Abstract
Patients with hyperkinetic disorders (hereafter named ADHD) have diagnostic features such as attention problems, hyperactivity, and impulsivity. In addition to these core symptoms, coexisting disorders or problems are frequent, increasing the complexity of the symptomatic picture (Gillberg, 2010). ADHD is a heterogeneous disorder and studies have shown that the addition of coexisting problems, including emotional and conduct problems, substantially increase its impairment (Steinhausen et al., 2006), and risk of an unfavorable long-term outcome (Althoff, Verhulst, Rettew, Hudziak, & van der Ende, 2010; Biederman et al., 2009). Current evidence indicates that ADHD is a heritable disorder (Nigg, Nikolas, & Burt, 2010). In addition to genetic risk factors, adverse family and environmental risk factors are frequent among children and adolescents with ADHD (Biederman et al., 1995), and family problems have particularly been associated with coexisting externalizing problems. Recent studies have found a greater risk for psychopathology in parents of children with ADHD and additional externalizing problems (Chronis et al., 2003; Pfiffner, McBurnett, Rathouz, & Judice, 2005).
The evaluation of family functioning often focuses on the different types of interaction among family members in a number of areas, including the rules of behavior, family members’ roles, and affective concerns within the family (Epstein, Baldwin, & Bishop, 1983). To date, most studies have assessed the family functioning of children with ADHD and coexisting externalizing problems, while few have assessed family functioning in children with ADHD and internalizing problems (Deault, 2010).
The multidimensional concept of QoL has various definitions (Coghill, Danckaerts, Sonuga-Barke, Sergeant, & A. E. G. Group, 2009), but is commonly referred to as subjectively perceived well-being and satisfaction within several life domains (Mattejat & Remschmidt, 1998), such as physical and mental health, friends, family, school, and time alone. It has been shown that it is possible to improve QoL without reducing symptoms, which demonstrates the importance of assessing QoL (Bastiaansen, Koot, & Ferdinand, 2005). Nevertheless, treatment studies of children with ADHD reveal inconsistent findings of improved QoL over time (Danckaerts et al., 2010). A large European study found that QoL outcomes for adolescents with ADHD were negatively associated with emotional and conduct problems, physical health problems, coordination problems, and maternal and paternal mental health problems (Riley et al., 2006). However, an important limitation of the study was the lack of adolescent self-reports. Another study, which included self-reports, found worse QoL among adolescents with ADHD and conduct problems (Becker, Roessner, Breuer, Dopfner, & Rothenberger, 2011); however, this study did not include adolescents with ADHD and both emotional and conduct problems.
The use of multiple informant assessment is widely accepted in science and clinical practice (Achenbach & Rescorla, 2001). By including self-reports, the adolescents’ perspective and perceptions are acknowledged in the decision-making process (Coghill et al., 2009). A recent review demonstrated inflated perception of self-competence among children with ADHD (Owens, Goldfine, Evangelista, Hoza, & Kaiser, 2007). Even though the studies in the Owens et al. review were in the age group below 13 years, the findings suggest the importance of adding reports from parents and adolescents when assessing family functioning and QoL in adolescents with ADHD. Thus, although each informant provides a unique perspective that may be of great relevance in the assessment of the adolescents and the family, most ADHD studies to date have included only parent reports of family stress or functioning and QoL, and not self-reports of adolescents (Danckaerts et al., 2010; Deault, 2010). In addition, only a few studies in a recent review have assessed adolescents perceptions of family functioning; as most studies have focused on parent reports of children below the age of 11 (Deault, 2010).
To our best knowledge, no studies to date have assessed family functioning and QoL in adolescents with ADHD and coexisting emotional and conduct problems in a multi-informant perspective. Therefore, the present study was aimed at the following:
To compare the levels of self-reported family functioning and QoL among subgroups of adolescents with ADHD.
To compare the levels of parent-reported family functioning and QoL among subgroups of adolescents with ADHD.
To compare reports of adolescents with ADHD and their parents with regard to family functioning and QoL.
