Psychosocial Functioning Among Children With and Without Differences of Sex Development

Abstract

Objective

 To assess health-related quality of life (HRQoL), psychosocial adjustment, and family functioning of children with differences of sex development (DSD) or cleft lip and/or palate (CL/P).

Methods

 In this cross-sectional study, parents of children with DSD (n = 67), CL/P (n = 121), and a comparison group of unaffected youth (n = 126) completed standardized measures assessing family functioning and their children’s HRQoL and psychosocial adjustment. Medical charts were abstracted for youth with either congenital condition.

Results

 Children with DSD were rated as having significantly lower HRQoL and greater internalizing problems compared to youth with CL/P and unaffected youth. Children in the DSD group were also significantly more likely to fall into the clinical risk categories for total and internalizing problems relative to the CL/P and unaffected groups. Caregivers of children with DSD were significantly more likely to endorse items about child suicidality compared with caregivers in the CL/P and unaffected groups. No significant differences were found between groups for externalizing problems or the expressiveness domain of family functioning; parents of children with DSD reported significantly less family conflict relative to the other groups and greater cohesion relative to the unaffected group. Conclusions Youth with DSD appear to be at greater risk for psychosocial problems relative to children with CL/P and unaffected peers. Results underscore the need for integrated interdisciplinary care and ongoing psychosocial risk monitoring in youth with DSD.

Introduction

Structural and/or functional physical differences are estimated to affect 3% of all births in the U.S. (Centers for Disease Control and Prevention [CDC], 2008) and may be readily apparent at or before birth or may only become evident over time. Cleft lip with or without cleft palate (CL/P) results from incomplete closure of the lip and/or roof of the mouth during early prenatal development and is estimated to affect one out of every 940 live births, making it the second most common birth defect in the U.S. (Parker et al., 2010). CL/P is associated with speech, hearing, dental problems, and often some degree of residual appearance differences even after corrective surgeries. In contrast, differences of sex development (DSD), which involve variation in genetic, gonadal, or phenotypic sex, are significantly more rare, affecting an estimated one out of every 4,500 live births, with prevalence rates varying between diagnoses (Hughes et al., 2006). DSD can pose a range of challenges for affected children and their families, including diagnostic uncertainties, decisions about selecting a gender of rearing, when and how to share information with other people, and whether or not (and when) to pursue gonadectomy or genital surgery—decisions which can have implications for future fertility and urogenital functioning (Wisniewski, 2017).

Both CL/P and DSD typically involve surgical management and associated care during childhood and adolescence to improve both the function of the affected structures and to make the appearance more typical (American Cleft Palate-Craniofacial Association [ACPA], 2018; Hughes et al., 2006). While these conditions are usually not life-threatening, they can have a profound impact on physical and psychosocial functioning and quality of life for affected children and their families (Bennecke et al., 2017; de Vries et al., 2019; Stock & Feragen, 2016). Both conditions also involve physical differences and related concerns about social stigmatization. However, DSD often impact genital appearance, with potential implications for gender dysphoria, sexual function, and fertility. These issues may convey additional risks for stigma and psychosocial distress for families of children with DSD. To date, few contemporary studies (Oner et al., 2009) have compared youth with DSD to other pediatric populations with other congenital conditions. Comparing psychosocial functioning across similar conditions with respect to onset, course, treatment (Rolland, 1987) and psychosocial risk factors can be helpful in conceptualizing areas of risk and resiliency and offer implications for clinical care, particularly for understudied conditions like DSD.

Over the past several decades, a large body of research has developed about the psychosocial impact of CL/P for affected youth, likely due to the higher prevalence rate of these conditions relative to DSD and early adoption of an interdisciplinary care model for their treatment (ACPA, 2018). Studies of youth with CL/P have documented risks for emotional and behavioral problems including anxiety and depression; body image concerns; teasing and bullying; and impairments in health-related quality of life (HRQoL) (Broder et al., 2014; Crerand et al., 2017; Stock & Feragen, 2016). However, there is also growing evidence that many children with CL/P exhibit healthy levels of adjustment and quality of life (Stock & Feragen, 2016), highlighting the need to further study these outcomes using appropriate comparison groups and measures that are normed.

