Introduction

Disruptive behaviors (DB), such as aggression, defiance, and temper tantrums, are among the most commonly reported types of behavioral difficulties in early childhood.¹ This early clinical phenomenology reflects deviation from normative developmental patterns within an age period.² Although some DB are considered a normal part of the developmental process of preschool children, they are the most common complaints among parents of young children in multiple settings.^3,4 According to previous studies, severe DB in early-life is predictive of a child’s performance as they grow into school age, adolescence, and adulthood. Furthermore, early childhood diagnosis of disruptive behavior disorders (DBDs) is a strong predictor of long-lasting disruptive psychopathology and adverse life outcomes.^5-7

Earlier research has highlighted various risk factors linked to disruptive behaviors, such as parental depression⁸, conflict and maltreatment in the family⁹, and poverty.¹⁰ Parenting styles have been among the most commonly studied risk factors.^11-13 Most observational studies have consistently shown that children whose parents have negative parenting styles are more likely to develop DB problems. In contrast, parents who employ positive parenting styles are less likely to have children displaying disruptive behaviors.^11,12 In particular, exposure to physical and verbal punishment during preschool years is found to be associated with high levels of DB, which persists or increases across development.^14,15

Studies conducted in different cultures have shown similar results. For example, a recent study involving a large cohort from United States with low-income urban families found that maternal spanking at age three predicted externalizing behaviors at age five, independent of other parental factors.¹⁶ A study of clinic-referred children in Belgium found that high levels of maternal-reported coerciveness, including physical punishment, were associated with an increase in externalizing behaviors over time.¹⁷ Considering the long-term negative consequences of conduct problems, it is important to recognize the associated factors in the early period. Although there are some studies investigating parenting practices in preschool children in Turkey^18,19, to our knowledge there is no study in the Turkish sample examining the relationship between behavioral problems and parenting attitudes, including functionality in preschool-aged children through the Turkish adaptation of the Preschool Age Psychiatric Assessment (PAPA).

Although many studies assume parents have a one-directional influence on their children’s behaviors, findings suggest a reciprocal relationship, such that children’s behaviors may have an impact on their parents too.^13,20,21 There have however been a few studies that have examined this bidirectional relationship between various externalizing behavioral problems and parenting styles, yet findings have been inconsistent.^22-24 This bidirectional relationship has been mostly studied in attention deficit hyperactivity disorder (ADHD), which often accompanies conduct problems and has been shown to be bidirectionally impaired in relation to parental functionality.^25,26 The results support both parent and child effects in the relation between child ADHD symptoms and family functioning.²⁵ Moreover, existing research and cross-cultural theory suggest that children’s socialization, developmental patterns, and parent-child interactions may vary across cultures.²⁶ Therefore, necessitating an investigation into this bidirectional relationship in our own culture in a clinical preschool sample.

In the current study, we hence aimed to evaluate parenting characteristics (i.e. emotional warmth, control attempts, discipline) family functionality, and disruptive problems in a sample of Turkish preschool children. We hypothesized that children with more DB symptoms would have more problematic parental attitudes and poorer family functioning than children with fewer DB symptoms. Additionally, we expected DB symptoms and other factors such as ADHD diagnosis and parental psychopathology to affect family functioning in a negative way.

Materials and Methods

Participants

The participants of the study were selected from the Turkish Preschool Age Psychiatric Assessment (PAPA) Reliability and Validity Study. Parents of all children who applied to our university’s Department of Child and Adolescent Psychiatry outpatient clinic, between the ages of 2 and 6 years, were administered the structured interview PAPA. The Turkish PAPA study involved 300 patients. Because behavioral problems are often accompanied by other disorders, patients diagnosed with neurodevelopmental disorders such as autism spectrum disorders and global developmental delay were excluded so as to not confound the results. Diagnostic assessments for autism spectrum disorder and global developmental delay were performed by experienced clinicians based on the criteria outlined in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5). A total of 223 patients were included in the current study.