Method
The Clinical Sample
The study is part of The Health Survey in the Department of Child and Adolescent Psychiatry (CAP) in a hospital in Norway. This was a cross-sectional study of a defined clinical population. The catchment area is a county in Norway with 303,664 inhabitants, which includes urban and rural areas. The Department of CAP at the University Hospital covers all inhabitants in the county. Inclusion criteria were referred adolescents, aged 13 to 18 (20) years, who had at least one personal attendance at the clinic between February 15, 2009, and February 15, 2011. Exclusion criteria were major difficulties in answering the questionnaire due to their psychiatric state, cognitive function, visual impairments or lack of sufficient language skills. Emergency patients were invited to take part once they entered a stable phase.
In the study period, 2,032 adolescent patients had at least one attendance at the CAP clinic. Of these, 289 were excluded on the basis of the exclusion criteria. In addition, 95 were lost in the registration process (i.e., missing). Hence, 1,648 (81.1 %) were eligible and were invited to participate. Of these, a total of 717 (43.5%) participated in the CAP survey: 393 girls (54.8%) and 324 boys (45.2%).
To explore the representativeness of the study sample, anonymous information about the reference population was collected from annual reports from St. Olav’s University Hospital, 2009 to 2011. All adolescents in the study period (n = 2,032) minus those excluded (n = 289) were defined as the reference population (n = 1,743). In accordance with the permission given by the Norwegian Social Science Data Services, The Data Protection Official for Research, we compared age, sex, and main reason for referral between participants (n = 717) and nonparticipants (n = 1,026) of the reference population. Participants were 0.27 years (95% CI [0.10, 0.45]) older than nonparticipants, (M = 15.66, SD = 1.65) vs. (M = 15.39, SD = 1.95), p < .001. There were more girls in the study group than in the nonparticipating group: (n = 393; 54.8%) vs. (n = 509; 49.6%), p = .032. The main reason for referral did not differ between participants and nonparticipants (data not shown, Pearson Exact Chi-Square test; p = .11).
The adolescents diagnosed with ADHD (N = 194) were included in the present study (see Figure 1). By using the Strength and Difficulties Questionnaire (SDQ), the adolescents with ADHD were further divided into the following subgroups: no additional problems (ADHD Only), additional emotional problems (ADHD + EMO), conduct problems (ADHD + COND) or emotional and conduct problems (ADHD + EMO + COND) All other diagnostic groups and undiagnosed cases in the CAP survey were excluded from the present study (n = 474). ADHD patients with a missing SDQ questionnaire (n = 49) were also excluded.
Flowchart of the sample.
Procedure
Newly referred patients and patients already enrolled at the CAP clinic received oral and written invitations to participate in the study at their first attendance after the project commenced. Written informed consent was obtained from adolescents and parents prior to inclusion, according to the CAP survey procedures. The participating adolescents responded to an electronic questionnaire and data were collected from clinical charts. Parents also responded to a questionnaire with some information of socioeconomic status (SES), including educational level of both parents.
Measures
Sociodemographic information
Participants completed a demographic form with information about age, sex, and SES. The highest educational level of parents on an 8-point Hollingshead-type scale was used to estimate SES (Hollingshead, 1958).
Clinical diagnosis
Diagnoses were collected from clinical charts and followed the International Statistical Classification of Diseases and Related Health Problems, 10th revision (ICD-10) multiaxial diagnostic system (i.e., Axes I-VI; World Health Organization, 1992). All diagnoses were made by a clinical psychologist or a child and adolescent psychiatrist based on all clinical information. The CAP clinic’s standardized procedure for the assessment and diagnosis of hyperkinetic disorders is based on the national Guideline for Assessment and Treatment of ADHD (Norwegian Directorate of Health, 2007). This guideline, similar to other established ADHD guidelines (Subcommittee on Attention-Deficit/Hyperactivity Disorder et al., 2011), requires a clinical diagnostic interview assessing ADHD and possible coexisting disorders, a somatic assessment, and an interview with the teacher; it recommends the use of questionnaires filled out by the adolescent, parent, and teacher to obtain ADHD symptom scores. The ICD-10 diagnosis of hyperkinetic disorder is referred to as ADHD in this study. Diagnostic criteria for hyperkinetic disorder are nearly identical to criteria for ADHD combined type in the Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; DSM-IV-TR; American Psychiatric Association, 2000); however, specifiers such as mainly attention problems or mainly hyperactivity/impulsivity problems are not utilized in the ICD-10. A recent study of adults showed that DSM-IV ADHD inattentive and hyperactive-impulsive types are less likely to qualify for a diagnosis of hyperkinetic disorder (Gomez, 2016).