In contrast to CL/P, there has been little empirical evaluation of psychosocial functioning in youth with DSD, likely due to the rarity and heterogeneity of DSD conditions. Most studies have focused on gender identity and psychosexual development (Sandberg et al., 2012). While prior studies have focused on HRQoL or adjustment in specific DSD conditions (e.g., proximal hypospadias, Schonbucher et al., 2007; congenital adrenal hyperplasia [CAH], Oner et al., 2009), few studies have evaluated HRQoL or adjustment (Kleinemeier et al., 2010) among a wider range of DSD conditions nor have these aspects been studied across the developmental spectrum. In addition, limited work has explored psychosocial functioning in comparison to typically developing youth or those with other congenital conditions (Oner et al., 2009). Of significant concern, studies of psychosocial functioning in adults with DSD document high rates of suicidality (de Vries et al., 2019; Schützmann et al., 2009), psychosocial distress, and impaired sexual functioning (de Vries et al., 2019; Kleinemeier et al., 2010). Such findings underscore the need to evaluate psychosocial risks in youth with DSD to improve care and to prevent poor outcomes.

Despite the importance of healthy relationships in supporting well-being and adaptive development among children with pediatric conditions (Herzer et al., 2010), family functioning has received limited attention in youth with DSD or CL/P. Because DSD can affect sexual and reproductive functioning, parents may be uncomfortable discussing condition-related information with their child. Such dynamics may negatively impact family communication or lead to conflict if caregivers disagree about information-sharing with the child and/or others and about health care-related decision-making. However, to our knowledge, no studies have examined family functioning in pediatric DSD samples. Two of the few studies to evaluate family functioning in youth with CL/P reported comparable results, namely high levels of cohesion and expressiveness and low levels of conflict on average (Crerand et al., 2015) and greater cohesion, expressiveness, and lower conflict compared to families of unaffected youth (Wyszynski et al., 2005).

While interdisciplinary care that includes mental health services has been recommended for both DSD and CL/P (ACPA, 2018; Hughes et al., 2006; Sandberg et al., 2017), there is considerable variation in how these services are rendered, and many youth have inadequate access to psychosocial care (Ernst, Liao, et al., 2018). Because families look to their medical teams to provide anticipatory guidance and psychoeducation on expected psychosocial outcomes, a better understanding of psychosocial risks in children and families with DSD or CL/P is needed. Such information can also inform the timing and provision of psychoeducation, support, interventions, and advocacy for integrated mental health services (Ernst, Liao, et al., 2018).

To address these gaps, this cross-sectional study was designed to obtain preliminary data about parent-rated HRQoL, emotional and behavioral adjustment, and family functioning among youth with DSD, youth with CL/P, and typically developing, unaffected youth. The hypothesis of this study was that youth with DSD and their families would be at greater risk than CL/P youth or unaffected youth for several reasons. DSD are rarer conditions which are not always visibly or readily apparent to others (unlike CL/P and associated speech/appearance differences), which may predispose families to hide the condition from both the child and others due to shame or fear of stigmatization. Furthermore, DSD treatments are often more complex, and DSD conditions often carry infertility risks and potential implications for gender that can pose decisional challenges for parents (Wisniewski, 2017). In addition, it can be challenging for parents of children with DSD to explain a child’s medical issues, gender changes and other factors to siblings, extended family and friends, issues that are typically not present for CL/P. Finally, as described above, studies of adults with DSD suggest heightened psychosocial risks including elevated suicidality. For these reasons, we hypothesized that parents of youth with DSD would report lower HRQoL, greater emotional and behavioral risks, and poorer family functioning (e.g., greater conflict, lower cohesion, and expressiveness) relative to the CL/P and unaffected comparison groups.

Methods

Participants

All participants were parents and/or legal guardians of children 0 to 19 years old, were proficient in English and at least 18 years old, and had no impairments that would impact questionnaire completion. Participants were primarily recruited from one pediatric academic medical center; eight parents from the DSD group were recruited from a second pediatric academic medical center.