The high DB group (n=93) included children who had more than 3 conduct problem symptoms whose symptoms were related to functional impairment due to disruptive problems and had no chronic medical conditions. Patients with 3 or fewer conduct problem symptoms, who had no functional impairment due to disruptive problems and no chronic medical condition were considered as the Low DB group (n=130). Since oppositional defiant disorder (ODD) in the PAPA includes the same diagnostic criteria as DSM-5, and the diagnosis requires meeting at least four criteria and impaired functioning, this was accepted as the threshold. Other psychiatric disorders (ADHD, anxiety disorders etc.) were also assessed using PAPA and were not exclusion criteria for either group.

Measures

Sociodemographic data form:

The form was created by researchers based on the original sociodemographic form of the PAPA and included questions to gather information about the age, gender, education level of the child and parents, family structure, and number of siblings.

The ‘Child behavior checklist for ages 1.5-5’ (CBCL/1½−5):

The CBCL/1½−5 is a parent-rated questionnaire designed to assess children’s problem behaviors and consists of 100 items about children’s behaviors and emotional issues.²⁷ The CBCL/1½−5 consists of seven subscales into which the items can be clustered: emotionally reactive, anxious/depressed, somatic complaints, withdrawn, sleep problems, attention problems, and aggressive behavior. The Turkish standardization study was conducted by Erol et al.²⁸ This scale was used in this study to assess and compare behavioral and emotional problems.

Preschool age psychiatric assessment (PAPA):

PAPA, developed in 1999, is an interviewer-based, caregiver-reported diagnostic assessment method. This structured interview evaluates the symptoms in four main areas: (1) diagnostic criteria of all diagnoses in DSM-5 and ICD-11, (2) all of the Research Diagnostic Criteria-Preschool Age (RDC-PA) items, (3) all of the Diagnostic Classification of Mental Health and Developmental Disorders of Infancy and Early Childhood 5 (DC: 0-5), (4) potential behaviors and symptoms that are not merely diagnostic criteria like sleeping rituals and peer relationships. In addition, the interview assessed the family environment and relationships, family psychosocial status, and functional impairments. It covered not just disorders and problems, but all aspects that affect young children’s mental health.²⁹

The reliability study of PAPA was conducted by Egger et al., and no significant difference was found in reliability according to age, gender, and race. The interview has been shown to be a valid tool for children aged between 2 and 6 years, although its use up to age 8 years has been established.^30,31 The psychometric study was conducted in Turkey, and reliability and validity were demonstrated in the Turkish population.³² The PAPA conduct section covers all DSM-5 criteria for ODD and conduct disorder (CD). The PAPA was used in the current study to evaluate conduct problems, family demographics, and family functionality. For this assesment, Family section, Conduct Problems section, Incapacity section and Child and Adolescent Impact Assessment sections were used. In the interview, the interviewer asked about symptoms related to each relevant module. Following the guidelines in the PAPA dictionary, each symptom was assessed on a three-point scale — absent (0), sometimes present (1), or definitely present (2) —. The symptom scales provided a continuous measure of observed symptoms. These scores were summed to obtain the total score. Similarly, a numerical variable was obtained by summing the scores of the variables related to functionality. Higher scores indicate more severe impairment in functioning across more areas.

Procedure

Patients aged between 2 and 6 years who applied to our outpatient clinic were evaluated, and written and verbal consent was obtained from the parents who agreed to participate in the study. Following this, a child psychiatrist interviewed the parents and children. After the interview, the clinician completed the sociodemographic data form and PAPA short forms, and the CBCL/1½−5 was completed by the parents. Of the parents interviewed, 215 (96%) were mothers, 5 were fathers (2.2%), and in 3 cases (1.34%) were both mothers and fathers. The study was approved by the Ankara University Faculty of Medicine Non-Invasive Clinical Research Ethics Committee on October 10, 2019 (Decision No: I4-151-19).