SDQ
Coexisting problems were measured using the Norwegian version (Van Roy, Groholt, Heyerdahl, & Clench-Aas, 2006) of the SDQ (Goodman, 1997). This clinical and research instrument contains 25 items, and includes self-report and parent report, addressing behavioral problems, emotional problems, aggressive behavior, and attention problems (Goodman, 1997). The SDQ subscales have shown satisfactory to good internal consistency, and the stability of the basic psychometric properties of the SDQ have been demonstrated across clinical samples (Becker et al., 2006). The SDQ adolescent self-report has shown satisfactory construct validity and internal consistency in a study by the main author (Cronbach’s α = 0.73; Goodman, 2001), as well as a recent Norwegian study (Van Roy, Veenstra, & Clench-Aas, 2008).
We used the Norwegian cut-off points of borderline level (80th percentile) for subscale scores on the adolescent self-report (Van Roy et al., 2006). Thus, the ADHD group was divided into four groups according to coexisting emotional and conduct problems: ADHD without self-reported emotional or conduct problems (ADHD Only), ADHD with emotional problems (ADHD + EMO), ADHD with conduct problems (ADHD + COND), and ADHD with emotional and conduct problems (ADHD + EMO + COND). SDQ adolescent self-report has shown satisfactory construct validity and internal consistency (Cronbach’s α = 0.73) (Goodman, 2001), supported by a recent Norwegian study (Van Roy et al., 2008).
General Functioning Scale (GFS)
Family functioning was measured using the GFS of the Norwegian version (Reigstad, Jorgensen, Sund, & Wichstrom, 2010) of the McMaster Family Assessment Device (FAD; Epstein et al., 1983). The 12-item parent proxy-report and self-report inventories measure family functioning reflecting six different areas, including problem solving, communication, roles, affective responsiveness, affective involvement and behavioral control (Byles, Byrne, Boyle, & Offord, 1988). Each item was rated on an ordinal scale (1 = strongly agree, 4 = strongly disagree), which were summed (range = 12-48) with higher scores indicating poorer functioning. The unhealthy functioning is indicated by scores of 24 or higher (2 × 12), representing a family member’s perception of emotional and physical health of the family members (Ryan, Epstein, Keitner, Miller, & Bishop, 2005). The reliability of the GFS is good, with a Cronbach’s alpha of 0.92 (Epstein et al., 1983). The construct validity of the GFS was supported by findings in the Ontario Child Health study, a large epidemiological study of all children from 4 to 16 years (Byles et al., 1988).
Inventory of Life quality of Children and adolescents (ILC)
QoL was measured using the Norwegian version (Jozefiak, Mattejat, & Remschmidt, 2012) of the ILC (Jozefiak, Larsson, Wichstrom, Mattejat, & Ravens-Sieberer, 2008; Mattejat & Remschmidt, 2006). The seven-item parent proxy-report and self-report inventories include one item for global evaluation of QoL and six items addressing the child’s physical and mental health, perception of activities when the child is alone, perceived relationships with friends and family, and functioning in school. The ILC yields a score on a 0 to 100 scale (0 = very low QoL, 100 = very high QoL). Reliability testing of the Norwegian version indicates satisfactory internal consistency (Cronbach’s α of parents report = 0.78, child report = 0.80-0.81) and a 2-week test–retest reliability of 0.86. Construct validity is also satisfactory.
Ethics
Written informed consent was obtained from adolescents and parents prior to inclusion, according to the study procedures in the CAP survey. Study approval was given by the Regional Committees for Medical and Health Research Ethics (reference numbers CAP survey: 4.2008.1393, present study: 2011/1772) and by the Norwegian Social Science Data Services (reference number CAP survey: 19976).
