Sage Journals: Discover world-class research

Abstract

Siblings of children with chronic disorders are at an increased risk of internalizing and externalizing problems, but the extent to which this is true varies depending on who rates sibling mental health. We examined interrater discrepancies between teachers and parents who rated the behavior of siblings of children with chronic disorders on the Strengths and Difficulties Questionnaire (SDQ) and investigated possible predictors of these discrepancies. We used Latent Profile Analysis (LPA) to identify classes of siblings with higher and lower levels of discrepancy between raters. Predictors of interrater discrepancy were also examined, including sibling age and sex, and the impact of the disorder. We identified four distinct classes of siblings with differing levels of interrater discrepancy and internalizing and externalizing problems. Sibling sex had a small effect on classes, with girls less likely to be included in the discrepant classes. Neither sibling age nor the impact of the disorder predicted being included in discrepant classes. Our findings reflect previous research showing that sibling experiences and outcomes tend to vary between individuals and across settings and that teachers tend to report fewer problems of different types than parents. Further research is needed to better understand predictors of these interrater discrepancies.

Keywords

siblings mental health interrater agreement/discrepancy chronic disorders teachers

Siblings of children with chronic disorders (herein referred to as “siblings”) may experience a range of experiences and outcomes. Qualitative studies have found that siblings often note both positive and negative aspects of their family situation, and a small number of quantitative studies suggest that siblings may have stronger self-esteem and social skills than their peers (Levante et al., 2025; Perenc et al., 2015; Wolff et al., 2022). Overall, however, several meta-analyses have established that having a brother or sister with a chronic disorder—such as a neurodevelopmental disability, rare genetic condition, or somatic illness—puts children at a heightened risk for internalizing and externalizing problems (Martinez et al., 2022; Pinquart, 2023; Shivers et al., 2019). This increased risk can be attributed to factors such as increased family stress, parents being occupied with additional caregiving tasks, and siblings’ complex emotions regarding their brother or sister with a chronic disorder (Fjermestad et al., 2019; Fredriksen et al., 2023; Hanvey et al., 2022). Severity of the disorder and its intrusiveness in everyday family life are known to be associated with poorer outcomes for siblings, while findings remain mixed regarding sibling age, gender, and specific chronic disorder diagnosis (Pinquart, 2023; Wolff et al., 2022).

Studies of sibling well-being mainly use parent and sibling self-reports, with less consideration of teachers’ perspectives (Shivers et al., 2019). Teacher ratings provide essential information about children’s behavior in a school setting—an environment where their behavior may be substantially different from at home (Rescorla et al., 2014). Furthermore, the school years strongly influence children’s subsequent vocational, economic, and health functioning (Davies et al., 2018; Kytties et al., 2021). Siblings may particularly benefit from the inclusion of teacher ratings in assessments, since siblings have previously reported that school can provide respite from home stresses and be a place where they experience “normality” (Gan et al., 2017; Pavlopoulou et al., 2022).

We recently reported mental health measured by the Strengths and Difficulties Questionnaire (SDQ) for siblings as rated by teachers, mothers, and fathers, and found that teachers rated siblings as having lower levels of problems than parents did, and more so for boys than for girls (Prentice et al., 2024). In the present paper, we build on this knowledge by investigating interrater discrepancies between parent and teacher SDQ scores. We aim to identify groups of siblings for whom these discrepancies are higher or lower, and explore factors that may be associated with higher levels of discrepancy.

Interrater Agreement

Interrater agreement regarding child behavior is generally low to moderate across a range of child behavior measures and cultural contexts (Carneiro et al., 2021; De Los Reyes et al., 2019). This finding is consistent in contexts as diverse as the United States (Gresham et al., 2018), Portugal (Santos et al., 2020), Estonia (Streimann et al., 2021), and Nepal (Ma et al., 2022). A meta-analysis found the mean interrater agreement to be r = .28 across a range of child behavior measures in studies using ratings by parents, teachers, and children (De Los Reyes et al., 2015). A meta-analysis of parent–teacher agreement on the Child Behavior Checklist (CBCL) Parent and Teacher Report Forms found agreement to range from r =.18 to .35 across subscales (Huang, 2017), and a cross-European study of parent–teacher agreement on the SDQ found correlations of r = .23 to .43 across subscales (Cheng et al., 2018). In studies comparing parent and teacher reports, teachers normally report fewer problematic behaviors than parents (Carneiro et al., 2021; Murray et al., 2021; Winsler & Wallace, 2002). For example, in an international comparison of parent-reported CBCL scores and corresponding Teacher Report Form scores, Rescorla et al. (2014) found that despite variance in agreement levels across 21 national contexts, parents consistently reported higher levels of problems across subscales.

Historically, these discrepancies between informants’ ratings have been seen as problematic measurement error, but in more recent decades, they have been recognized as an important opportunity to gain detailed information about how children’s behavior varies across different contexts (Achenbach, 2011; De Los Reyes & Kazdin, 2005). Home and school environments, for example, place different demands on children and provide different types of support, often resulting in divergent behavior between home and school. Parents and teachers have different roles and characteristics, and therefore provoke different responses from children. In addition, informants may have divergent perceptions of what type of behavior is problematic due to differing goals and requirements of school and home environments (Carneiro et al., 2021). Thus, consideration of multiple informants’ reports helps provide a fuller picture of children’s behavior and well-being, and careful consideration of discrepancies between raters can improve assessment and access to interventions for those who require them (De Los Reyes & Epkins, 2023).

Factors Associated With Interrater Discrepancy

A growing body of literature has examined factors that may be associated with higher and lower levels of interrater discrepancy on child behavior measures. Knowledge about such factors can help practitioners to come to more accurate clinical conclusions, target interventions, and address potential biases in the referral process. The most robust finding in this field is that interrater discrepancies are higher for internalizing problems than externalizing problems, likely because externalizing problems are more observable with less interpretation required (Achenbach et al., 1987; Cheng et al., 2018; De Los Reyes et al., 2015). Demographic factors such as sex and age of the child have been found to have effects on interrater agreement, but inconsistently (Carneiro et al., 2021; Cheng et al., 2018).