Parents of Youth With DSD (DSD Group)

Caregivers of children with DSD conditions receiving endocrinology or urology care at two pediatric academic medical centers with interdisciplinary DSD clinics were invited to participate. At both sites, participants were recruited prospectively as they presented for their children’s appointments; additionally, at the primary site, participants were also recruited retrospectively (previously seen from 2006 to 2018). Parents were included if their child had a DSD diagnosis involving atypical genital appearance or discordance between genetic sex and internal or external reproductive structures. Parents were excluded if their child had (a) presence of a life-threatening or other significant comorbid condition (e.g., cancer, uncontrolled diabetes) that could impact psychosocial functioning independent of the DSD or CL/P; or (b) genetic disorders affecting sex chromosomes that have known comorbid physical, medical, and neurocognitive differences (e.g., Klinefelter syndrome, Turner syndrome). Children (197) with DSD were identified across both sites (16 from BCH); 44 did not meet eligibility criteria after screening (including one from BCH). Thirty-four (including six from BCH) were eligible but not enrolled due to missing appointments or not being able to be contacted. Sixteen actively declined participation, and two withdrew after enrolling (all from NCH). Overall, 101 families were enrolled; 9 completed surveys from BCH and 58 from NCH for a total of 67 complete surveys.

Parents of Youth With CL/P (CL/P Group)

Parents of patients with CL/P at the primary pediatric hospital were recruited as a part of a larger questionnaire validation study for children with craniofacial conditions (Crerand et al., 2018). For this study, parents were excluded if their child had syndromic CL/P or complex craniofacial conditions (e.g., Apert syndrome) given risks for other physical, psychosocial, and/or neurocognitive problems (Feragen & Stock, 2014). In the larger craniofacial study, 653 caregivers were consecutively identified from clinic schedules. Thirty-nine were unable to be reached due to incorrect phone numbers or addresses; 62 actively declined participation. Five hundred and fifty-two expressed interest in participation and provided informed consent; of these enrolled caregivers, 44% (n = 242 caregivers) returned completed surveys. For this study, 121 parents of children with CL/P were included in analyses.

Parents of Unaffected Children (Comparison Group)

Parents who were employees at the primary pediatric hospital were invited to participate in the comparison group arm of the DSD study described above. Parents were excluded if their child had current or a history of life-threatening or chronic physical or mental health conditions that could affect family and psychosocial functioning (e.g., congenital heart disease, bipolar disorder). For the comparison group, 257 caregivers responded to advertisements. Of these, one caregiver was ineligible, and 40 were contacted but did not return calls. Eighty-three were eligible but were not contacted due to age (we attempted to match on age such that at least one comparison group participant would match the age of at least one DSD participant). Ultimately, 133 eligible parents agreed to participate and 126 returned the survey.

Procedures

Caregivers of patients from both clinical groups were recruited through letters, phone calls, and/or in-person visits during clinic appointments. Caregivers of healthy children were recruited via email advertisements sent to a hospital employee listserv. A research assistant screened participants in all groups for eligibility and obtained verbal informed consent. All surveys were completed either via secure online surveys or paper versions of questionnaires (according to caregiver preference). Items on the child behavior checklist (CBCL) assessing suicidality were reviewed upon survey completion or receipt; if any safety items on the CBCL were endorsed, clinical follow-up with the family was promptly conducted by one of the study PIs (both of whom are pediatric psychologists). Caregivers who returned completed surveys were given a monetary incentive ($25). Study procedures were approved by the hospital Institutional Review Board.

Measures

CBCL/Adult Behavior Checklist

The CBCL/adult behavior checklist (ABCL) was administered to caregivers of children ages 1.5 years and older to assess total, internalizing, and externalizing problems. These parent-report measures are widely-used, normed, reliable, and valid (Achenbach & Rescorla, 2001). For the CBCL, internal consistency was good to excellent for the total problem scores across groups (DSD: α  =  .94; CL/P: α  =  .94; Comparison group: α  =  .88); for the ABCL total problems scale, internal consistency fell within the same ranges (DSD: α  =  .87; Comparison: α  =  .98; no caregivers in CL/P group completed ABCL).

