Ego Development, Psychopathology, and Parenting Problems in Substance-Abusing Mothers

Abstract

The authors examined maternal ego development in relation to psychopathology and parenting problems in a sample of substance abusing mothers. Given predilections at higher levels of ego development for introspection and guilt, the authors expected mothers at higher levels to report more psychopathology. Given predilections at lower levels of ego development for dichotomous perceptions and limited conceptions of causation, the authors expected mothers at low levels to report more problematic parenting behaviors. Intelligence was expected to correlate but not overlap with ego development. Subjects were 182 mothers who expressed interest in a randomized clinical trial for a new parenting intervention. Measures included the Washington University Sentence Completion Task—Short Form, the Parental Acceptance-Rejection Questionnaire, the Brief Symptom Inventory and the Kaufman Brief Intelligence Test. Results of correlation and multivariate analyses of variance confirmed predictions. Implications for future development of interventions for substance abusing mothers are discussed.

Although not all mothers who abuse substances have difficulty parenting their children, as a group, substance abusing women are at least twice as likely as nonsubstance using women to lose custody of their children because of reoccurrences of substance abuse and child neglect (Department of Health & Human Services [DHHS], 1999). Over the last 20 years, research has shown that, although some mothers who use illicit substances are able to parent their children in ways that support their development, as a group, mothers with substance use disorders are at greater risk for mal-adaptive parenting than mothers who do not have substance use problems. Parental substance abuse has been linked with poor discipline skills (use of coercive control, harsh discipline, and failure to follow through), limited or absent parental monitoring, ineffective control of children’s behaviors, and problems regulating aggression (Keller, Cummings, & Davies, 2005; Pears, Capaldi, & Owen, 2007; Smyth, Miller, Mudar, & Skiba, 2003). Parental substance abuse has also been linked with lower levels of parental involvement, poorer synchrony, and more negative parental behaviors (Pears et al., 2007).

Parenting problems of substance abusing mothers also seem to be a function of other factors, most notably sociodemographic risk and psychological dysfunction (DHHS, 1999). The presence of comorbid psychiatric disorders appears to exacerbate the parenting problems of many substance abusing mothers and increases the chance that a biological child will be permanently placed out of home (DHHS, 1999; Dunnet al., 2002; Hans, Bernstein, & Henson, 1999; Nair et al., 1997). A less studied aspect of psychological functioning that may also have implications for parenting practices of substance abusing mothers is ego development, or the way in which one makes coherent sense of the world (Loevinger, 1976). Belsky (1993) and Abidin (1992) have suggested that the core psychological functioning of a parent is critical to his or her capacity to manage ongoing environmental demands and to buffer children from their negative effects. Ego development can be understood as one such core psychological capacity that may have critical implications for parenting and child outcomes. For mothers with substance use disorders, understanding ego development in relation to psychopathology and parenting may have important implications for intervention development in terms of timing and approach to intervention delivery.

Loevinger’s Theory of Ego Development

Jane Loevinger (Hy & Loevinger, 1996; Loevinger, 1976) has conceptualized ego development as a staged organization of experience beginning with infancy, with each ego development level (e.g., E2, E3) characterizing a period of child development but also observable as levels of arrested ego development in adults (see Table 1). The Impulsive Stage (E2) is characterized by preoccupations with physical needs and impulses; dependence on others for control; dichotomous evaluations of others as simply “good” or “bad”; poor understanding of rules and causation; and defiance as the only means for enacting autonomy. At this level, physical and emotional symptoms are merged and there is little sense of an inner life. The Self-Protective Stage (E3) is characterized by a capacity to delay immediate gratification; rudimentary impulse control; self-serving awareness of the advantages of rules; use of interpersonal skills for exploiting others; wariness and self-protection in relation to others; and an embracement of rituals, traditions, rules, and controls. Together, the first two levels comprise the Pre-Conformist Stages.

Table 1.

Levels of Ego Development and Example of a Scored Item (Hy & Loevinger, 1996)