Statistical analysis

For evaluating the tests and scales, IBM SPSS Statistics for Windows 30.0 software was used. Descriptive statistics are given as mean ± standard deviation or median and interquartile range (IQR) for continuous variables. Frequencies (percentages) are given for categorical variables. The Kolmogorov-Smirnov test is used to check whether continuous variables follow a normal distribution. For comparisons of groups, the independent sample t-test was used for independent samples of normally distributed data. The Mann-Whitney U test was performed to test the significance of pairwise differences. The chi-square or Fisher’s exact test was used for comparison of categorical variables. A multiple linear regression model was used to identify independent predictors of functionality. Goodness-of-fit statistics were used to assess model fit. The significance level for this study was set to 5%. The p<0.05 was considered statistically significant.

Results

A total of 223 children, with a mean age of 45.74 months (standard deviation: 12.93), were included in the study. Among the children, 116 (52.07%) were boys. The sociodemographic characteristics of the groups are shown in Table I.

We assessed maternal and paternal attitudes that could be related to DB. Spanking with the hand, verbal dispraise, and selective negative view were more common in the DB group (Table II).

We also analyzed parents’ relationship, perceptions about their partners, and mental health status, as factors that could be related to higher DB. Both mothers and fathers have more psychopathology in the high DB group. Parental arguments were seen as more frequent in the high DB group. Socioeconomic levels were similar in the groups (Table III).

Another aspect that we evaluated was the areas of impairment in the PAPA. There was a serious deterioration in the child’s functionality in almost all areas (Table IV). When the correlation between CBCL and impairment in child’s functioning was evaluated, it was found that the domains of relationships with parents, helping in cooperation, leisure time activities, problems with adults outside the home/nursery and ability to act appropriately outside home or daycare/school were positively correlated with the CBCL domains of ADHD (relationship with mother: r=0.24, p<0.001; relationship with father: r=0.24, p<0.001; helping in cooperation: r=0.2, p=0.002, problems with adults outside the home/nursery: r=0.21, p=0.001; ability to act appropriately outside home or daycare/school: r=0.21, p=0.001), aggressive behaviors (relationship with mother: r=0.4, p<0.001; relationship with father: r=0.44, p<0.001; helping in cooperation: r=0.28, p<0.001, problems with adults outside the home/nursery: r=0.19, p=0.002; leisure time activities: r=0.25, p<0.001; ability to act appropriately outside home or daycare/school: r=0.21, p=0.001; peer relationships: r=0.19, p=0.003), somatic (helping in cooperation: r=0.23, p<0.001, problems with adults outside the home/nursery: r=0.17, p=0.008; leisure time activities: r=0.27, p<0.001; ability to act appropriately outside home or daycare/school: r=0.22, p<0.001), sleep (relationship with mother: r=0.24, p<0.001; relationship with father: r=0.22, p<0.001; sibling problems: r=0.24, p<0.001; helping in cooperation: r=0.25, p<0.001, problems with adults outside the home/nursery: r=0.22, p<0.001; leisure time activities: r=0.19, p=0.003), anxious/depressed (relationship with mother: r=0.24, p<0.001; relationship with father: r=0.18, p=0.007; helping in cooperation: r=0.18, p=0.005, problems with adults outside the home/nursery: r=0.26, p<0.001; peer relationships: r=0.2, p=0.002), withdrawn (problems with adults outside the home/nursery: r=0.29, p<0.001; leisure time activities: r=0.25, p<0.001; ability to act appropriately outside home or daycare/school: r=0.19, p=0.004; peer relationships: r=0.27, p=0.001), emotional reactivity (relationship with mother: r=0.25, p<0.001; relationship with father: r=0.22, p<0.001; helping in cooperation: r=0.21, p=0.008, problems with adults outside the home/nursery: r=0.22, p<0.001).