Interrater Agreement and Siblings of Children With Chronic Disorders

Siblings of children with chronic disorders exhibit varying degrees of internalizing and externalizing problems, contingent upon the rater. There are only a few studies of interrater discrepancies pertaining to siblings, which found moderate agreement (r = .52–.68) between mother and father ratings on the SDQ for siblings of children with autism spectrum disorder (Griffith et al., 2014) and nonsignificant to moderate agreement (r = .12–.57) between parent and sibling self-ratings on the SDQ prosocial subscale (Orm et al., 2022). However, no known study has examined the interrater agreement between teachers and parents of siblings of children with chronic disorders. This is pertinent because teachers have regular contact with siblings in an environment outside the home and have a wide range of other children to compare them with, so they can help to provide a more detailed picture of siblings’ behavior across multiple life contexts (Prentice et al., 2024). In previous studies, siblings have noted that they compartmentalize their lives between home and school and that being at school can provide a break from potentially stressful home situations (Brennan et al., 2013; Gan et al., 2017). Siblings have also expressed a desire to avoid worrying their parents (Deavin et al., 2018), making a multi-informant approach particularly relevant to identifying any needs for support.

LPA

Previous studies have predominantly employed variable-oriented approaches, such as correlations, to investigate interrater discrepancies in siblings’ psychological functioning (Griffith et al., 2014; Orm et al., 2022). However, these approaches often neglect the interplay between psychological symptoms, which manifest across different dimensions and are contingent on the source of the ratings (Haapasalo et al., 2000). Person-centered approaches, such as latent profile analysis (LPA), have been noted as appropriate in addressing this heterogeneity by incorporating multiple dimensions of symptoms and considering different raters’ perspectives concurrently (De Los Reyes & Epkins, 2023).

LPA, a data-driven method, identifies latent subgroups based on their observed variable patterns. This approach offers an analytical strategy for pinpointing participant groups with similar patterns concerning the variables under consideration, making it valuable for examining rating patterns from different sources regarding various dimensions of the behaviors being rated. Numerous studies have utilized LPA to explore symptom clusters reported by different informants (Aitken et al., 2018; De Los Reyes et al., 2011; Ringoot et al., 2015; Rubin et al., 2013). A smaller number of studies have previously employed LPA to investigate profile membership in parent–teacher interrater discrepancies and have shown promising results in detecting such discrepancies (Kang et al., 2023; Lerner et al., 2017). However, notably absent from the existing literature are studies examining parent–teacher discrepancies with regard to sibling mental health.

The present study draws on pre-intervention data from the SIBS-RCT, an evaluation of an intervention to improve siblings’ well-being and sibling-parent communication (Fjermestad et al., 2020). We used an LPA approach to investigate interrater discrepancies between teacher and parent SDQ scores and to explore how discrepancy is associated with factors including sex of the sibling, age of the sibling, and the sibling-reported impact the diagnosis has on their lives. We expected to find low to moderate levels of agreement on the SDQ, with higher levels of agreement for externalizing problems (De Los Reyes et al., 2015). Moreover, we expected to find higher interrater discrepancies for boy siblings, given sex-specific results of our previous study (Prentice et al., 2024) and higher discrepancies for older siblings (De Los Reyes & Kazdin, 2005).

In terms of expectations for sibling adjustment and interrater discrepancies, different hypotheses could be justified. On the one hand, poorer sibling adjustment could mean siblings display more externalizing symptoms that were more easily observable across settings, indicating it could be expected that low discrepancy was associated with poorer adjustment. However, poorer adjustment could also be associated with a pattern of high discrepancy, indicating that the sibling struggles to balance their behavior across the home-school contexts. Moreover, if the sibling is very well-adjusted, this could also be associated with a pattern of low discrepancy, with teachers and parents agreeing that the sibling shows few symptoms. Due to this potentially complex relationship between sibling adjustment and interrater discrepancy, we explored these associations openly without a priori direction expectations.

Methods

Participants

Families in the study had a child with a chronic disorder, including neurodevelopmental disorders, mental illness, and somatic diseases (Table 1). Siblings had a mean age of 10.3 years (SD = 2.0; range = 8–16 years). The mean ages of parents in the study were 42.5 years (SD = 5.5; range = 28–54) for mothers and 44.6 years (SD = 6.0; range = 31–66) for fathers. Families came from both urban and rural parts of the country. Demographic data about participating teachers was not collected.

Table 1.

Participant Background Information.

Baseline characteristics	n	%
Gender of siblings
Boys	57	45.2
Girls	68	54.4
Age of siblings
8–10 years old	74	59.2
11–13 years old	40	32.0
14–16 years old	11	8.8
Ethnicity of siblings
European	114	93.4
Asian	1	0.8
African	1	0.8
Other	6	4.9
Number of children in family
2	78	63.9
3	35	28.7
4+	9	7.4
Diagnosis of sibling with a chronic disorder^a
ADHD	31	24.8
Tourette’s syndrome	9	7.2
Asperger syndrome	18	14.4
Autism	18	14.4
Specific developmental disorder	1	.8
Emotional or conduct disorder	3	2.4
Eating disorder	5	4.0
Cerebral palsy	2	1.6
Intellectual disability	4	3.2
Down syndrome	12	9.6
Somatic diseases	2	1.6
Rare disorders^b

Note. Missing data: 3.1%–4%.

Primary, parent-reported diagnosis.

Rare disorders are defined as affecting fewer than 1 out of 2,000 people.

Procedures

Families chose to take part in a randomized controlled trial evaluating the SIBS intervention, which targets sibling well-being and parent–child communication. Families were recruited to participate via hospitals, municipal health care centers, user organizations, and a center for rare disorders. Inclusion criteria were having both a child with a diagnosis (aged 3–18 years) and one or more typically developing children aged 8 to 16 years. Siblings of the child with the disorder were excluded from participation if they had a diagnosis of any of the included disorders themselves.