Family Environment Scale

The Family Environment Scale (FES) is a reliable, valid, and normed measure of family social and environmental characteristics (Moos & Moos, 2002). All caregivers completed the Relationship Dimension subscales (27 items) to assess cohesion, expressiveness, and conflict. Internal consistency for each subscale across groups revealed poor to acceptable Cronbach’s alpha values: DSD (Cohesion: α  =  .18; Expressiveness: α  =  .55; Conflict: 0.60); CL/P (Cohesion: α  =  .75; Expressiveness: α  =  .52; Conflict: α  =  .61); and Comparison (Cohesion: α  =  .63; Expressiveness: α  =  .50; Conflict: α  =  .61).

Pediatric Quality of Life Inventory

The Pediatric Quality of Life Inventory (PedsQL 4.0) was completed by all caregivers to evaluate children’s HRQoL across four areas of functioning: physical, emotional, social, and school. This 23-item normed measure is psychometrically sound, and versions are available across all age ranges (infancy through young adulthood) (Varni et al., 2001). For the PedsQL Total Score, Cronbach’s alpha statistics were calculated across study groups with ranges reported for specific age versions within each diagnostic group (DSD: α  =  .82–.98; CL/P: α  = .85–.95; Comparison group: α  =  .73–.86); all values fell within the acceptable to excellent range.

Demographic Questions

All caregivers provided their age, race, ethnicity, sex, level of education, employment status, occupation, and marital status.

Medical Chart Review

Chart reviews for all patients with DSD or CL/P were completed by trained research assistants; information about diagnosis, medical history, insurance type, and receipt of social work and/or psychology consultation were abstracted.

Statistical Analysis

Data were managed using Research Electronic Data Capture (Harris et al., 2009). Analyses were run in SPSS 24.0 (Armonk, NY) and SAS 9.4 (Cary, NC). Missing data was handled in accordance with instrument guidelines or via case exclusion. Normality assumptions and equality of group variances were verified. Nonparametric statistics were used if indicated.

Descriptive statistics were calculated for both demographic and outcome measures. Demographic characteristics were first compared among the three groups (DSD, CL/P, comparison) using one-way ANOVA, Chi-square, and Fisher’s exact tests as appropriate.

Multivariate linear regression was used to examine study group (DSD vs. CL/P vs. comparison) as a predictor of continuous outcomes, controlling for child age, sex of rearing (female vs. male), race (Asian, Black, or other vs. White); and parent education (some college or less vs. college graduates), marital status (single/separated/divorced/widowed vs. married/cohabitating), employment (employed vs. unemployed), and race (Black or other vs. White) as these variables differed significantly between groups (see Table I). Stepwise forward regression was applied to select independent variables that were used to generate the final model. At each step, Akaike’s information criterion was calculated to decide whether the variable with the smallest effect should be dropped from the model. Total FES cohesion, expressiveness and conflict were log transformed due to skewness of the data. Chi-square or Fisher’s exact tests were used to compare frequencies of clinical risk status between groups on the CBCL/ABCL internalizing problems, externalizing problems, and total problems and the PedsQL physical, psychosocial, and total subscales.

Table I.

Demographics for Parents and Children

	DSD (n = 67)		CL/P (n = 121)		Unaffected (n = 126)
Child characteristics	M	SD	M	SD	M	SD	p a
Age, years	7.8	6.3	6.9	4.4	6.8	5.8	.43
	n	%	n	%	n	%
Sex of rearing							.04
Male	46	68.7%	63	52.1%	64	50.8%
Female	21	31.3%	58	47.9%	62	49.2%
Race							<.0001
White	53	79.1%	100	82.6%	108	85.7%
Black	8	11.9%	0	0%	6	4.8%
Asian	0	0%	19	15.7%	3	2.4%
Other	6	9.0%	2	1.7%	9	7.1%
Ethnicity							.84
Hispanic	1	1.5%	1	0.8%	3	2.4%
Insurance type							.14
Private	43	64.2%	77	63.6%	–	–
Medicaid	19	28.4%	40	33.1%	–	–
Self-pay	3	4.5%	0	0%	–	–
Other	2	3.0%	4	3.3%	–	–
Parent characteristics	M	SD	M	SD	M	SD
Age, years	38.2	8.5	39.1	8.5	37.1	6.98	.14
	n	%	n	%	n	%
Sex
Male	1	1.5%	14	11.6%	5	4.0%	.01
Female	66	98.5%	107	88.4%	121	96.0%
Race							.0009
White	57	89.1%	118	97.5%	115	91.3%
Black	7	10.9%	0	0%	5	4.0%
Other	0	0%	3	2.5%	6	4.8%
Highest level of education							<.0001
Less than college graduate	28	43.8%	45	37.8%	14	11.1%
College graduate or higher	36	56.3%	74	61.2%	112	88.9%
Employment
Employed full or part-time	50	78.1%	105	86.8%	114	91.2%	.04
Unemployed due to any reason	14	21.9%	16	13.2%	11	8.8%
Parent marital status							.002
Single, separated, divorced, or widowed	22	34.4%	19	15.7%	18	14.3%
Married or living with significant other	42	65.6%	102	84.3%	108	85.7%