Stage	Characteristics	Response to sample item: “A wife should…”
E2 Impulsive	Focus on physical needs and impulses; dependence on others for control and limits; rules are poorly understood; exploitation of others for one’s own good; Physical and emotional needs are merged; no sense of inner emotional experience	“…be good.”
E3 Self-protective	Capacity to delay immediate gratification; self-serving appreciation for rules; interpersonal wariness	“…clean the house and cook if she’s not working.”
E4 Conformist	Belonging to group is paramount; “right” and “wrong” simply determined by social convention; self and others viewed in terms of “fit” with group norms; simplistic and banal inner life	“…take care of her children and love her husband.”
E5 Self-awareness	Allowance for qualifications of “right” and “wrong” based on demographic differences; capacity for loneliness and self-consciousness; expanded inner life; relationships experienced in terms of feelings rather than behaviors	“…feel she is loved and needed.”
E6 Conscientious	“Right” and “wrong” and decisions based on personal feeling; capacity for guilt; concern about hurting others; feeling responsible for others	“…be a good mother, lover, listener and business person.”
E7 Individualistic	Greater tolerance for individual differences; differentiation between inner and outer states; emergence of long-term perspective; Interpersonal relationships are deeper and more intense	“…try to understand her husband’s feelings.”
E8 Autonomous	Freedom from excessive striving and achievement; search for self-fulfillment; recognition of individual human complexity; tolerance for ambiguity and paradox; deepened respect for autonomy of others	“…try to develop herself as a person and not try to be all things to all people.”
E9 Integrated	Self-actualization; not fully described	Not fully described

Note. E = ego development level.

The next two levels comprise the Conformist Stages of development where group pressure to conform provides the motivation for movement beyond the Pre-Conformist level. The Conformist Stage (E4) is characterized by a preoccupation with group membership; use of group conventions to guide definitions of right and wrong; views of self and others in terms of social group stereo-types and ways they fit (or don’t fit) in with social norms; and an inner life that is understood only in simple terms (e.g., sad, happy, glad, angry). The Self-Awareness Stage (E5), is characterized by self-examination (in response to a growing awareness that competing authorities can make conformity impossible); recognition of multiple alternatives; qualification of choices in terms of demographic differences (not deep personal or individual desire); prevalence of loneliness and self-consciousness; and a greater capacity to conceptualize inner life and experience relationships in terms of feelings rather than actions.

At the Conscientious Stage (E6), the first Post-Conformist Stage, judgments about right or wrong are made on the basis of personal feelings, and decisions are made for personal reasons rather than reflexively or normatively. Long-term goals and ideals are characteristic. Guilt arises out of the possibility of hurting another person rather than breaking a rule and there is a tendency to feel excessively responsible for others. Inner states and individual differences are described in vivid and differentiated terms. At the Individualistic Stage (E7), a sense of individuality emerges and there is greater tolerance for individual differences and differentiation between inner and outer state. Relationships with others become deeper and more intense.

Starting at the Individualistic Stage (E7), and continuing in the Autonomous Stage (E8), there is an emergent freedom from excessive striving and responsibility as the search for self-fulfillment replaces the striving for achievement. There is also a growing recognition of the complexity and multifaceted character of “real people” in “real situations” (Hy & Loevinger, 1996, p. 6) and a high tolerance for ambiguity and paradox. There is a deepened respect for other people and their need for autonomy, particularly the need of family members and children to find their own way and to make their own mistakes. As well, there is an increased focus on psychological causation and development. The Integrated Stage (E9) – the highest and rarely achieved level—represents the “theoretical high point,” similar to Maslow’s self-actualizing person, but not been fully described because of a lack of empirical evidence or rater agreement on how it manifests in individuals.

Ego Development, Substance Abuse, Psychopathology and Parenting

Ego development and substance abuse

Only a handful of published studies involving opiate abusers have focused on ego development in substance abusing populations (Noam, Young, & Jilnina, 2006). The modal level of ego development among opiate abusers in previous studies has been Self-Protective (E3), although a significant percentage have fallen at the Conformist (E4) and Self-Awareness (E5) Stages as well, indicating that some opiate users are functioning at higher levels of ego development (Blatt et al., 1984; Fineman, Beckwith, Howard, & Espinosa, 1997; Wilber, Rousnaville, Sugarman, Casey, & Kleber, 1982). Notably, almost no opiate users (male or female) have been found to have ego development levels above the Self-Awareness Stage.

Ego development and psychopathology

In a review of studies examining ego development and psychopathology, Noam and colleagues (2006) reported that, whereas psychosis and personality disorders inhabit the Impulsive (E2) and Self-Protective (E3) levels, diagnostic categories characterized by anxiety and affective disorders are represented in higher proportions at the Conformist (E4) and Self-Awareness (E5) Stages. With regard to depression, there is considerable support for a curvilinear hypothesis, that depression is highest at the Conformist Stage, where emphasis on conformity to group norms may suppress individual emotional awareness and expression. At Pre-Conformist levels, where thinking is insufficiently complex and the self too undifferentiated to accommodate depression, and at the Post-Conformist levels, where meeting group standards becomes a lesser priority and failure to meet them less catastrophic, depression levels are lower. Studies measuring associations between anxiety and ego development have similarly shown a curvilinear relationship, with anxiety peaking at the Self-Awareness Stage (E5), where concern about meeting group norms is still strong (Noam et al., 2006).