Teachers were not directly recruited by the project’s research team; rather, participant parents were asked to send a copy of the SDQ to a teacher who knew their sibling child well. Participant teachers reported knowing siblings for a mean of 3.5 semesters (range = 0–12) and currently teaching the siblings for a mean of 15 hours per week (range = 0–35). They assessed themselves as knowing the target sibling “a little bit” (7.2%), “well” (62.4%), and “very well” (30.4%). A total of 125 siblings out of 288 had SDQ data from both a teacher and a parent (43.4%). This subsample was considered adequate for LPA. While larger samples (e.g., N ≈ 300–1000) are often recommended to ensure stable fit indices, acceptable model estimation can be obtained with smaller samples under certain conditions, such as when the number of indicators is limited, and class separation is strong (Marsh et al., 2009; Nylund-Gibson & Choi, 2018). Simulation studies further suggest that, although models with as few as N = 30 may be estimable in ideal cases (Nylund-Gibson & Choi, 2018), more commonly N ≥ 100–200 is preferable to ensure reliable recovery of class structure (Tein et al., 2013). Our analyses included four indicators (parent- and teacher-reported internalizing and externalizing scores). Because the number of indicators was limited, the model required fewer parameters to be estimated, reducing sample size demands and supporting the use of this subsample. Multiple fit indices were then used to evaluate model adequacy.

All participants were provided with age-appropriate information about the study and informed that they could end their involvement at any time. Participants aged 16 and older provided written consent, and parents provided written consent for children younger than 16. The local institutional board for research ethics approved the study.

Measures

The SDQ is a measure of child mental health that can be completed by parents, caregivers, and teachers, and self-rated by children and young people (Goodman, 1997; https://www.sdqinfo.org/a0.html). It has satisfactory internal consistency and test–retest reliability and has been validated across a number of languages, cultures, and contexts (Goodman, 2001; Woerner et al., 2004).

The SDQ consists of 25 items and is divided into five subscales made up of five items each. The subscales of emotional problems and peer difficulties can be combined to create an internalizing problems composite score, while conduct problems and hyperactivity-inattention can be combined to create an externalizing problems composite score. The four subscales comprising internalizing and externalizing problems are normally combined to make a total difficulties score. The final subscale, prosocial behaviors, is reported separately, with higher scores representing a strength rather than a difficulty.

The Negative Adjustment Scale (NAS) is a measure of the impact of chronic disorder on siblings via siblings’ reported adjustment to their family situation (Lobato & Kao, 2002). The NAS is a simplified version of an earlier measure, the Sibling Perception Questionnaire (Sahler & Carpenter, 1989), and in its current version has been shown to be a valid measure of sibling adjustment/maladjustment (Orm et al., 2021). This version of the NAS includes 16 items that are self-reported by the sibling, such as “I wish my parents could spend more time with me” and ‘I feel angry because of my brother or sister’s disability or illness.’

Data Analytics Plan

We analyzed the data using R Software (v4.1.2; R Core Team, 2021). We conducted LPA to estimate distinct classes related to SDQ dimensions and raters (parent vs. teacher) utilizing the TidyLPA package (Rosenberg et al., 2019). We estimated various model types, incorporating different variances and covariances, with up to five classes computed and compared using various fit indices (Ferguson et al., 2020). A good model fit was indicated by lower values for Akaike Information Criterion (AIC), adjusted Bayesian Information Criterion (aBIC), Bayesian Information Criterion (BIC), and the Consistent Akaike Information Criterion (CAIC) (Masyn, 2013). In addition, we considered an entropy value higher than .80 and significant results on the Bootstrapped Likelihood Ratio Test (BLRT) (Nylund-Gibson & Choi, 2018).

To compare latent classes, we employed chi-square tests and the Mann–Whitney U test due to the non-normal distribution of the age variable, as confirmed by the Shapiro-Wilk test. The NAS was normally distributed, and we therefore used an independent samples t-test. Bonferroni correction was used for the correction of multiple testing.

For the examination of associations between predictors (sibling NAS rating, sibling age, and sibling sex) and class membership, we utilized multiple logistic regression.

Results

Agreement Correlations

We found agreement between parents and teachers to be low for internalizing problems (r = .32) and moderate for externalizing problems (r = .51).

LPA

Statistical fit indexes indicated that, for the present data, a four-class solution with varying variances and covariances provided the best fit (Supplemental Table 1). The fit indices were close to the five-class solution with varying variances and covariances; however, the BLRT was not significant. Figure 1 illustrates the identified classes, highlighting distinct patterns in parent–teacher ratings on SDQ subscales.

Figure 1.

Mean Subscale Scores for the Strengths and Difficulties Questionnaire (SDQ) Parent and Teacher Ratings, Stratified by Class.

The first class comprised discrepant parent–teacher ratings for internalizing problems but not externalizing problems, with parents reporting higher levels of internalizing problems than teachers. This group was made up of 37 siblings (30%) and was termed the “discrepant internalised problems class.” The second class, termed the “discrepant externalized & internalized problems class,” comprised 25 siblings (20%) with consistently high parent–teacher ratings on the externalized and internalized problems subscales, but lower teacher ratings on both problem subscales. In contrast, the third class, termed the “non-discrepant externalized & internalized problems class,” consisted of 40 siblings (32%) whose parents and teachers rated both externalized and internalized problems as slightly elevated. The fourth class, consisting of 23 individuals (18%), was termed the “non-discrepant internalized problems class” and had slightly elevated parent and teacher ratings on the internalized problems subscale and lower ratings on the externalized problems subscale.

Further details about the characteristics of each class are provided in Supplemental Table 2.

Comparisons Between Classes and Logistic Regression Analysis

For the subsequent analyses, the two discrepant classes were consolidated into a single class, and the two non-discrepant classes were consolidated into a single class, as the distinctions related to discrepancy were of particular interest. Descriptive statistics for the resulting two classes are presented in Table 2.

Table 2.

Descriptives Divided by Class.