Note. CL/P = cleft lip and/or palate; DSD = differences of sex development.

^a p-Value from one-way ANOVA, Chi-square, or Fisher’s exact test, as appropriate.

Results

Demographic and Clinical Characteristics

As shown in Table I, children’s mean ages across groups ranged from 6.8 to 7.8 years. Ages ranged from 1 month to 19 years for the DSD and healthy comparison group, and from 2 months to 17 years for the CL/P group. There were significantly more males than females in the DSD group compared to CL/P and comparison groups. Child race also differed, with higher percentages of Black children represented in the DSD group and Asian children in the CL/P group. The mean age of caregivers across groups was late 30s (37–39 years of age). There were significant differences across groups for parent race, with more parents in the DSD group identifying as Black. Education level also differed, with a greater percentage of parents from the comparison group reporting having at least graduated from college compared to parents of children with DSD or CLP. Employment status differed, with a greater percentage of families in the DSD group reporting that they were unemployed. Finally, a significantly higher percentage of caregivers of children with DSD reported being single parents compared to the other groups.

The most commonly represented DSD diagnoses in this sample were 46, XY DSD (e.g., hypospadias plus other anomalies like micropenis or bifid scrotum without an identified underlying cause), isolated severe hypospadias, and CAH (see Table II). Most youth in the CL/P group had combined cleft lip and palate.

Table II.

Child Clinical Characteristics

Children With DSD	n	%
Diagnosis
46, XY DSDa	17	25.4
Isolated severe hypospadias	15	22.4
CAH	13	19.4
Testicular regression syndrome	6	9.0
Mixed gonadal dysgenesis	5	7.5
Pure gonadal dysgenesis	4	6.0
Partial androgen insensitivity syndrome	3	4.5
Ovotesticular DSD	2	3.0
Complete androgen insensitivity syndrome	1	1.5
Mayer–Rokitansky–Küster–Hauser syndrome	1	1.5
Surgical history, yes	62	92.5
Genital surgeryb	48	72.7
Gonadal surgeryc	29	43.9
Exploratory surgery	8	12.1
	M	SD
Number of DSD-related surgeries	1.58	.97
Children With CL/P	n	%
Diagnosis
Cleft lip and palate	72	59.5%
Cleft palate only	30	24.8%
Cleft lip only	19	15.7%
	M	SD
Number of CL/P surgeries	1.84	.95

Note. CL/P = cleft lip and/or palate; DSD = differences of sex development.

^aPatients were classified to have 46, XY DSD if they had hypospadias in combination with another DSD trait (e.g., bifid scrotum, microphallus, cryptorchidism) of unknown cause.

^bGenital surgery includes feminizing genitoplasty, vaginoplasty, hypospadias repair, scrotoplasty, clitoroplasty, or perineoplasty.

^cGonadal surgery includes orchidectomy/gonadectomy, orchiopexy/orchidopexy, testicular prosthetic insertion, or inguinal hernia repair.

Health-Related Quality of Life

As shown in Table III, youth with DSD had lower mean scores on the PedsQL total, physical, and psychosocial subscales relative to both the CL/P and healthy comparison group. These differences were statistically significant even after controlling for child and caregiver demographic characteristics (see Table III). Clinical risk was also evaluated by calculating the percentages in each group whose mean PedsQL total, psychosocial and physical scores fell below clinical cut-off scores (Varni et al., 2001). As shown in Table IV, significantly greater percentages of youth with DSD fell within the clinical risk ranges across all three scores compared to both the CL/P and healthy comparison groups.