We also assessed the specific impact of symptoms on aspects of the parent’s life. We found that DB symptoms had a more significant negative impact on the parent’s relationship with other children, relationship with other family members, and participation in social and personal activities, in the high DB group compared to the low DB group (Table V).

In line with our hypothesis and correlation analysis, multiple linear regression analysis was performed to investigate the factors that affected children’s functionality (Table VI). We used the backward linear regression method. First, we included the child’s age, gender, maternal psychopathology, paternal psychopathology, presence of ADHD, and number of symptoms of DB in the model. Then, the maternal psychopathology and gender of the child, which was found to be unrelated, were excluded from the model. It was found that the gender of the child, presence of paternal psychopathology, presence of ADHD, and number of symptoms of DB were associated with deterioration of child’s functionality, as shown in Table VI.

We also performed correlation analyses to investigate the influence of factors related to parental impact. There was a weak positive correlation between impact on parents and symptom number of DB (r=0.16, p=0.01 and presence of ADHD (r=0.19, p=0.004). In the multiple regression analysis, no significant effect of these factors was found.

Discussion

The current study investigated the relationship between familial factors, family functioning and disruptive behavior problems in preschool children in a Turkish clinical sample. Our hypothesis was that children exhibiting more behavioral problems were more likely to have parental problems and worse familial functioning than the control group. It was found that spanking with hand, verbal dispraise, and selective negative view of the child were more frequent in the DB group than in the control group. Additionally, although problems such as physical violence between parents were not more frequent, arguments between parents and parental psychopathology were found to be reported more frequently in the DB group. Finally, in the current study, DB symptoms were shown to have a negative impact in several areas on both the child’s functioning and the parent’s life. We also showed that not only disruptive symptoms but also various factors such as the presence of ADHD, parental psychopathology, and the age of the child predicted impairment in their functionality.

The first finding of the current study was that sociodemographic characteristics were similar in both groups. Previous research has shown that families of children with ODD/CD are more likely to experience socioeconomic disadvantage, including lower parental educational status.³³ According to the current study, the DB group showed no significant differences compared with the control group. In previous studies conducted in our country, socioeconomic status was associated with DB problems.³⁴ The difference in our study was attributed to the clinical sample. It was also thought that the similarity in these characteristics may have minimized confounding factor in terms of the results.

Coercive, harsh, and conflictual parenting practices are known risk factors for developing of clinically meaningful DB problems.³⁵ Consistent with previous studies, we found that spanking by hand, verbal dispraise and selective negative view of the child were more common in the DB group. We also hypothesized that difficulties in parents’ marital relationships may affect their parenting and thereby constitute a risk to their child’s DBs, as shown in previous studies.³⁶ However, we did not find any significant differences in parental relationship dissatisfaction. We thought that we did not find a difference between the two groups, because both groups were clinical samples. Additionally, temperamental traits influenced by genetic factors also play a role in conduct disorders, including genetic variants that affect emotional reactivity and social affiliation.³⁷ Therefore, we may find the impact of environmental factors as more limited.

Parental psychopathology, especially depression, is one of the best known risk factors for child mental health problems.³⁶ We found that parental psychopathology was more common in both mothers and fathers, consistent with the findings reported in existing literature regarding familial mental health trends. Parental psychopathology appears to be both an important risk factor for DB and a potential area for various interventions.

A mother’s negative perception of her child has been examined as a possible risk factor for child conduct problems. However, it was also suggested that the strong associations between parental perceptions and child conduct problems cannot confirm a causal connection between the variables or the direction of any such association if there is a causal link. It is highly likely that child conduct problems also promote negative parental cognitions and vice versa.³⁸ The cross-sectional structure of our study also does not allow such a casual inference.

