological distress (SPD) and sociodemographic characteristics and for main effects analyses among stratified samples predicting binge drinking among adolescents living with mothers (n = 4734) or fathers (n = 3176)
PredictorAdolescents living with mothersAdolescents living with fathers
OR(95% CI)OR(95% CI)
Moderation models
SPD × Adolescent male gender1.39(.62–3.12)1.31(.31–5.45)
SPD × Adolescent age
SPD × 12–13.64(.15–2.63)1.07(.15–7.70)
SPD × 14–152.48⁎(1.01–6.09).98(.17–5.62)
SPD × Adolescent race/ethnicitya
SPD × Black.47(.05–4.20).06⁎(.01–.60)
SPD × Hispanic2.45(.73–8.15).73(.11–4.76)
SPD × Education of the parent
SPD × HS graduate or less.56(.24–1.32).57(.15–2.15)
SPD × Total annual family income
SPD × <$20,000.64(.15–2.68)—b—b
SPD × $20,000-$49,9991.34(.38–4.66)1.25(.19–8.20)
SPD × $50,000-$74,999.54(.14–2.16)3.05(.49–18.80)
Stratified models
Adolescents aged 12–13.56(.13–2.33)
Adolescents aged 14–152.52⁎⁎(1.38–4.60)
Adolescents aged 16–171.16(.61–2.21)
White adolescents 1.54(.70–3.41)
Black adolescents .08⁎(.01–.79)
Hispanic adolescents 1.10(.20–6.17)

CI = confidence interval; OR = odds ratio

Analyses controlled for binge alcohol use or past-month illicit drug use by the parent and family structure. Reference categories are age 16–17 years, female gender, white race/ethnicity, some college or more, two parents, and total annual income of $75,000 or more. Sample sizes are those shown in Table 1.

aAdolescents from other racial/ethnic groups are excluded from this model.

bThis interaction could not be calculated because at least one cell had zero observations.

⁎p < .05;

⁎⁎p < .01; ***p < .001; ****p < .0001.

Mothers' SPD was positively associated with adolescent illicit drug use (OR = 1.55, 95% CI = 1.08–2.23) after controlling for mothers' substance use and family structure, but fathers' SPD was not associated with adolescent illicit drug use. As shown in Table 3, the association between father SPD and adolescent illicit drug use was not moderated by adolescent age, gender, or race/ethnicity, or by father education or family income. However the association between mothers' SPD and adolescent illicit drug use was moderated by adolescent gender (OR = .46, 95% CI = .22–.95), such that mothers' SPD was positively associated with illicit drug use among daughters (OR = 2.14, 95% CI = 1.24–3.70) but not among sons. Adolescent race/ethnicity moderated the relationship between mothers' SPD and adolescent illicit drug use, such that mothers' SPD was positively associated with illicit drug use among white adolescents (OR = 1.78, 95% CI = 1.19–2.68) but not among black or Hispanic adolescents.

Table 3. Summary of logistic regression moderation models for interactions between mother or father serious psychological distress (SPD) and sociodemographic characteristics and for main effects analyses for stratified models predicting past-month illicit drug use among adolescents living with mothers (n = 4734) or fathers (n = 3176)
PredictorAdolescents living with mothersAdolescents living with fathers
OR(95% CI)OR(95% CI)
Moderation models
SPD × Adolescent male gender.46⁎(.22–.95)3.51(.73–17.00)
SPD × Adolescent age (y)
SPD × 12–131.10(.30–3.98)2.67(.38–18.68)
SPD × 14–151.51(.66–3.47).53(.09–3.27)
SPD × Adolescent race/ethnicitya
SPD × Black.18⁎⁎(.05–.66)1.53(.23–10.34)
SPD × Hispanic.61(.19–1.94)1.03(.15–7.10)
SPD × Education of parent
SPD × high school graduate or less1.23(.56–2.68).41(.10–1.71)
SPD × Total annual family income
SPD × <$20,0001.82(.54–6.11).18(.02–1.91)
SPD × $20,000–49,9992.54(.89–7.25).36(.05–2.53)
SPD × $50,000–74,9992.04(.51–8.19).37(.05–2.95)
Stratified models
Adolescent males1.12(.72–1.76)
Adolescent females2.14⁎⁎(1.24–3.70)
White adolescents1.78⁎⁎(1.19–2.68)
Black adolescents.33(.10–1.12)
Hispanic adolescents1.04(.32–3.38)

Reference categories are age 16–17 years, female gender, white race/ethnicity, some college or more, two parents, and annual income of $75,000 or more. Sample sizes are those shown in Table 1. Stratified models controlled for binge alcohol use or past-month illicit drug use by the parent and family structure.

aAdolescents from other racial/ethnic groups are excluded from this model.