Table III.

Measure Scores by Group and Multivariate Linear Regression Results Examining Group Differences

	Group Measure Scores						Group Comparisons
	Unaffected		CL/P		DSD		DSD vs. Unaffected		CL/P vs. Unaffected		DSD vs. CL/P
Measure	M	SD	M	SD	M	SD	Adjusted Estimates	p	Adjusted Estimates	p	Adjusted Estimates	p
PedsQL total	85.56	8.26	85.40	11.76	78.27	20.34	−6.72	.002	−0.69	.71	−7.41	.001
PedsQL physical	90.59	8.34	90.26	13.05	80.28	25.23	−8.41	.001	−0.68	.74	−9.09	.0003
PedsQL psychosocial	82.33	10.54	82.43	13.12	76.84	19.30	−5.73	.01	−0.87	.45	−6.60	.005
CBCL/ABCL total problems T-score	43.59	7.58	44.80	10.28	47.41	12.25	2.67	.14	−0.90	.55	1.77	.33
CBCL/ABCL externalizing problems T-score	43.92	7.34	44.78	9.85	45.29	10.35	0.54	.75	−0.61	.67	−0.07	.97
CBCL/ABCL internalizing problems T-score	44.27	8.81	45.00	9.82	49.76	12.71	4.27	.02	−0.50	.75	3.76	.04
FES cohesion	8.17	1.29	8.07	1.55	8.31	.82	0.47	.03	−0.05	.77	0.41	.06
FES expressive	6.75	1.58	6.55	1.69	6.66	1.73	0.28	.31	0.01	.97	0.29	.30
FES conflict	1.64	1.60	1.81	1.63	1.39	1.50	−0.51	.04	−0.07	.76	−0.58	.03

Note. Child age, sex, race and parent education, race, employment, and marital status included as covariates. ABCL = adult behavior checklist; CBCL = child behavior checklist; CL/P = cleft lip and/or palate; DSD = differences of sex development; FES = Family Environment Scale; PedsQL = pediatric quality of life inventory.

Table IV.

Clinical Risk by Group

	DSD (n = 33)		CL/P (n = 68)		Unaffected (n = 61)
CBCL/ABCLa	n	(%)	n	(%)	n	(%)	p
Harms self or attempts suicide, sometimes or often	1	(3.03)	2	(2.90)	1	(1.64)	.99c
Talks about killing self, sometimes or often	4	(12.12)	1	(1.47)	0	(0.00)	.006 c

	DSD (n = 51)		CL/P (n = 106)		Unaffected (n = 95)
CBCL/ABCLb	n	(%)	n	(%)	n	(%)	p
Total problems clinical risk	4	(7.84)	3	(2.83)	0	(0)	.02c
Externalizing problems clinical risk	1	(2.0)	3	(2.8)	0	(0)	.26
Internalizing problems clinical risk	7	(13.7)	5	(4.7)	2	(2.1)	.02c

	DSD (n = 67)		CL/P (n = 120)		Unaffected (n = 126)
PedsQL	n	(%)	n	(%)	n	(%)	p
Total	11	(16.42)	9	(7.50)	0	(0)	<.0001c
Physical	11	(16.42)	6	(5.00)	1	(.79)	<.0001c
Psychosocial	13	(19.40)	11	(9.17)	10	(7.94)	.04

Note. ABCL = adult behavior checklist; CBCL = child behavior checklist; CL/P = cleft lip and/or palate; DSD = differences of sex development; PedsQL = pediatric quality of life inventory.

^aThese items were only asked to parents of children 6 years and older.

^bThe CBCL/ABCL were only completed by parents of children 1.5 years and older; clinical risk reflects percentage of participants with T-scores falling within the borderline or clinical range on these instruments.

^c p-Values reported from Fisher’s exact tests as some cells had expected counts <5.

Emotional and Behavioral Adjustment

On the ABCL/CBCL, mean scores for the total problem and externalizing scales were similar across groups, and there were no significant differences between groups when controlling for parent demographic characteristics. Scores for the DSD group were significantly higher on the internalizing subscale compared to the CL/P and comparison group (see Table III).