Ego development and psychopathology in substance abusing adults

Preliminary evidence from studies examining depression in substance abusing adults (Luthar, Doyle, Suchman, & Mayes, 2001; Wilber et al., 1982) suggests a positive linear association between ego development and depression. However, because the highest level of ego development reported in these studies was the Self-Awareness Stage, it is impossible to rule out a curvilinear relationship. Similarly, in one study examining anxiety in a sample of substance abusing adults (e.g., Wilber et al., 1982), a positive correlation between anxiety and ego development was found, although the restricted range of ego development range again precludes ruling out a curvilinear relationship. We expect that the association between ego development and psychopathology in substance abusing adults is positive rather than curvilinear. The greater capacity for guilt about personal transgressions (e.g., hurting others and breaking rules) at the Post-Conformist Stage of ego development likely increases the likelihood that substance-abusing adults will experience depression or anxiety in relation to their addictive behaviors.

Ego development and psychopathology in substance abusing mothers

Findings from previous studies with substance abusing mothers suggest that higher levels of maternal ego development predispose them to symptoms of depression and anxiety, perhaps because of marked discrepancies in their self-aspirations as parents and their addictive behaviors. In a demographically diverse sample of 91 mothers that included opiate-dependent and non-drug-using women, Luthar and colleagues (2001) found that associations between ego development and mothers’ emotional responses to the parenting role depended upon whether or not a mother was opiate dependent. Whereas ego development was positively associated with maternal satisfaction and negatively associated with maternal parenting distress in non-opiate-dependent women, the reverse was true for opiate-dependent women (maternal ego development was negatively associated with maternal satisfaction and positively associated with maternal parenting distress). However, a curvilinear relationship between ego development and emotional response to parenting among opiate addicted women could not be ruled out because the highest level of ego development represented in the opiate-dependent sample was Self-Awareness (E5).

Ego development and parenting in substance abusing mothers

Previous studies suggest that higher levels of ego development in substance abusing mothers may confer advantage for their parenting practices. In a sample of 74 substance abusing mothers of infants, Fineman and colleagues (1997) found a positive association between maternal ego development and maternal sensitivity during interactions with infants (r = .35, p < .001). In post hoc analyses, the authors found that mothers at the Impulsive Level (E2), the lowest level in the sample, differed significantly in maternal sensitivity from mothers at the highest level found in the sample, the Self-Awareness Level (E5), but there were no other group differences. The authors also examined relations of intelligence, ego development, and maternal sensitivity and determined that, although intelligence and ego development had considerable overlap (r = .41, p < .001), only ego development was correlated with maternal sensitivity.

In another study with 361 opiate-dependent and non-drug-using mothers, Zelazo (2006) reported that opiate-dependent mothers at Pre-Conformist levels were at greater risk for maladaptive parenting (aggression, neglect, rejection, and withholding warmth) and limit setting problems than non-opiate-dependent mothers. Zelazo controlled for maternal intelligence and found that ego development effects were above and beyond intelligence effects and that intelligence did not predict parenting factors.

Study Aims

The primary aim of the current study was to examine within the same sample of substance abusing mothers whether mothers at higher levels of ego development exhibited more symptoms of depression and anxiety and fewer parenting deficits (e.g., aggression, neglect, and lack of warmth) than mothers at lower levels of ego development. A secondary aim was to determine whether maternal intelligence is a related but distinct construct in relation to maternal ego development. We expected that maternal intelligence would increase across levels of ego development, but that maternal ego development would still be associated with higher levels of psychopathology and fewer parenting deficits after maternal intelligence was taken into account.

Method

Overview of Procedures

Data used in this study were collected during baseline assessments of methadone-maintained mothers who expressed interest in a clinical trial testing the efficacy of a new parenting intervention called the Relational Mothers’ Parenting Group (RPMG; for a full report on the randomized clinical trial study, see Luthar, Suchman, & Altomare, 2007). Opiate-addicted mothers interested in participating in parenting groups were recruited at three methadone clinics in New Haven, CT. Recruitment occurred via referrals by counselors, visits made by research assistants to counseling groups and medication lines, and referrals from mothers who had already participated in the study. To be eligible for inclusion, mothers had to (a) have at least one child less than 16 years of age in their care, and (b) report problems with parenting. Exclusion criteria included conditions that would impede ability to benefit from group therapy, such as cognitive deficits, psychotic thought processes, suicidality, and homicidality.