⁎p < .05;

⁎⁎p < .01.

We tested for mediation among significant associations between mothers' or fathers' SPD and adolescent substance use (Table 4). Parental involvement was a significant mediator of the relationship between mothers' SPD and adolescent illicit drug use, but only for daughters (z = 2.05, p = .04). Mothers' SPD was associated with lower levels of parental involvement, which in turn was associated with an increased probability of adolescent illicit drug use. The direct effect (i.e., impact of mothers' SPD not mediated by parent involvement) was also significant, with mothers' SPD associated with significantly greater odds of illicit drug use among daughters (OR = 1.88, p = .03). Parental involvement did not mediate the relation between fathers' SPD and adolescent illicit drug use.

Table 4. Summary of tests of mediation by parental involvement of the relation between mothers' or fathers' serious psychological distress and adolescent substance use among subsamples of adolescents living with their mothers or fathers
Subsamplez
Adolescent binge drinking
Adolescents aged 14–15 years living with their mother.80
Black adolescents living with their father.84
Adolescent past-month illicit drug use
Daughters living with their mother2.05⁎
White adolescents living with their mother1.26

Analyses controlled for parent binge alcohol use or past-month illicit drug use and family structure.

⁎p < .05.

Back to Article Outline

Discussion 

Our estimates that fathers' SPD affects 4.1% of adolescents living with their father and that mothers' SPD affects 11.5% of adolescents living with their mother are comparable to findings in a recent report that, in 2002, an estimated 8.9% of parenting adults experienced serious psychological distress during the past year [30], and to the estimate of frequent mental distress as assessed by the CDC's Behavioral Risk Factor Surveillance System (BRFSS) (8.6%) [31]; but our findings are higher than estimates of SPD found by the National Health Interview Survey (3.1%), which measured past-month SPD [17] instead of SPD during the worst month of the past year. These studies are not completely comparable to the present analyses, however, because they measure SPD in the adult population instead of percentages of adolescents who live with a mother or father with SPD.

Overall, more adolescents living with their mother had a mother with SPD than did adolescents living with their father who had a father with SPD. Other studies have similarly found that psychological problems are more common among women than men [12], [13], [30], [31]. Mothers' SPD was more consistently linked with adolescent substance use than was fathers' SPD, perhaps because of mothers' increased role in caretaking compared with that of fathers [32]. Also, because 95% of adolescents who live with fathers also live with a second parent, the presence of the second parent may offer protection from the effects of paternal SPD.

We found that maternal SPD increased the risk of binge drinking among adolescents aged 14–15 years but not among those from other age groups. Mothers' SPD may have a greater impact during middle adolescence, a period of vulnerability for binge drinking [33], when alcohol use trajectories diversify. Escalation from initiation of alcohol use to heavier use, such as binge drinking, has been associated with family dysfunction [34].

Adolescent gender moderated the association between mothers' SPD and adolescent illicit drug use, such that the impact was greater for girls. Past research has shown that poor family functioning resulting from maternal depression predicts conduct problems in girls but not in boys [7]. Same-gender parent–adolescent dyads also intensify during adolescence, and mother–daughter relationships become more contentious [35].

Unexpectedly, fathers' SPD was associated with decreased risk of binge drinking among adolescents of black ethnicity. Families of this ethnicity are more likely to be composed of single mothers, and they make greater use of extended family members [36]. The absence of fathers from many black families may spur adolescents and other family members to rally around a psychologically distressed father in the home. This idea, as well as the associations between paternal SPD and adolescent binge drinking in black families, should be further investigated in future research.

Maternal SPD was associated with illicit drug use among white adolescents but not among black or Hispanic adolescents. Racial/ethnic differences in parenting style [37] may protect minority adolescents from adolescent drug use over and above the effects of maternal SPD. In addition minority adolescents appear to be more influenced by environmental aspects of substance use, such as opportunities for involvement in drug selling and exposure to community violence [38].