Statistical Analysis
According to ILC scoring guidelines for individuals with 1 to 3 items missing (less than 43% of each inventory), data were substituted by using expectation maximization algorithm procedures (Little & Rubin, 1987). Equivalent procedures were performed for the GFS scale. Across all variables, 0.1% to 6.1% of the data were missing.
Adjustment for SES in the cases with SES (n = 506) gave substantially similar results (results not shown). Therefore, SES was excluded from the analysis to maintain statistical power.
Differences between diagnostic group means were examined by one-way analysis of variance (ANOVA) and by analysis of covariance (ANCOVA), adjusting for age and gender. Post hoc tests were carried out according to the Sidak procedure. Normality of the residuals was confirmed by inspection of the Q–Q plots.
Two-sided p-value tests with p < .05 were taken to indicate statistical significance.
Results
Descriptive analyses of the data showed that age distribution of the adolescents were similar for all groups, while the sex distribution was different. Mothers were the responding parent in the main number of cases, and the response rate was lower for the ADHD + COND group. Educational levels among the four groups were similar (Table 1). Adolescent self-reports of family functioning were similar for unadjusted (Table 2) and adjusted results for age and gender (Table 3). Family functioning among adolescents with ADHD without additional emotional or conduct problems (ADHD Only; M = 21.76, 95% CI [20.3, 23.3]) was significantly better than for adolescents with ADHD and additional emotional and conduct problems (ADHD + EMO + COND; M = 29.30, CI 95% [27.2, 31.4]) and ADHD and additional conduct problems (ADHD + COND; M = 25.98, 95% CI [23.7, 28.3]; Table 3). Adolescents with ADHD and additional emotional problems (ADHD+EMO; M = 24.03, 95% CI [22.1, 25.9]) reported significantly better family functioning than adolescents with ADHD + EMO + COND. Adolescents with ADHD + EMO + COND reported significantly lower QoL (M = 51.34, 95% CI [46.4, 56.3]) than all the other ADHD groups after adjusting for age and gender (Table 3).
Descriptive Data of the Study Sample.
Note. SES = socioeconomic status (Highest educational level of parents); ADHD Only = ADHD without additional emotional or conduct problems; ADHD + EMO = ADHD with additional emotional problems; ADHD + COND = ADHD with additional conduct problems; ADHD + EMO + COND = ADHD with additional emotional and conduct problems.
Means of Self- and Parents Reports of Family Functioning and QoL.
Note. QoL = quality of life; ADHD Only = ADHD without additional emotional or conduct problems; ADHD + EMO = ADHD with additional emotional problems; ADHD + COND = ADHD with additional conduct problems; ADHD + EMO + COND = ADHD with additional emotional and conduct problems; GFS = General Functioning Scale of the McMaster family assessment device.
Means of Self- and Parents Reports of Family Functioning and QoL, Adjusted for Age and Gender.
Note. Marginal means, adjusted for age and gender. QoL = quality of life; ADHD Only = ADHD without additional emotional or conduct problems; ADHD + EMO = ADHD with additional emotional problems; ADHD + COND = ADHD with additional conduct problems; ADHD + EMO + COND = ADHD with additional emotional and conduct problems; GFS = General Functioning Scale of the McMaster family assessment device.
Parents reported no significant differences in family functioning between the groups in either unadjusted or adjusted analysis. Parents reported lower QoL for adolescents with ADHD + EMO + CD (M = 56.78, 95% CI [50.6, 63.0]) than for the ADHD Only group (M = 72.52, 95% CI [68.4, 76.6]; Table 2).
A paired t-test showed that compared with parents, adolescents with ADHD and additional problems reported significantly worse family functioning (Table 4). Adolescents from the ADHD Only group reported (M = 78.9, 95% CI [77.1, 80.6]) significantly better QoL than parents (M = 73.0, 95% CI [71.1, 75.0]), while parents and adolescents from all ADHD groups with additional problems reported similar levels of QoL (Table 4).
Comparison of Self- and Parents-reports by Paired Samples t-Test.
Note. ADHD Only = ADHD without additional emotional or conduct problems; ADHD + EMO = ADHD with additional emotional problems; ADHD + COND = ADHD with additional conduct problems; ADHD + EMO + COND = ADHD with additional emotional and conduct problems; GFS = General Functioning Scale of the McMaster family assessment device; QoL = quality of life.