Clinical risk was evaluated by calculating the percentages in each group whose ABCL/CBCL score fell within the borderline or clinical range. As shown in Table IV, youth with DSD were significantly more likely to fall into these risk categories for total problems and internalizing symptoms compared to the CL/P and comparison groups. Caregivers of children with DSD were significantly more likely to endorse that their child has had suicidal ideation and self-harm behaviors compared with caregivers in the CL/P and comparison group.

Family Functioning

On the FES, parents of children with DSD reported significantly lower conflict relative to both the CL/P and comparison group and greater cohesion relative to the unaffected group (see Table III). There were no significant differences between groups for the expressiveness subscale.

Discussion

Consistent with the initial hypotheses, this study found that parents of children with DSD reported poorer HRQoL and higher levels of internalizing symptoms and suicidality for their children than either the CL/P or unaffected comparison groups. The DSD group reported similar (and on some domains, better) family functioning in comparison to the unaffected and CL/P groups.

Our findings of elevated symptoms for the psychosocial and physical HRQoL domains in youth with DSD are consistent with a European study of youth with DSD that showed impaired self-esteem, physical, and school functioning domains on the KINDL-R (Jürgensen et al., 2014) while a study in Malaysia demonstrated impaired social and school but not physical scores on the PedsQL (Selveindran et al., 2017). Total PedsQL scores obtained here are also similar to scores in Yau et al.’s (2015) cohort of children with CAH.

To date, most studies examining psychosocial functioning in DSD have focused on parent and family adjustment. Approximately a third of families of children with DSD are at risk for psychosocial problems (Ernst, Gardner, et al., 2018), consistent with risk levels in other pediatric populations (Crerand et al., 2018). The limited studies conducted on mental health in adolescents and adults with DSD have reported mixed findings regarding emotional concerns (Bennecke et al., 2017; Kleinemeier et al., 2010). Mean scores for our DSD sample on the CBCL/ABCL fell within the nonclinical range, consistent with Sandberg et al. (2017). However, in their sample, more children with DSD fell within the clinical risk ranges on the CBCL relative to our sample, which could be due to their inclusion of children with sex-chromosome aneuploidies (e.g., Turner and Klinefelter syndromes) which are associated with neurodevelopmental and psychiatric issues (Hong & Reiss, 2014).

Studies examining rates of suicidal ideation in DSD have found elevations when compared to control samples (Schweizer et al., 2017). In our sample, 12% of youth with DSD had a history of suicidal ideation, and 3% had harmed themselves or attempted suicide. de Vries et al. (2019) reported overall rates of 18.6% for suicidal ideation and 6% for attempted suicide in a large sample of adults with a range of DSD including XY-DSD and CAH. Our control sample may be an underrepresentation of suicidal ideation and behavior as compared to a larger population-based sample which suggests caregiver reports of 8.1% of children with suicidal ideation and 0.5% with a history of suicidal behavior (Janiri et al., 2020). Despite our small sample size, our findings underscore the importance of screening for safety concerns in youth with DSD. Importantly, it is likely that rates of suicidality would be higher if age-appropriate youth in our sample had been queried directly, as parents may be unaware of and underestimate their child’s suicidal ideation (DeVille et al., 2020; Jones et al., 2019).

Youth with CL/P in our sample had better HRQoL and lower risks for adjustment problems relative to the DSD group. This difference could reflect the fact that older DSD participants may not have had access to interdisciplinary care as standards and team composition have changed substantially since 2006. However, we were unable to compare receipt of interdisciplinary care across CL/P and DSD groups, and there can be inconsistencies within and across teams regarding access to and inclusion of psychosocial care. Mean scores on the CBCL/ABCL and PedsQL for the CL/P group were similar to the unaffected group and suggest that youth in the CL/P group were functioning in the normative range overall. Our results are also consistent with other studies in the CL/P literature (Collett et al., 2012). Nonetheless, this study did not assess condition-related risks (e.g., teasing, speech problems) for poor psychosocial adjustment, and youth with syndromes were excluded. Psychosocial functioning is often poorer among youth with CL/P who have other comorbid medical and/or developmental conditions (Feragen & Stock, 2014). Moreover, risks were apparent in this sample, with ∼3% having experienced suicidal ideation and/or self-harm, underscoring the need for appropriate risk screening.