All eligible mothers who expressed interest in the study met with a research assistant who explained the nature of the study as a randomized trial and completed consent procedures with mothers. Each mother was asked to select one child to be the focus of her work in the intervention and assessments. Initial assessments were scheduled with mothers. After mothers completed the baseline assessment, they were scheduled for a second meeting during which they were randomized to either RPMG or to Recovery Training (RT), a comparison condition.

The RPMG and RT conditions each entailed weekly group meetings in addition to standard treatment at the clinic. Mothers were enrolled in their respective interventions for 24 weeks and in the study for one year. Mothers and children completed assessments about the mothers’ parenting, and mothers’ and children’s behavioral and psychological adjustment seven times during the year at 8-week intervals (Weeks 0, 8, 16, 24, 32, 40, and 48). To compensate mothers for time spent in assessments, a staggered reimbursement schedule was used, such that mothers were paid $20 at the baseline visit, $25 at Weeks 8, 16, 24 and 32, $30 at Week 40, and $40 at Week 48. Mothers received bonus payments of $5 for completing their assessments on time.

Sample

A total of 182 mothers who expressed interest in the study were screened and found eligible for the study and completed baseline assessments. Mothers in the sample were, on average, 36.26 years old, Caucasian, unemployed, had completed high school, and were caring for 1.45 children. The average age of focal children was 9.59 and approximately half were male (see Table 2).

Table 2.

Descriptive Data (n = 182)

	Mean (SD)a or percent
Maternal age	36.26 (7.14)
Ethnicity
Caucasian	57.1
African American	28.0
Hispanic/Latino	14.8
Marital status
Never married	53.3
Married or remarried	16.5
Separated or divorced	24.7
Widowed	5.5
Education
Less than high school	34.1
High school or GED	39.6
College or tech school	26.4
Unemployed	81.9
Number of children living with mother	1.45 (1.21)
Focal child age	9.59 (4.37)
Male focal child	54.4
Ego development level
Impulsive (E2)	2.7
Self-protective (E3)	28.0
Conformist (E4)	30.2
Self-awareness (E5)	35.7
Conscientious (E6)	3.3
Intelligence	92.08 (11.19)
Parenting
Aggression/hostility	25.72 (5.99)
Neglect/indifference	22.80 (5.82)
Undifferentiated rejection	15.71 (4.03)
Warmth/affection	71.37 (8.60)
Psychopathology
Somatization	54.96 (13.91)
Obsessive – compulsive	57.75 (15.5)
Interpersonal sensitivity	55.90 (14.20)
Depression	56.87 (12.95)
Anxiety	52.85 (15.05)

^aItalicized numbers represent T Scores.

Measures

Maternal ego development

The Washington University Sentence Completion Test—short form (WUSCT; Hy & Loevinger, 1996) is a semiprojective technique that consists of 18 open-ended sentences that subjects are asked to complete (the full version contains 36 items) without guidance about content or length of response. Each response is then coded by a trained rater according to a detailed scoring manual (Hy & Loevinger, 1996). Item level scores range from E2 to E9 (very rarely attained) and follow a developmental progression from Impulsive (E2) to Integrated (E9; See example in Table 1). The intermediary developmental categories are Self-Protective (E3), Conformist (E4), Self-Aware (E5), Conscientious (E6), Individualistic (E7), and Autonomous (E8). Total protocol scores (categorical scores assigned to the entire protocol) are then calculated for each subject in accordance with Hy & Loevinger (1996). For this study, protocols were rated by two trained raters who were blind to all other subject data. Independent ratings of the two raters on 10% of the protocols indicated 93.1% agreement.

Noam and colleagues (2006) conducted a comprehensive review of hundreds of studies examining the validity and reliability of the WUSCT. Generally, the test–retest reliability, interrater reliability, and internal consistency have been found to be adequate in normal and clinical populations. Ego level assessed with the WUSCT has also been found to correlate with ego level assessed by interview and other staged and projective tests of personality (Manners & Durkin, 2001). The developmental theory on which the WUSCT is based has been validated by evidence in sequential studies that ego development increases during high school and college and has been validated in studies testing asymmetry of comprehension that have demonstrated that people can understand ego levels lower than their own but not those much higher (Manners & Durkin, 2001; Noam et al., 2006). Substantial support has also been found for the unitary nature of ego development proposed by Loevinger (Manners & Durkin, 2001). In tests of discriminant validity, the WUSCT has shown moderate correlations with intelligence, fluency, and diverse socioeconomic variables suggesting that ego development is a related but separate construct (Manners & Durkin, 2001; Noam et al., 2006).