Variables	Non-discrepant classes(n = 63)	Discrepant classes(n = 62)	p
SDQ
Teacher: internalized problems, M (SD), Range	3.0 (2. 6)	2.9 (3.0)
Teacher: externalized problems, M (SD), Range	2.4 (2.3)	4.8 (3.9)
Parent: internalized problems, M (SD), Range	2.6 (2.0)	6.7 (3.3)
Parent: externalized problems, M (SD), Range	2.6 (2.1)	6.8 (3.2)
NAS
Sibling: total score, M (SD), Range	2.1 (0.5)	2.2 (.4)	.545
Sociodemographic variables
Sibling age, M (SD), Range	10.4 (2.1)	10.2 (1.9)	.718
Sibling sex female, n (%)	41 (65%)	27 (44%)	.075

Note. M = mean. n = sample size. SD = standard deviation. p = p-value from logistic regression analysis, Bonferroni corrected. SDQ = Strengths and Difficulties Questionnaire. NAS = Negative Adjustment Scale.

No significant differences were observed between the discrepant class and the non-discrepant class concerning sibling age and the NAS (Table 2). However, significant differences emerged between the two classes for the sibling sex variable (χ²(1) = 5.004, p = .025), with more girl siblings in the non-discrepant class. However, this was no longer the case when using the Bonferroni correction, corrected for three tests (p = .076).

When using the two non-discrepant classes as the reference group, class membership in the discrepant class was found to be unrelated to sibling age or NAS ratings, but significantly associated with sibling sex (Table 3) (p = .025). Specifically, girl siblings had lower odds of belonging to the discrepant class. However, again, this was also no longer the case when Bonferroni correction was applied, corrected for four tests (p = .100).

Table 3.

Predictors of the Discrepant Class Using a 2-Class Logistic Regression Model.

Effect	OR	B	SE	Wald	95% CI		p
Effect	OR	B	SE	Wald	LL	UU	p
Intercept	1.42	.35	1.38	.066	.10	21.30	.798
Sibling: total score NAS	1.32	.28	.43	.408	.57	3.08	.523
Sibling sex^a	.42	−.86	.38	5.017	.20	.90	.100
Sibling age	.95	−.05	.09	.277	.79	1.14	.598

Note. N = 109. B = unstandardized regression coefficient; OR = Odds Ratio; CI = confidence interval; LL = lower limit; UL = upper limit, p = p-value, Bonferroni corrected.

0 = male, 1 = female.

Discussion

This study investigated discrepancies between parent and teacher ratings on the SDQ for siblings of children with chronic disorders. Agreement between raters was low to moderate. Overall, teachers reported lower SDQ scores (fewer problems) than parents, and there was a higher discrepancy for internalizing problems than externalizing problems. Latent class analysis identified four distinct classes of siblings with different internalizing/externalizing profiles and levels of discrepancy between parent and teacher raters. Neither the age of the sibling nor the impact of the disorder on the sibling predicted membership in the discrepant classes. Girls were slightly less likely than boys to be included in the discrepant classes, but this difference was not significant when the Bonferroni correction was applied.

Our findings support the value of using a multi-rater approach in research and clinical work with siblings. We found agreement between parents and teachers to be low for internalizing problems (r = .32) and moderate for externalizing problems (r = .51). These findings concur with a robust literature showing that interrater agreement on measures of child behavior tends to be in the low to moderate range, and that interrater agreement tends to be lower for internalizing than externalizing problems (Cheng et al., 2018; De Los Reyes et al., 2015; Huang, 2017). The few previous studies of interrater agreement regarding sibling behavior—between parents and siblings rather than parents and teachers—have also found low to moderate levels of agreement between raters (Griffith et al., 2014; Orm et al., 2022).

Our findings specifically highlight the value of including teacher ratings when addressing sibling functioning and well-being. Consistent with research on children’s behavior generally, we found that teachers reported fewer problems for siblings than parents did (Cheng et al., 2018; De Los Reyes & Kazdin, 2005). Such discrepancies between parent and teacher reports likely reflect differing characteristics of the raters, the home and school environments, and children’s context-specific behavior related to these differences (Carneiro et al., 2021; De Los Reyes & Epkins, 2023). Teachers may be less likely to notice internalizing problems in busy classroom environments (Liljequist & Renk, 2007), but they likely also have more experience with a wide range of child behaviors on which to base their observations (Winsler & Wallace, 2002). In the case of siblings, school environments may also represent an opportunity to be removed from additional stressors at home and to receive support from people outside of the family (Brennan et al., 2013; Gan et al., 2017; Pavlopoulou et al., 2022). While a recent review found that siblings can experience isolation or bullying at school due to their brother or sister’s condition, social support was identified as a key resilience factor for siblings (Wolff et al., 2022).

Findings from this study add nuance to previous research showing that siblings, on average, are at a slightly raised risk of internalizing and externalizing problems (Pinquart, 2023; Shivers et al., 2019). We used LPA to identify four distinct classes of siblings with different levels of internalizing and externalizing problems and different levels of discrepancies between parent and teacher raters. Two classes had significant discrepancies between raters. In the first discrepant class, teachers reported low levels of internalizing problems and high levels of externalizing problems, while parents reported high levels of both internalizing and externalizing problems. In the second discrepant class, teachers reported low levels of internalizing and externalizing problems, while parents reported high levels of internalizing and externalizing problems. The two non-discrepant classes had lower levels of internalizing and externalizing behaviors, with high agreement between teacher and parent raters. This analysis suggests that while some siblings are doing well, others have raised levels of problems across one or more contexts and therefore may benefit from additional support.

Our findings also add to research using LPA to examine parent–teacher agreement in other areas of child behavior. Studies comparing parent and teacher ratings of autism symptoms have also identified four distinct patterns of interrater agreement, with two discrepant classes and two non-discrepant classes (Kang et al., 2023; Lerner et al., 2017). Two key differences can be noted between our results and those of Kang et al. (2023) and Lerner et al. (2017), however. First, our study’s two discrepant classes were both made up of low-teacher and high-parent ratings (one class for internalizing problems and one class for both externalizing and internalizing problems), while Kang et al. (2023) and Lerner et al. (2017) identified one class where parents made higher ratings and one class where teachers made higher ratings. Second, our two non-discrepant classes showed only slightly elevated levels of symptoms from both raters, while Kang et al. (2023) and Lerner et al. (2017)’s non-discrepant classes showed agreement between raters for both low autism symptoms and high autism symptoms.