Mothers' SPD influenced daughters' illicit drug use through decreased parental involvement. When mothers with psychological problems fall short of fulfilling their parenting responsibilities, daughters may initiate adult roles earlier, such as taking on excessive responsibility in running the household, taking care of siblings, and addressing the emotional or physical health needs of their impaired parent [39]. These adolescents may perceive substance use as a means to escape, to cope with stress, or to validate their adult roles. The lack of significant mediated effects among adolescents living with fathers with SPD as well as the significant direct effect of SPD on illicit drug use among daughters living with SPD mothers suggest that further investigation is needed to explore potential casual chains through mediators other than parental involvement. More research is also needed to clarify what types of parental involvement mediate the relationship between mothers' SPD and daughters' drug use. Moreover the hypothesized caretaking role of daughters of psychologically ill mothers should be more carefully studied. Additional pathways, such as mother–daughter conflict and mothers' co-occurring substance use, should also be explored.

Unfortunately, research suggests that only about one-third of children with a parent who has psychological problems receive any mental health services [6]. It is important to link mothers with SPD to family-centered programs that address parenting skills and that provide daughters with skills and additional support known to be protective against substance use. In addition it may be beneficial to assess the mental health status of mothers of adolescents identified as substance users to determine whether these parents need intervention as well. Screening and outreach may be needed to identify these parents, as they may not exhibit symptoms reaching clinical levels of distress or seek treatment on their own.

Study limitations 

The present study had several limitations. Cross-sectional data do not capture the changing dynamics of SPD symptomatology, family relations, and adolescent substance use patterns over time. Moreover cross-sectional designs do not provide evidence of temporality as would be desirable for some of the associations considered. For example it is plausible that mothers' or fathers' SPD is “caused” by adolescent substance use behaviors, and it is possible that parent involvement is a consequence rather than cause of adolescent substance use.

The self-report nature of the NSDUH may limit the accuracy of the data, and the exclusion of some high-risk populations (e.g., homeless persons) limits the generalizability of the results. Moreover, although most adolescents live with two parents, we did not have data from the second parent in these homes. Furthermore a parent with SPD who is not living in the household may still influence adolescent substance use behaviors, but we were unable to measure this type of relationship. It is also possible that having SPD would make a person more likely to refuse to participate or to be unable to participate in the survey. In addition parents with the most severe psychological distress may have had to place their children in alternative care [3] and thus would not be included in our sample of parent–child pairs. Thus the results likely underestimate adolescents' exposure to parents' SPD, and the effects of such underestimation on the associations, moderations, and mediations cannot be assessed.

By using a cut-off for the K6 screening scale, we may have neglected the impact of minor psychological distress on parenting and adolescent substance use. We conducted analyses using a continuous measure of SPD but did not identify a dose–response relationship between parents' SPD and adolescent substance use, possibly because of the heavy skewness of the SPD variable toward zero values among the population. Lack of additional information on noncustodial parental influences, extrafamilial influences, history and severity of psychological problems, and other family and adolescent characteristics further limits the conclusions that can be drawn from these data. Unexpected null findings may be attributable to small cell sizes (particularly Hispanic and black father–child pairs). We considered that the significant moderations found could be caused by the number of comparisons; but all analyses stemmed from past literature and were planned tests of hypotheses.

Finally, the questions about parental involvement lack specificity because they refer collectively to all parents in the household and are subject to interpretation by the participating adolescent. Our parental involvement construct assesses a basic and important level of involvement of parents in the life of their child from the perspective of the adolescent. Ultimately, we are limited to measures available in the data set, a common limitation of secondary data analyses of national studies [40].

Back to Article Outline

Conclusions 

Mothers' or fathers' SPD was associated with different substance use outcomes among subgroups of adolescents. Thus prevention, early identification, and outreach targeting adolescents may need to be sensitive to important factors such as gender of the parent with SPD; race/ethnicity, gender, and age of the adolescent; and different substance use behaviors. Our moderation findings render consideration only of direct effects of mothers' or fathers' SPD inappropriate. In particular daughters, white adolescents, and adolescents aged 14–15 years living with mothers with SPD were most at risk for substance use, so helping resources should be prepared to address these populations. Mothers' SPD was linked with decreased parental involvement, which in turn was associated with greater likelihood of illicit drug use among daughters; and the diminished maternal nurturing and support are implicit in these data. Future research should continue to explore the effects of mothers' and fathers' SPD, the question of why some subgroups are more affected than others by a parent's SPD, and the subtleties of associated parenting behaviors.