Maternal intelligence

The Kaufman Brief Intelligence Test (K-BIT; Kaufman & Kaufman, 1990) is a widely used-standardized screening measure of verbal, nonverbal, and composite intelligence with good reliability and correlations with other widely used measures of intelligence (Miller, 1995; Young, 1995). Standard scores for verbal IQ were computed, resulting in values of 70–79 indicating a “well below average” score; 80–89 indicating a “below average” score; 90–109 indicating an “average” score; 110–119 indicating an “above average” score; 120–129 indicating a “well above average” score; 130–160 indicating a score in the “upper extreme.” in this sample, cronbach’s alpha coefficient for maternal intelligence was .95.

Maternal psychopathology

The Brief Symptom Inventory (BSI; Derogatis, 1983) is a widely used 53-item self-report measure of psychopathology that yields scores on nine primary symptom dimensions, including the Somatization, Obsessive Compulsive, Interpersonal Sensitivity, Depression, Anxiety, Hostility, Paranoid Ideation, Psychoticism, and Phobic Anxiety dimensions. Four scales, including Somatization, Interpersonal Sensitivity, Depression, and Anxiety, were used in this study. Each of the scales has demonstrated a high degree of internal consistency with similar populations (Benishek, Hayes, Bieschke, & Stoffelmayr, 1998; Boulet & Boss, 1991). Cronbach’s alpha coefficients in this sample were .62 for Somatization, .70 for Interpersonal Sensitivity, .81 for Depression, and .85 for Anxiety. Although studies examining the factor structure have been inconclusive, they have shown the utility of the scales as a measure of global distress (Beneshik et al., 1998; Boulet & Boss, 1991).

Maternal parenting

The Parental Acceptance-Rejection Questionnaire (PARQ; Rohner, 1991) is a 60-item self-report measure rated on a 4-point scale. Three subscales measure mal-adaptive parenting, including the Aggression/Hostility Scale (with high scores indicating more use of verbal and physical aggression), the Neglect/Indifference Scale (with higher scores indicating less attention to children’s needs), and the Undifferentiated Rejection Scale (with higher scores indicating emotional coolness toward the child). The fourth subscale that measures adaptive parenting is the Warmth/Affection Scale (with higher scores indicating more expressed warmth and interest). Adequate internal consistency has been documented for the PARQ subscales in similar populations (Khaleque & Rohner, 2002; McGuire & Earls, 1993). In this sample of mothers, coefficient alphas for the four subscales were .73 for Aggression/Hostility, .79 for Neglect/Indifference, .62 for Undifferentiated Rejection, and .89 for Warmth/Affection. Because the Undifferentiated Rejection coefficient alpha was below .70, this subscale was not included in data analyses.

Data Analysis

Preliminary analyses

Because sociodemographic factors (e.g., child age, maternal education, and employment, etc.) have been found to be related to parenting and psychiatric status in prior investigations (Suchman & Luthar, 2000), these associations were initially examined using Pearson’s Product Moment Correlation Coefficient. Sociodemographic variables that were associated with at least two parenting or psychiatric factors were included as covariates in all subsequent analyses.

To determine whether maternal verbal intelligence covaried with maternal ego development, an analysis of variance was first conducted testing for a positive linear trend in verbal intelligence across levels of ego development. The significance of the trend determined whether or not maternal intelligence would be included as a covariate in subsequent analyses.

Multivariate analyses of variance

In two separate multivariate analyses of variance (MANOVAs), after controlling for confounding sociodemographic factors and maternal verbal intelligence, parenting variables and psychiatric variables were entered, respectively, as dependent variables and level of maternal ego development served as the independent variable. Because only one direction of association was predicted for each MANOVA, a one-tailed significance test was used. For models that indicated significant overall effects, univariate effects and predicted trends for each domain of parenting (e.g., aggression, neglect, warmth) and psychopathology (e.g., depression anxiety, etc.) across levels of ego development were then tested using a one-tailed significance test. Although fewer than the ideal number of 20 cases per cell were expected in cells for the lowest and highest levels of ego development, a minimum of five subjects per cell (given at least three cells) was expected to yield enough power to detect moderate effects (e.g., .40) and also insured that the number of cases per cell was greater than the number of dependent variables (Tabachnick & Fidell, 1996).

Results

Descriptive Data

As shown in Table 2, the modal level of ego development was the Self-Awareness Stage (E5). Large representation of the Conformist (E4) and Self-Protective (E3) stages were also evident. The average group scores for parenting were all within normal levels, although the Warmth/Affection mean was at the clinical cut-off, suggesting that many mothers in the group were at risk for with-holding warmth and affection from their children (see Table 4 for clinical cut-off scores). The average group T-Scores for maternal intelligence and psychopathology were all within normal limits.