Previous research on interrater discrepancies has aimed to uncover factors that are associated with higher or lower levels of discrepancy, but a consistent pattern has not been identified for factors such as the child’s age, sex, or ethnicity, family socio-economic status or parenting stress (Cheng et al., 2018). Similarly, our study did not find a significant effect of sibling age on the likelihood of belonging to the discrepant versus non-discrepant classes, and found only a borderline significant effect of sex. Siblings’ adjustment to the disorder (NAS) also did not predict membership in the discrepant versus non-discrepant classes. In the case of age, it has been argued that effects may be difficult to detect because many studies of child behavior have relatively narrow age ranges (De Los Reyes & Kazdin, 2005). Our study sample ranged from 8 to 16 years old, but with more than half of the sample aged between 8 and 10. Perhaps an effect would have been detected if a wider, or more evenly distributed, range of ages had been included. In the case of the NAS, while impact of the disorder and how siblings adjust have been shown to be associated with sibling well-being (Pinquart, 2023), results of this study suggest that adjustment to the disorder is not related to siblings’ behavior in the school versus home contexts and/or the way in which this behavior is perceived by teachers versus parents.

In terms of sex, we expected that interrater discrepancies would be higher for boy siblings, based on the findings of our previous study reporting parent and teacher SDQ scores (Prentice et al., 2024), but this hypothesis was only marginally supported by the results, with a significant p-value only before the Bonferroni correction was applied. Previous studies of interrater discrepancies have found mixed results in terms of sex effects. For example, a compilation of parent and teacher SDQ scores across several European countries found that sex was significantly associated with parent–teacher agreement for externalizing but not internalizing problems (Cheng et al., 2018). Some studies have suggested that agreement between parents and teachers is higher for boys (Berg-Nielsen et al., 2012; Satake et al., 2003), but a recent meta-analysis of sex and interrater discrepancies did not find a significant effect (Carneiro et al., 2021). It is possible that the borderline effect of sex that we detected in this study was due to boys, in general, exhibiting more externalizing behaviors, and externalizing behaviors, in turn, being linked to higher levels of interrater agreement.

Implications

The results of this study show that while many siblings fare well in terms of mental health and well-being, others experience considerable problems and could benefit from support. Further, given that parent and teacher ratings were significantly discrepant for half of the siblings in the sample, the use of multiple informants in assessing sibling well-being is advisable, and could help facilitate communication about sibling well-being between parents, teachers, and clinicians. It is now well established that interrater discrepancies provide domain-relevant information regarding children’s mental health and behavior across different contexts, such as home and school (De Los Reyes & Epkins, 2023). However, it is less clear how information from multiple informants should be integrated in research and clinical practice. Frameworks such as the Operations Triad Model (De Los Reyes, 2023) suggest ways to take account of both domain-relevant information and measurement confounders, but more work is needed in this area to operationalize conceptual work. In the meantime, the implementation of school-based support systems for siblings could be useful as an additional layer of support outside of health care systems, and training teachers to better identify internalizing problems could help improve early detection and support for those siblings who would benefit.

Previous studies of interrater discrepancies in clinical youth populations have found links between level of interrater discrepancy and clinical characteristics, for example with lower parent and teacher discrepancy regarding autism symptoms linked to higher observed clinical severity (Lerner et al., 2017) and higher parent and youth discrepancy regarding internalizing symptoms linked to a higher risk of suicidality in youth in inpatient psychiatric care (Makol et al., 2019). Our study did not identify factors that were significantly associated with parent–teacher interrater discrepancies for siblings. Future research should examine additional predictors of parent–teacher discrepancies related to siblings, such as mental health and socio-economic status of parents. Better identification of such predictors is crucial for developing person-centered interventions and establishing stronger processes for identifying those who are at risk.

Strengths and Limitations

While numerous studies have examined parent–teacher discrepancies on measures of child behavior, this is the first such study to do so for siblings of children with chronic disorders. Furthermore, the study used a person-oriented approach, LPA, which allowed the examination of multiple behavior ratings by multiple raters at the same time. This provided insight into how different groups of siblings behave in different settings, adding important detail to the overall picture of siblings being at a raised risk of experiencing problems.

Limitations related to the study should also be noted. It is possible that the study was underpowered, given that the sample size was calculated to detect changes across two groups in a randomized controlled trial rather than for sub-analysis of baseline data. Since parents elected to take part in an intervention to improve sibling well-being, it is possible that parents who were more worried about their sibling child were more likely to participate. This could mean that participant siblings were at a higher risk of problems than siblings generally, and may have contributed to the higher levels of concern voiced by parents compared with teachers. Participant parents were more likely to be highly educated than the general population, and 93.4% of participant siblings identified as “European” in origin, raising questions about generalizability to more diverse populations. Finally, no demographic data were available for participant teachers, so it was not possible to examine whether personal characteristics such as age or years of practice may have played a role in how they perceived siblings’ behavior.

Conclusion

This study identified four distinct classes of siblings in terms of parent and teacher reporting of internalizing and externalizing problems. Girls were slightly less likely than boys to be included in the discrepant classes—that is, parents and teachers were more likely to agree regarding girls’ internalizing and externalizing behaviors—but neither sibling age nor the impact of the disorder had a significant effect. These results reflect the larger literature on interrater agreement and discrepancies, but more research is needed to better understand interrater discrepancies for siblings of children with chronic disorders. This would promote the development of screenings and interventions that are tailored to those siblings who are identified as being at higher risk, ensuring that they receive appropriate support in both home and school environments.

Supplemental Material

sj-docx-1-ebx-10.1177_10634266261417609 – Supplemental material for Parent–Teacher Discrepancies About the Mental Health of Siblings of Children With Chronic Disorders: A Latent Profile Analysis

Supplemental material, sj-docx-1-ebx-10.1177_10634266261417609 for Parent–Teacher Discrepancies About the Mental Health of Siblings of Children With Chronic Disorders: A Latent Profile Analysis by Caitlin M. Prentice, Susanne Wehrli, Erica Zahl, Stian Orm, Matteo Botta, Torun M. Vatne, Trude Fredriksen, Solveig Kirchhofer and Krister Fjermestad in Journal of Emotional and Behavioral Disorders

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by grants from the Norwegian Research Council (#321027; PI Krister Fjermestad).