Back to Article Outline

References 

  1. Goodman SH, Gotlib IH. Risk for psychopathology in the children of depressed mothers: A developmental model for understanding mechanisms of transmission. Psychol Rev. 1999;106:458–490
  2. Hans SL, Auerbach JG, Styr B, Marcus J. Offspring of parents with schizophrenia: Mental disorders during childhood and adolescence. Schizophr Bull. 2004;30:303–315
  3. Nicholson J, Biebel K, Hinden B, et al. Critical issues for parents with mental illness and their families. Rockville, MD: Center for Mental Health Services, Substance Abuse and Mental Health Services Administration; 2001;
  4. Oyserman D, Mowbray CT, Meares PA, Firminger KB. Parenting among mothers with a serious mental illness. Am J Orthopsychiatry. 2000;70:296–315
  5. DelBello MP, Geller B. Review of studies of child and adolescent offspring of bipolar parents. Bipolar Disord. 2001;3:325–334
  6. Mowbray CT, Bybee D, Oyserman D, et al. Diversity of outcomes among adolescent children of mothers with mental illness. J Emot Behav Disord. 2004;12:206–221
  7. Davies P, Windle M. Gender-specific pathways between maternal depressive symptoms, family discord, and adolescent adjustment. Dev Psychol. 1997;33:657–668
  8. Mowbray CT, Schwartz S, Bybee D. Mothers with a mental illness: Stressors and resources for parenting and living. Fam Soc. 2000;81:118–129
  9. US Department of Health and Human Services. Mental health: A report of the Surgeon General. Rockville, MD: Substance Abuse and Mental Health Services Administration, Center for Mental Health Services, National Institutes of Health, National Institute of Mental Health; 2000;
  10. Williams DR, Neighbors HW, Jackson JS. Racial/ethnic discrimination and health: Findings from community studies. Am J Public Health. 2003;93:200–208
  11. Harnish JD, Dodge KA, Valente E Conduct Problems Prevention Research Group. 1995. Mother–child interaction quality as a partial mediator of the roles of maternal depressive symptomatology and socioeconomic status in the development of child behavior problems. Child Dev. 1995;66:739–753
  12. Substance Abuse and Mental Health Services Administration. Results from the 2002 National Survey on Drug Use and Health: National findings. NHSDA Series H-22, DHHS Publication No. SMA 03-3836 Rockville, MD: Substance Abuse and Mental Health Services Administration; 2003;
  13. Substance Abuse and Mental Health Services Administration. Results from the 2003 National Survey on Drug Use and Health: National findings. Office of Applied Studies, NSDUH Series H-25, DHHS Publication No. SMA 04-3964 Rockville, MD: Substance Abuse and Mental Health Services Administration; 2003;
  14. Chromy JR, Singh AC. Estimation for person-pair drug-related characteristics in the presence of pair multiplicities and extreme sampling weights. Proceedings of the Annual Meeting of the American Statistical Association Atlanta, GA: American Statistical Association; 2001;
  15. Furukawa TA, Kessler RC, Andrews G. The performance of the K6 and K10 screening scales for psychological distress in the Australian and National Survey of Mental Health and Well-Being. Psychol Med. 2003;33:357–362
  16. Kessler RC, Andrews G, Colpe LJ, et al. Short screening scales to monitor population prevalences and trends in non-specific psychological distress. Psychol Med. 2002;32:959–976
  17. Centers for Disease Control and Prevention. Early release of selected estimates based on data from the 2004 National Health Interview Survey. [Online] http://www.cdc.gov/nchs/data/nhis/earlyrelease/earlyrelease200506.pdfAccessed October 9, 2005
  18. Kessler RC, Barker PR, Colpe LJ, et al. Screening for serious mental illness in the general population. Arch Gen Psychiatry. 2003;60:184–189
  19. Substance Abuse and Mental Health Services Administration. Results from the 2004 National Survey on Drug Use and Health: National findings. Office of Applied Studies, NSDUH Series H-28, DHHS Publication No. SMA 05-4062 Rockville, MD: Substance Abuse and Mental Health Services Administration; 2005;