Table 4.

Results of Univariate and Multivariate Analyses of Variance Comparing Means for Maternal Verbal Intelligence, Parenting and Psychopathology Across Levels of Ego Development (n = 182)

	Ego development (TPR) level Meana (SD)					Between-group effects		Polynomial contrast
Construct	Impulsive (E2)	Self-protective (E3)	Conformist (E4)	Self-aware (E5)	Conscientious (E6)	F	df	Linear contrast estimate (SE)	p	Clinical cut-off score
Intelligence
Verbal IQ	86.40a (6.58)	86.69 (8.03)	92.62 (8.24)	97.03 (7.94)	103.50 (9.14)	15.23***	4,177	14.09 (3.13)	.000	<80
Psychopathology
Somatization	41.00a (0.00)	51.92 (13.72)	58.20 (13.51)	55.40 (14.10)	58.00 (13.73)	3.17**	4,175	14.11 (5.60)	.000	>63
Interpersonal	41.00 (0.00)	53.27 (14.18)	57.33 (13.93)	56.95 (14.11)	66.17 (13.66)	2.87**	4,175	18.27 (5.67)	.001	>63
Depression	46.20 (9.39)	52.76 (13.35)	58.15 (12.34)	59.00 (12.32)	66.00 (13.10)	2.92**	4,175	14.75 (5.18)	.003	>63
Anxiety	43.00 (11.18)	49.35 (14.62)	53.95 (15.35)	55.03 (14.99)	57.00 (15.05)	2.16*	4,175	13.07 (6.07)	.015	>63
Parenting
Aggression/hostility	31.80 (9.09)	25.07 (5.52)	25.67 (6.06)	25.80 (6.09)	25.72 (5.99)	1.46†	4,175	−3.65 (2.42)	.065	>35
Neglect/indifference	31.60 (7.64)	23.35 (5.83)	22.09 (5.63)	22.51 (5.50)	20.33 (3.98)	3.43***	4,175	−7.46 (2.28)	.000	>27
Warmth/affection	61.60 (7.20)	69.65 (8.90)	72.16 (8.05)	72.42 (8.57)	75.50 (6.25)	2.49*	4,175	9.74 (3.45)	.001	<71

^aScores in italics represent T Scores

^†p < .10

^*p < .05

^**p < .01

^***p < .001, one-tailed.

Covariates

Correlations among all variables are reported in Table 3. One sociodemographic factor—focal child’s age - was significantly correlated with more than one parenting or psychiatric variable and was therefore treated as a covariate in subsequent analyses. As shown in Table 4, maternal verbal intelligence demonstrated a significant positive linear trend across levels of ego development and was therefore treated as a covariate in subsequent analyses as well.

Table 3.

Pearson’s Product Moment Correlation Coefficient for All Variables

	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
1. Age
2. High school	−.02
3. Married	−.06	−.02
4. Unemployed	−.02	.04	.06
5. Family size	−.10	−.21**	.10	−.06
6. Child age	.38**	−.09	−.01	−.04	.10
7. Child gender	.12	−.02	−.02	−.17*	.10	.07
8. Verbal IQ	.06	.44**	.02	.16*	−.05	−.19**	.01
9. Ego development	.01	.35**	.05	.06	−.18*	.01	−.02	.50**
10. Aggression	.04	.05	.02	.06	−.05	.17*	.00	−.03	−.02
11. Neglect	.10	−.06	−.11	−.04	−.04	.21**	.00	−.10	−.17*	.56**
12. Warmth	−.06	.14	.05	.05	−.02	−.11	−.05	.11	.21**	−.42**	−.75**
13. Somatization	.00	−.02	−.09	−.08	−.11	−.02	.10	.00	.15**	.02	−.14	.12
14. Interpers. distress	−.18*	.00	−.08	−.03	−.09	−.14	.01	.08	.20**	.01	−.15	.22**	.55**
15. Depression	−.03	.07	−.09	.00	−.14	−.07	.01	.13	.26**	.05	−.15*	.23**	.60**	.70**
16. Anxiety	−.07	−.03	−.03	−.06	−.15	.06	.06	.04	.18*	.09	−.13	.14	.72**	.63**	.69**

^*p < .05

^**p < .01.