Youth with CL/P also face significant stressors, including real or feared stigmatization and other challenges related to appearance and speech differences. Families of children with CL/P are also vulnerable to psychosocial problems including stigmatization, limited social support, and anxiety about their child’s treatment (Nelson et al., 2012; Nidey et al., 2016). Prolonged parental distress and poor adjustment can increase the affected child’s risk for psychosocial problems (Pope et al., 2005). While parents of youth with CL/P must cope with visible or readily apparent differences (e.g., scars, speech differences), parents of those with DSD may experience more worry and fear that their child’s often concealable physical difference or surgical history will be exposed. Little has been studied in pediatric psychology about the potential differences in how stigma is experienced. The social psychology literature suggests that internalized or concealable stigma may be more related to psychological distress and lower physical well-being than enacted or experienced stigma (Chaudoir et al., 2013). Although this study did not assess perceived stigmatization, results highlight the need to directly assess this construct in future studies and particularly in relationship to psychosocial adjustment. Furthermore, future research is needed to identify specific condition- and/or treatment-related stressors and understand how they contribute to psychosocial distress across both DSD and CL/P populations.

Notably, this study found some evidence of family functioning differences across groups, namely lower overall conflict in the DSD group relative to both the CL/P and comparison groups, and greater cohesion in the DSD group relative to the comparison group. The clinical significance of these differences is likely limited given that overall conflict scores across groups were low, and cohesion scores were high across groups and relative to norms (Moos & Moos, 2002). Furthermore, internal consistency was poor for the FES cohesion scale for the DSD group, although concerns about FES internal consistency have been raised in other studies (Boyd et al., 1997; Moos, 1990). Research has typically demonstrated similar levels of family functioning in healthy and pediatric chronic disease populations (Herzer et al., 2010). For CL/P and DSD, care is frequently episodic (e.g., staged surgical procedures), and this cross-sectional evaluation which included a wide age range of children may have obscured critical periods (e.g., at time of diagnosis) when the impact of DSD or CL/P may have a more acute impact on family functioning.

Limitations

This study has several limitations including possible selection bias in families from the two medical centers who participated. Furthermore, the unaffected sample was recruited from only one of the two sites and comprised hospital employees. Thus, there were demographic differences across groups which may account for findings observed here, although analyses controlled for demographic differences. Furthermore, parents who volunteered to participate may have other unmeasured differences which may bias their questionnaire responses (Rosenthal & Rosnow, 1975). Sample racial and ethnic diversity was not representative of the U.S., and experiences may differ based on cultural background in families of children with DSD (Ediati et al., 2016) or CL/P (Broder et al., 2012). Future studies should attempt to match DSD and comparison groups on relevant demographic variables or consider recruiting from more representative populations (e.g., families presenting for pediatric well-visits). The study is also limited by its cross-sectional design. Longitudinal data are greatly needed in these patient populations to understand their adjustment trajectories. Because of the large age range investigated, the implausibility of collecting survey data from younger children, and the fact that data from two separate studies which used the same parent-report instruments was combined to conduct these analyses, only parent-reported data was used for the present study. Parental report of their child’s adjustment is not always well-correlated with youth self-report, particularly for internalizing symptoms (Stock & Feragen, 2016). Future studies should assess both parent and youth perspectives whenever feasible.

Clinical Implications

Given this study’s limitations, our findings should be viewed as preliminary. Nonetheless, they support the need for ongoing monitoring of psychosocial risks and well-being in both congenital populations. In line with a 2018 Consensus Statement, families of youth with DSD should receive follow-up from an interdisciplinary team at regular intervals, even when the child has no significant medical needs, to ensure that families are prepared to manage the medical and psychosocial aspects of their child’s condition (Cools et al., 2018). Similarly, the ACPA Parameters of Care (2018) advocate for routine psychosocial risk screening beginning at diagnosis and continuing through adulthood. Psychosocial functioning and quality of life may fluctuate primarily during times of more intensive treatment or surgery or during developmental stages when aspects of their medical condition may pose more significant risks to well-being. Future longitudinal studies incorporating both child and caregiver perceptions are needed to evaluate trajectories of psychosocial functioning and quality of life for both groups.