ORCID iDs

Caitlin M. Prentice

Solveig Kirchhofer

Supplemental Material

Supplementary material for this article is available on the Journal of Emotional and Behavioral Disorders website at .

References

Achenbach

T. M.

(2011). Commentary: Definitely more than measurement error: But how should we understand and deal with informant discrepancies? Journal of Clinical Child & Adolescent Psychology, 40(1), 80–86. https://doi.org/10.1080/15374416.2011.533416

Achenbach

T. M.

McConaughy

S. H.

Howell

C. T.

(1987). Child/adolescent behavioral and emotional problems: Implications of cross-informant correlations for situational specificity. Psychological Bulletin, 101(2), 213–232. https://doi.org/10.1037/0033-2909.101.2.213

Aitken

Henry

Andrade

B. F.

(2018). Distilling heterogeneity among children with disruptive behavior: Associations between symptom patterns and social functioning. Journal of Abnormal Child Psychology, 46(6), 1241–1252. https://doi.org/10.1007/s10802-017-0350-1

Berg-Nielsen

T. S.

Solheim

Belsky

Wichstrom

(2012). Preschoolers’ psychosocial problems: In the eyes of the beholder? Adding teacher characteristics as determinants of discrepant parent–teacher reports. Child Psychiatry and Human Development, 43(3), 393–413. https://doi.org/10.1007/s10578-011-0271-0

Brennan

Hugh-Jones

Aldridge

(2013). Paediatric life-limiting conditions: Coping and adjustment in siblings. Journal of Health Psychology, 18(6), 813–824. https://doi.org/10.1177/1359105312456324

Carneiro

Soares

Rescorla

Dias

(2021). Meta-analysis on parent-teacher agreement on preschoolers’ emotional and behavioural problems. Child Psychiatry & Human Development, 52(4), 609–618. https://doi.org/10.1007/s10578-020-01044-y

Cheng

Keyes

K. M.

Bitfoi

Carta

M. G.

Koç

Goelitz

Otten

Lesinskiene

Mihova

Pez

Kovess-Masfety

(2018). Understanding parent-teacher agreement of the Strengths and Difficulties Questionnaire (SDQ): Comparison across seven European countries. International Journal of Methods in Psychiatric Research, 27(1), Article e1589. https://doi.org/10.1002/mpr.1589

Davies

N. M.

Dickson

Davey Smith

van den Berg

G. J.

Windmeijer

(2018). The causal effects of education on health outcomes in the UK Biobank. Nature Human Behaviour, 2(2), 117–125. https://doi.org/10.1038/s41562-017-0279-y

Deavin

Greasley

Dixon

(2018). Children’s perspectives on living with a sibling with a chronic illness. Pediatrics, 142(2), Article e20174151. https://doi.org/10.1542/peds.2017-4151

10.

De Los Reyes

Augenstein

T. M.

Wang

Thomas

S. A.

Drabick

D. A. G.

Burgers

D. E.

Rabinowitz

. (2015). The validity of the multi-informant approach to assessing child and adolescent mental health. Psychological Bulletin, 141(4), 858–900. https://doi.org/10.1037/a0038498

11.

De Los Reyes

Epkins

C. C

. (2023). Than measurement error: Toward guidelines for integrating data from multi-informant assessments of youth mental health. Journal of Clinical Child & Adolescent Psychology, 52(1), 1–18. https://doi.org/10.1080/15374416.2022.2158843

12.

De Los Reyes

Kazdin

A. E

. (2005). Informant discrepancies in the assessment of childhood psychopathology: A critical review, theoretical framework, and recommendations for further study. Psychological Bulletin, 131(4), 483–509. https://doi.org/10.1037/0033-2909.131.4.483

13.

De Los Reyes

Lerner

M. D.

Keeley

L. M.

Weber

R. J.

Drabick

D. A. G.

Rabinowitz

Goodman

K. L

. (2019). Improving interpretability of subjective assessments about psychological phenomena: A review and cross-cultural meta-analysis. Review of General Psychology, 23(3), 293–319. https://doi.org/10.1177/1089268019837645

14.

De Los Reyes

Youngstrom

E. A.

Pabón

S. C.

Youngstrom

J. K.

Feeny

N. C.

Findling

R. L

. (2011). Internal consistency and associated characteristics of informant discrepancies in clinic referred youths age 11 to 17 years. Journal of Clinical Child and Adolescent Psychology: The Official Journal for the Society of Clinical Child and Adolescent Psychology, American Psychological Association, Division, 40(1), 36–53. https://doi.org/10.1080/15374416.2011.533402

15.

Ferguson

S. L.

Moore

E. W. G.

Hull

D. M.

(2020). Finding latent groups in observed data: A primer on latent profile analysis in Mplus for applied researchers. International Journal of Behavioral Development, 44(5), 458–468. https://doi.org/10.1177/0165025419881721

16.

Fjermestad

K. W.

Haukeland

Y. B.

Mossige

Vatne

T. M.

(2019). Children’s perspectives on the experiences of their siblings with chronic disorders. Clinical Social Work Journal, 47(3), 290–299. https://doi.org/10.1007/s10615-019-00705-3

17.

Fjermestad

K. W.

Silverman

W. K.

Vatne

T. M.

(2020). Group intervention for siblings and parents of children with chronic disorders (SIBS-RCT): Study protocol for a randomized controlled trial. Trials, 21(1), 851. https://doi.org/10.1186/s13063-020-04781-6

18.

Fredriksen

Marie Vatne

Bjartveit Haukeland

Tudor

Fjermestad

K. W.

(2023). Siblings of children with chronic disorders: Family and relational factors as predictors of mental health. Journal of Child Health Care : For Professionals Working with Children in the Hospital and Community, 27(1), 145–159. https://doi.org/10.1177/13674935211052157

19.

Gan

L. L.

Lum

Wakefield

C. E.

Nandakumar

Fardell

J. E.

(2017). School experiences of siblings of children with chronic illness: A systematic literature review. Journal of Pediatric Nursing, 33, 23–32. https://doi.org/10.1016/J.PEDN.2016.11.007

20.

Goodman

(1997). The Strengths and Difficulties Questionnaire: A research note. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 38(5), 581–586. https://doi.org/10.1111/j.1469-7610.1997.tb01545.x

21.