Maternal Ego Development and Psychopathology

Results of multivariate analyses are reported in Table 4. A significant overall model effect, F = 1.50, p < .05, after controlling for the focal child age and maternal verbal intelligence, was found for the MANOVA investigating associations between maternal ego development and psychopathology. Significant between-groups differences across levels of maternal ego development were found for somatization, interpersonal sensitivity, and depression. Marginally significant group differences were found for maternal anxiety. Polynomial contrasts confirmed a significant positive linear trend in each of the four psychiatric variables across levels of ego development. Also shown in Table 4, mean scores for Interpersonal Sensitivity and Depression for mothers at the Conscientious Stage (E6) fell beyond the clinical T-Score cut-off of 63 (A T-Score of 63 and higher is considered a positive diagnosis).

Maternal Ego Development and Parenting

Results of the multivariate analyses of variance (see Table 4) indicated a significant overall model effect, F = 1.54, p < .05, after controlling for the focal child’s age and maternal verbal intelligence. Between-groups differences across levels of maternal ego development were significant for parental neglect and parental warmth but not parental aggression. Polynomial contrasts confirmed a significant negative trend in maternal neglect across levels of ego development and a significant positive trend in maternal warmth across levels of ego development. As shown in Table 4, mean scores for Neglect among mothers at the Impulsive Stage (E2) fell above the clinical cut-off score of 27 indicating risk for child maltreatment. The mean Warmth score for mothers at the Impulsive (E2) and Self-Protective (E3) Stages of ego development was below the clinical cut-off score of 71, indicating risk for child maltreatment. Mean scores for Aggression were within the normal range for mothers across all levels of ego development.

Discussion

Building on findings of previous investigations, the aims of this study were to examine, within a sample of opiate abusing mothers of minor children, whether higher levels of maternal ego development were associated with higher risk for psychopathology and lower risk for parenting problems. A second aim involved examining whether ego development was a related but distinct construct from verbal intelligence, particularly in relation to parenting. Findings and limitations pertaining to each of these aims are discussed below.

Descriptive Data

Previous studies examining ego development among opiate using adults (e.g., Blatt et al., 1984; Wilber et al., 1982) have reported the modal level of ego development to be the Self-Protective (E3) Stage, although a significant percentage have fallen at the Conformist (E4) and Self-Awareness (E5) Stages. Studies examining ego development among opiate using mothers have reported similar or higher modal levels (e.g., Fineman et al., 1997; Luthar et al., 2001). In this study, the Self-Awareness (E5) Stage was found to be the modal level. Loevinger (Hy & Loevinger, 1996) has explained this level as containing its own aspects of conformity and characteristic experiences of loneliness and self-consciousness as the ability to conceptualize an inner life in terms of feelings develops. Mothers who enter treatment for their substance abuse often report their parental responsibilities and concerns about their children’s well-being as motivating factors for seeking help (DHHS, 1999). It is conceivable that a growing internal dissonance about competing social roles, awareness of alternative lifestyles, and increasing feelings of guilt and inadequacy may be driving their decisions to enter treatment.

Maternal Ego Development and Psychopathology

Although previous studies with opiate users have shown positive associations between ego development and psychopathology (Luthar et al., 2001; Wilber et al., 1982; Zelazo, 2006) results have been inconclusive, because they were based on a narrow range of ego development levels that did not include the Conscientious (E6) Stage. Results of studies conducted with non-substance-abusing adults have shown curvilinear associations between ego development and psychopathology. In this study, which included mothers at the Conscientious (E6) Stage, a linear association with dysphoric symptoms was predicted and results supported this prediction. For women at the Conscientious Stage whose substance use has placed their children at risk, higher levels of psychopathology might be explained by a greater awareness of inner feeling states and guilt about hurting other persons.

Maternal Ego Development and Parenting

Previous studies examining maternal ego development and parenting in opiate abusing women (Fineman et al., 1997; Luthar et al., 2001; Zelazo, 2006) have found that mothers at lower (e.g., Pre-Conformist) levels of ego development are at greatest risk for maladaptive parenting (e.g., aggression, neglect, inconsistent limit setting, absence of warmth, and maternal insensitivity). Similar findings were predicted in this study, given predilections at Pre-Conformist levels of ego development for dichotomous perceptions of others as “good” or “bad,” limited conceptions of causation, and limited self-control. Findings from this study are consistent with previous studies examining ego development and parenting in samples of opiate abusing mothers (Zelazo, 2006). Mothers at the Impulsive (E2) Stage of ego development reported (a) significantly higher levels of neglect and lower levels of warmth toward their children than mothers at higher levels of ego development, and (b) clinically significant levels of neglect and withholding of warmth. It is conceivable that mothers at the Impulsive (E2) Stage of ego development, who are significantly more likely to neglect their children’s physical and emotional needs, reject their children’s bids for attention and relatedness, maintain an emotional distance, evaluate their children’s behavior as they might evaluate their own, in relatively simple terms, as “right” and reward-worthy or “wrong” and punishable with an absence of recognition of underlying motivations or emotional needs. With a limited recognition of causation and consequences, a mother at the Impulsive Stage might also view her child’s behavior as erratic and unpredictable (and therefore more punishable) rather than having predictable and sequential patterns.