  20. Clark-Lempers DS, Lempers JD, Ho C. Early, middle, and late adolescents' perceptions of their relationships with significant others. J Adolesc Health. 1991;6:296–315
  21. Grolnick WS, Ryan RM. Parent styles associated with children's self-regulation and competence in school. J Educ Psychol. 1989;81:143–154
  22. Pilgrim CC, Schulenberg JE, O'Malley PM, et al. Mediators and moderators of parental involvement on substance use: A national study of adolescents. Prev Sci. 2006;7:75–89
  23. Epstein J, Barker P, Vorburger M, Murtha C. Serious mental illness and its co-occurrence with substance use disorders, 2002. Rockville, MD: Substance Abuse and Mental Health Services Administration, Office of Applied Studies; 2004;
  24. Walter M, Dammann G, Wiesbeck GA, Klapp BF. Psychosocial stress and alcohol consumption: Interrelations, consequences and interventions. Fortschr Neurol Psychiatr. 2005;73:517–525
  25. Kaplow JB, Curran PJ, Dodge KA The Conduct Problems Prevention Research Group. Child, parent, and peer predictors of early-onset substance use: A multisite longitudinal study. J Abnorm Child Psychol. 2002;30:199–216
  26. Research Triangle Institute. SUDAAN. Release 8.0. Research Triangle Park, NC: Research Triangle Institute; 2001;
  27. MacKinnon DP, Lockwood CM, Hoffman JM, et al. A comparison of methods to test mediated and other intervening variable effects. Psychol Methods. 2002;7:83–104
  28. Sobel ME. Asymptotic confidence intervals for indirect effects in structural equation models. Sociol Methodol. 1982;13:290–312
  29. MacKinnon DP, Dwyer JH. Estimating mediated effects in prevention studies. Eval Rev. 1993;17:144–158
  30. Herman-Stahl M, Ashley OS, Penne MA, et al. Serious psychological distress among parenting and non-parenting adults. Am J Public Health. 2007;97:2222–2229
  31. Centers for Disease Control and Prevention. Self-reported frequent mental distress among adults—United States, 1993–1996. MMWR. 1998;47:325–331
  32. Connell AM, Goodman SH. The association between psychopathology in fathers versus mothers and children's internalizing and externalizing behavior problems: A meta-analysis. Psychol Bull. 2003;128:746–773
  33. Tucker JS, Orlando M, Ellickson PL. Patterns and correlates of binge drinking trajectories from early adolescence to young adulthood. Health Psychol. 2003;22:79–87
  34. Colder CR, Campbell RT, Ruel E, et al. A finite mixture model of growth trajectories of adolescent alcohol use: Predictors and consequences. J Consult Clin Psychol. 2002;70:976–985
  35. Collins WA, Russell G. Mother–child and father–child relationships in middle childhood and adolescence: A developmental analysis. Dev Rev. 1991;11:99–136
  36. Garcia Coll C, Lamberty G, Jenkins R, et al. An integrative model for the study of developmental competencies in minority children. Child Dev. 1996;67:1891–1914
  37. Steinberg L, Darling N, Fletcher AC. Authoritative parenting and adolescent adjustment: An ecological journey. In:  Moen P,  Elder JGH,  Luscher K editor. Examining Lives in Context: Perspectives on the Ecology of Human Development. 4:Washington, DC: American Psychological Association; 1995;p. 23–66
  38. Luthar SS, D'Avanzo K. Contextual factors in substance use: A study of suburban and inner-city adolescents. Dev Psychopathol. 1999;11:845–867
  39. Valiakalayil A, Paulson LA, Tibbo P. Burden in adolescent children of parents with schizophrenia: The Edmonton High Risk Project. Soc Psychiatry Psychiatr Epidemiol. 2004;39:528–535
  40. Brooks-Gunn J, Phelps E, Elder GH. Studying lives through time: Secondary data analyses in developmental psychology. Dev Psychol. 1991;27:899–910

PII: S1054-139X(08)00111-0

doi:10.1016/j.jadohealth.2008.01.010

Journal of Adolescent Health
Volume 43, Issue 2 , Pages 141-150, August 2008

Article Tools

* Image Usage & Resolution