When the functioning of these children was evaluated, we showed that although they were at a very early age (mean 45.74 months), DB negatively affected their functioning in many areas, especially in family relationships. While a large body of parenting research exists, fewer works has concentrated on exploring family functioning. In a study conducted with boys aged 11 to 16 years, it was shown that families with children with conduct problems had poorer affective involvement, general family functioning, and more poorly defined family roles than families with typically developing children.³⁹ To the best of our knowledge, we have not found a study investigating the effect of DB on family functioning in the preschool period. The findings of our study are consistent with those of previous studies in older children.⁴⁰

We have shown that DB has a very negative impact not only on the child’s functioning but also on the parents’ lives. We have shown that it negatively affects their relationships with other children, other family members, their family social activities, and their personal activities. These factors may contribute to the bidirectional relationship shown in previous studies.⁴¹ Given that both parenting and home environment contribute to the development of DB, exploring the relationship between family functioning and DB in prospective studies could provide further insight into how the family functions as a whole.

Finally, we showed that the number of DB symptoms, presence of ADHD, paternal psychopathology and the child’s age may be predictive factors for functioning. The deterioration in functioning as the number of DB symptoms increased was a generally anticipated result. ADHD and ODD frequently co-occur during preschool years ⁴², and there is a correlation between the symptoms of these two disorders at this developmental stage.⁴³ It is also known that ADHD affects the functioning of both children and their families through many mechanisms.³⁶ Therefore, when evaluating and monitoring a child with one of these conditions, the effect on functionality should not be ignored.

A surprising finding of the study was that maternal psychopathology had no predictive effect on functioning, while only the father’s psychopathology had an effect. A different pattern for mothers and fathers has been shown in an ADHD study, although the study reported findings opposite to our findings.²⁵ We thought that the reason for this might be that fathers had more severe disorders compared to mothers. Additionally, similar to the disruptive problems seen in children, the father may also have issues related to emotional dysregulation, which could affect the functioning of both the family and the child. Similarly, some of the previous research suggested that fathers’ psychological health has a significant impact on children’s externalizing behaviors especially, and fathers play a crucial role in shaping their children’s development.³⁶ It would be useful to perform a detailed assessment of the psychopathologies of the parents in order to make sense of this relationship.

Follow-up and evaluation of these patients will continue. However, even in this form, the findings of our study have important practical implications for family assessments and interventions. First, interventions that focus solely on parents or children may have limited effectiveness in breaking the feedback loop between poor parenting styles, family functioning, and children’s disruptive behaviors. Second, both parents appear to be significantly impacted by their children’s disruptive problems, so family-based interventions should involve both parents and address their functioning. Third, early intervention for young children is crucial, as our findings show clear and severe impairments in children around the age of 4 years.

There are several limitations that should be noted. First, in this article, preliminary findings are presented cross-sectionally. This prevents the establishment of a causal relationship in the child-parent relationship loop. Second, the study relied largely on PAPA validity and reliability study variables, future research is needed to replicate the findings using multimethod assessment. Third, in the present study, we did not have detailed information about the parents’ psychopathology. Future studies should also use collateral information to assess parent psychopathology. Fourth, although a longitudinal design provides stronger support for a causal link than a cross-sectional design, it is important to consider potential confounding variables. Fifth, this sample included only two-parent families, which limits generalizability. Finally, the study sample is a clinical sample and many other psychopathologies may be confounding factors.

Acknowledgements

The authors would like to express their sincere gratitude to all of the participating families as well as the full staff of resident doctors at the Child Psychiatry Department at Ankara University for their contributions to this study.

Ethical approval

The study was approved by Ankara University Ethics Committee (date: 10.10.2019, number: İ4-151-19).

Author contribution

The authors confirm contribution to the paper as follows: Study conception and design: MCU, DBÖ; data collection: MCU, EY, GYE and DBÖ; analysis and interpretation of results: MCU, HKÜ, SAA ; draft manuscript preparation: MCU, EY, DBÖ, HKÜ, SAA. All authors reviewed the results and approved the final version of the manuscript.

Source of funding

The authors declare the study received no funding.

Conflict of interest

The authors declare that there is no conflict of interest.