It is important to note that the findings from this study are not consistent with a large body of research on the deleterious effects of maternal psychopathology on parenting and child outcomes (see Goodman & Gotlib, 1999; Lovejoy, Graczyk, O’Hare, & Neuman, 2000; Seifer, 2003). In this study, although mothers at the Conscientious Stage (E6) experienced more psychopathology, they were also reporting more optimal parenting. It may be that mothers with higher levels of ego functioning have a greater capacity to regulate and contain their psychopathology during interactions with children. In other words, their ego functioning may serve to prevent or limit the impact of psychopathology on their children in comparison with mothers at lower levels of ego development. In another related study (Suchman, Borelli, & David, 2007), higher levels of ego development in the same sample of mothers were found associated with higher levels of coping and readiness to change in therapy. Although they experience higher levels of psychopathology, these mothers also seem better able to actively address or cope with their psychopathology, which may also improve their interactions with children.

Ego Development and Intelligence

Findings in this study were consistent with previous work (Fineman et al., 1997; Zelazo, 2006) showing that, whereas maternal intelligence may vary across levels of ego development, it does not explain or serve as a proxy for associations between maternal ego development and psychopathology or parenting, respectively. It is possible that specific levels of maternal intelligence may be prerequisite to the attainment of specific levels of ego development. A capacity for abstract thinking, for example, may be a necessary prerequisite to recognizing group norms. Likewise, a capacity for metacognition may be prerequisite to recognizing inner states of oneself and others.

Study Limitations

The data used in this study were collected at a single time-point (baseline) and findings confirm correlation (but not causation) between ego development and other variables. Although ego development is often considered a stable trait during adult years, it is possible that fluctuations occur once an individual enters treatment (as all the women in this sample have done). Higher levels of psychopathology may have caused mothers to enter treatment where new insights led to upward shifts in ego development. Prospective research following mothers across a course of treatment would help clarify temporal associations between ego development and psychopathology and parenting.

The parenting data in this study represent mothers’ perceptions of their own behavior that may not correspond to objective observations by trained professionals. In general, parent self-reports tend to present parenting behavior in a more positive light than collateral views of parenting. This bias would either influence self-reports across all levels of ego development or increase the likelihood that mothers at lower levels of ego development would underreport behavioral problems in order to protect themselves from legal problems. In this study, the former scenario is more likely (since mothers at lower levels of ego development were more likely to report parenting problems).

Implications for Intervention Development

Given that the Self-Awareness Stage (E4) is the modal level of ego development among substance abusing mothers in treatment, it may be worth considering (during initial treatment planning) whether focusing on conflicting internal emotional states might be more effective than behavioral and cognitive treatments for substance abusing women who are at higher levels of ego functioning. Whereas a focus on concrete changes in cognition and behavior and supplemental psychopharmacological therapy have been the modal treatment for addressing comorbid psychopathology among patients in substance abuse treatment, these approaches may be less effective for mothers whose psychiatric distress may be precipitated by internal struggles about competing roles, and interpersonal and emotional needs. Insight-oriented therapies for substance abusing mothers that focus on strengthening maternal capacities for introspection (Luthar & Suchman, 2000; Luthar et al., 2007) and recognizing underlying emotional needs (Pajulo, Suchman, Kalland, & Mayes, 2006; Suchman, DeCoste, Castiglioni, Legow, & Mayes, in press; Suchman, Pajulo, DeCoste, & Mayes, 2006) may be more beneficial to women entering treatment with higher levels of ego functioning.

It may also be worthwhile to use a developmental understanding of parental ego functioning to design parenting interventions that create conditions that promote parental maturation from one level to the next. For example, exploring conventional norms for parenting and drug use with parents at the Impulsive Stage (E2) might encourage parents to consider the pros and cons of social standards. Similarly, exploring competing authorities on parenting and drug use with parents at the Conformist Stages (E3 and E4) might promote recognition that meeting demands of all authorities is impossible and encourage exploration of more personally meaningful guidelines for personal and parenting decisions. At higher levels of ego development (e.g., the Self-Awareness (E5), Conscientious (E6), and higher levels), a focus on recognizing and contextualizing painful affect related to parenting might help to promote greater psychological resilience.