Goodman

(2001). Psychometric properties of the Strengths and Difficulties Questionnaire. Journal of the American Academy of Child & Adolescent Psychiatry, 40(11), 1337–1345. https://doi.org/10.1097/00004583-200111000-00015

22.

Gresham

F. M.

Elliott

S. N.

Metallo

Byrd

Wilson

Cassidy

(2018). Cross-informant agreement of children’s social-emotional skills: An investigation of ratings by teachers, parents, and students from a nationally representative sample. Psychology in the Schools, 55(2), 208–223. https://doi.org/10.1002/pits.22101

23.

Griffith

G. M.

Hastings

R. P.

Petalas

M. A.

(2014). Fathers’ and mothers’ ratings of behavioral and emotional problems in siblings of children with autism spectrum disorder. Journal of Autism and Developmental Disorders, 44(5), 1230–1235. https://doi.org/10.1007/s10803-013-1969-6

24.

Haapasalo

Tremblay

R. E.

Boulerice

Vitaro

(2000). Relative advantages of person- and variable-based approaches for predicting problem behaviors from kindergarten assessments. Journal of Quantitative Criminology, 16(2), 145–168. https://doi.org/10.1023/A:1007512521780

25.

Hanvey

Malovic

Ntontis

(2022). Glass children: The lived experiences of siblings of people with a disability or chronic illness. Journal of Community & Applied Social Psychology, 32(5), 936–948. https://doi.org/10.1002/casp.2602

26.

Huang

(2017). Cross-informant agreement on the child behavior checklist for youths: A meta-analysis. Psychological Reports, 120(6), 1096–1116. https://doi.org/10.1177/0033294117717733

27.

Kang

Lerner

M. D.

Gadow

K. D.

(2023). The importance of parent-teacher informant discrepancy in characterizing autistic youth: A replication latent profile analysis. Journal of Clinical Child and Adolescent Psychology: The Official Journal for the Society of Clinical Child and Adolescent Psychology, American Psychological Association, Division, 52(1), 108–118. https://doi.org/10.1080/15374416.2022.2154217

28.

Kyttälä

Sinkkonen

H.-M.

Kõiv

(2021). Social, emotional, and behavioral strengths and difficulties among sixth grade students: Comparing student and teacher ratings in Finland and Estonia. International Journal of School & Educational Psychology, 9(1), 42–54. https://doi.org/10.1080/21683603.2019.1640148

29.

Lerner

M. D.

De Los Reyes

Drabick

D. A. G.

Gerber

A. H.

Gadow

K. D.

(2017). Informant discrepancy defines discrete, clinically useful autism spectrum disorder subgroups. Journal of Child Psychology and Psychiatry, 58(7), 829–839. https://doi.org/10.1111/jcpp.12730

30.

Levante

Martis

Del Prete

C. M.

Martino

Primiceri

Lecciso

(2025). Siblings of persons with disabilities: A systematic integrative review of the empirical literature. Clinical Child and Family Psychology Review, 28(1), 209–253. https://doi.org/10.1007/s10567-024-00502-6

31.

Liljequist

Renk

(2007). The relationships among teachers’ perceptions of student behaviour, teachers’ characteristics, and ratings of students’ emotional and behavioural problems. Educational Psychology, 27, 557–571. https://doi.org/10.1080/01443410601159944

32.

Lobato

D. J.

Kao

B. T.

(2002). Integrated sibling-parent group intervention to improve sibling knowledge and adjustment to chronic illness and disability. Journal of Pediatric Psychology, 27(8), 711–716. https://doi.org/10.1093/jpepsy/27.8.711

33.

Mahat

Brøndbo

P. H.

Handegård

B. H.

Kvernmo

Javo

A. C.

(2022). Teacher reports of emotional and behavioral problems in Nepali schoolchildren: To what extent do they agree with parent reports? BMC Psychiatry, 22(1), Article 584. https://doi.org/10.1186/s12888-022-04215-4

34.

Makol

B. A.

De Los Reyes

Ostrander

R. S.

Reynolds

E. K.

(2019). Parent-youth divergence (and convergence) in reports of youth internalizing problems in psychiatric inpatient care. Journal of Abnormal Child Psychology, 47(10), 1677–1689. https://doi.org/10.1007/s10802-019-00540-7

35.

Marsh

H. W.

Lüdtke

Trautwein

Morin

A. J. S.

(2009). Classical latent profile analysis of academic self-concept dimensions: Synergy of person- and variable-centered approaches to theoretical models of self-concept. Structural Equation Modeling: A Multidisciplinary Journal, 16(2), 191–225. https://doi.org/10.1080/10705510902751010

36.

Martinez

Pechlivanoglou

Meng

Traubici

Mahood

Korczak

Colasanto

Mahant

Orkin

Cohen

(2022). Clinical health outcomes of siblings of children with chronic conditions: A systematic review and meta-analysis. The Journal of Pediatrics, 250, 83–92.e8. https://doi.org/10.1016/j.jpeds.2022.07.002

37.

Masyn

K. E.

(2013). Latent class analysis and finite mixture modeling. In Little

T. D.

(Ed.), The Oxford handbook of quantitative methods in psychology: Vol. 2. Statistical analysis (pp. 551–611). Oxford University Press. https://doi.org/10.1093/oxfordhb/9780199934898.013.0025

38.

Murray

A. L.

Speyer

L. G.

Hall

H. A.

Valdebenito

Hughes

(2021). Teacher versus parent informant measurement invariance of the Strengths and Difficulties Questionnaire. Journal of Pediatric Psychology, 46(10), 1249–1257. https://doi.org/10.1093/jpepsy/jsab062

39.

Nylund-Gibson

Choi

A. Y.

(2018). Ten frequently asked questions about latent class analysis. Translational Issues in Psychological Science, 4(4), 440–461. https://doi.org/10.1037/tps0000176

40.

Orm

Haukeland

Vatne

Silverman

W. K.

Fjermestad

(2022). Prosocial behavior is a relative strength in siblings of children with physical disabilities or autism spectrum disorder. Journal of Developmental and Physical Disabilities, 34(4), 591–608. https://doi.org/10.1007/s10882-021-09816-7

41.

Orm

Vatne

Haukeland

Y. B.

Silverman

W. K.

Fjermestad

(2021). The validity of a measure of adjustment in siblings of children with developmental and physical disabilities: A brief report. Developmental Neurorehabilitation, 24(5), 355–358. https://doi.org/10.1080/17518423.2020.1869338

42.

Pavlopoulou

Burns

Cleghorn

Skyrla

Avnon

(2022). “I often have to explain to school staff what she needs”. School experiences of non-autistic siblings growing up with an autistic brother or sister. Research in Developmental Disabilities, 129, 104323. https://doi.org/10.1016/j.ridd.2022.104323

43.

Perenc

Radochoński

Radochońska

(2015). Prosocial competencies among adolescent siblings of the physically disabled. Current Issues in Personality Psychology, 3(4), 195–202. https://doi.org/10.5114/cipp.2015.53897

44.

Pinquart

(2023). Behavior problems, self-esteem, and prosocial behavior in siblings of children with chronic physical health conditions: An updated meta-analysis. Journal of Pediatric Psychology, 48(1), 77–90. https://doi.org/10.1093/jpepsy/jsac066

45.

Prentice

C. M.

Orm

Botta

Vatne

T. M.

Fredriksen

Kirchhofer

Fjermestad

(2024). Teacher-rated mental health of siblings of children with chronic disorders. School Mental Health, 16, 1183–1196. https://doi.org/10.1007/s12310-024-09685-3

46.

R Core Team (2021). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/

47.

Rescorla

L. A.

Bochicchio

Achenbach

T. M.

Ivanova

M. Y.

Almqvist

Begovac

Bilenberg

Bird

Dobrean

Erol

Fombonne

Fonseca

Frigerio

Fung

D. S. S.

Lambert

M. C.

Leung

P. W. L.

Liu

Marković

Markovic

. . .Verhulst

F. C.

(2014). Parent–teacher agreement on children’s problems in 21 societies. Journal of Clinical Child & Adolescent Psychology, 43(4), 627–642. https://doi.org/10.1080/15374416.2014.900719

48.

Ringoot

A. P.

van der Ende

Jansen

P. W.

Measelle

J. R.

Basten

Jaddoe

V. W. V.

Hofman

Verhulst

F. C.

Tiemeier

(2015). Why mothers and young children agree or disagree in their reports of the child’s problem behavior. Child Psychiatry and Human Development, 46(6), 913–927. https://doi.org/10.1007/s10578-014-0531-x

49.

Rosenberg

J. M.

Beymer

P. N.

Anderson

D. J.

van Lissa

C. J.

Schmidt

J. A.

(2019). TidyLPA: An R package to easily carry out latent profile analysis (LPA) using open-source or commercial software. Journal of Open Source Software, 3(30), 978. https://doi.org/10.21105/joss.00978

50.

Rubin

D. H.

Althoff

R. R.

Walkup

J. T.

Hudziak

J. J.

(2013). Cross-informant agreement on child and adolescent withdrawn behavior: A latent class approach. Child Psychiatry and Human Development, 44(3), 361–369. https://doi.org/10.1007/s10578-012-0330-1

51.

Sahler

O. J.

Carpenter

P. J.

(1989). Evaluation of a camp program for siblings of children with cancer. American Journal of Diseases of Children (1960), 143(6), 690–696. https://doi.org/10.1001/archpedi.1989.02150180068023

52.

Santos

Farrington

D. P.

da Agra

Cardoso

C. S.

(2020). Parent-teacher agreement on children’s externalizing behaviors: Results from a community sample of Portuguese elementary-school children. Children and Youth Services Review, 110, 104809. https://doi.org/10.1016/j.childyouth.2020.104809

53.

Satake

Yoshida

Yamashita

Kinukawa

Takagishi

(2003). Agreement between parents and teachers on behavioral/emotional problems in Japanese school children using the child behavior checklist. Child Psychiatry and Human Development, 34(2), 111–126. https://doi.org/10.1023/A:1027342007736

54.

Shivers

C. M.

Jackson

J. B.

McGregor

C. M.

(2019). Functioning among typically developing siblings of individuals with autism spectrum disorder: A meta-analysis. Clinical Child and Family Psychology Review, 22(2), 172–196. https://doi.org/10.1007/s10567-018-0269-2

55.

Streimann

Sisask

Toros

(2021). Children’s mental health in different contexts: Results from a multi-informant assessment of Estonian first-grade students. School Psychology International, 42(4), 379–397. https://doi.org/10.1177/01430343211000414

56.

Tein

J.-Y.

Coxe

Cham

(2013). Statistical power to detect the correct number of classes in latent profile analysis. Structural Equation Modeling: A Multidisciplinary Journal, 20(4), 640–657. https://doi.org/10.1080/10705511.2013.824781

57.

Vermaes

I. P. R.

van Susante

A. M. J.

van Bakel

H. J. A.

(2012). Psychological functioning of siblings in families of children with chronic health conditions: A meta-analysis. Journal of Pediatric Psychology, 37(2), 166–184. https://doi.org/10.1093/jpepsy/jsr081

58.

Winsler

Wallace

G. L.

(2002). Behavior problems and social skills in preschool children: Parent-teacher agreement and relations with classroom observations. Early Education & Development, 13(1), 41–58. https://doi.org/10.1207/s15566935eed1301_3

59.

Woerner

Fleitlich-Bilyk

Martinussen

Fletcher

Cucchiaro

Dalgalarrondo

Lui

Tannock

(2004). The Strengths and Difficulties Questionnaire overseas: Evaluations and applications of the SDQ beyond Europe. European Child & Adolescent Psychiatry, 13(S2), ii47–ii54. https://doi.org/10.1007/s00787-004-2008-0

60.

Wolff

Magiati

Roberts

Pellicano

Glasson

E. J.

(2022). Risk and resilience factors impacting the mental health and wellbeing of siblings of individuals with neurodevelopmental conditions: A mixed methods systematic review. Clinical Psychology Review, 98, 102217. https://doi.org/10.1016/j.cpr.2022.102217