Discussion
The aim of this study was to assess the impact of self-reported coexisting problems on family functioning and QoL among adolescents with ADHD by using multiple informants. Our results revealed that coexisting problems worsened the self-reported family functioning and QoL of adolescents with ADHD. Thus, adolescents with emotional and conduct problems reported lower QoL and family functioning. While the parents reported similar levels of QoL as the adolescents, the parents’ family functioning reports revealed no impact of coexisting problems. These results suggest that there might be differences in perspectives that should be considered in the assessment, diagnosis, and treatment of adolescents with ADHD.
Our study is the first to investigate self-reports of family functioning among adolescents with ADHD and coexisting emotional and conduct problems. Former studies of children with ADHD and additional emotional and conduct problems have revealed worse psychosocial functioning, more severe symptoms, and increased risk of other psychiatric disorders (Biederman et al., 2009; Peyre, Speranza, Cortese, Wohl, & Purper-Ouakil, 2015) indicating that this is a high-risk group. According to standardized cut-off scores of the GFS, our results show that adolescents with ADHD and additional problems reported unhealthy levels of family functioning (Ryan et al., 2005). Lower family functioning may indicate problems caused by the adolescents’ symptoms. However, it is important to consider parental factors too (Faraone, Kunwar, Adamson, & Biederman, 2009). A recent study found that ADHD symptoms were related to poorer family functioning (van Steijn, Oerlemans, van Aken, Buitelaar, & Rommelse, 2013). Studies of children with ADHD with and without additional conduct problems have shown that child ADHD plus conduct problems were associated with problems within the family (Chronis et al., 2003). In the present study additional conduct problems worsened self-reported family functioning, which is in agreement with findings in recent research (Deault, 2010). Thus, our findings add further evidence to the view that an increased awareness of ADHD patients with coexisting problems is needed in clinical assessment and research.
In our sample, parents reported no significant differences in family functioning between the groups, revealing no impact of coexisting problems. This result is in contrast to the report of the adolescents and shows that self-reports added important information to the results. Deault (2010) discussed the importance of assessing multiple informants, as there may be multiple risk factors for poor family functioning present within the same family (Deault, 2010). The adolescents’ psychiatric disorders could have limited their ability to assess themselves and thereby biased their reports. However, importantly, by excluding informants, particular aspects of disorders may be missed (van der Ende & Verhulst, 2005). Furthermore, concerning patients with internalizing disorders, differences between informants typically become greater with age (van der Ende, Verhulst, & Tiemeier, 2012), with parents reporting less problems. Externalizing disorders have been shown to be more burdensome for the parents than anxiety and depressive disorders, and parental burden associated with child psychiatric disorders is related to the use of specialist mental health services (Angold et al., 1998). Furthermore, maternal depression is associated with maternal reports of worse child behavior (Müller, Achtergarde, & Furniss, 2011). Because parental reports can be biased, it is important to include their children’s self-reports for more accurate assessments of well-being and functioning, especially during adolescence. Thus, in this study the value of self-reports was considered important and was used to categorize the ADHD group according to emotional and conduct problems. By including both perspectives on outcome measures, clinicians can improve the foundation for therapeutic discussion with the adolescent.
A higher level of coexisting psychiatric problems with ADHD in our sample had a significant impact on the QoL reports. These findings are consistent with a large study of ADHD and coexisting psychiatric problems (Steinhausen et al., 2006). In addition to parent reports, our results also included adolescents’ self-reports, which were not included in former studies. Our findings showed that adolescents’ self-reports and their parent reports regarding coexisting problems had similar impact on QoL. Nevertheless, parents and adolescents made their assessments based on different perspectives. As stated in a recent practitioner review by Coghill and coworkers, “Child self-completed QoL scales could help identify which outcomes are most important as treatment targets and provide a more ecologically valid measure of the positive and negative impact of a treatment regime. Using parent ratings can provide a sense of the wider impact of the treatment regime” (Coghill et al., 2009). It has previously been reported that most adolescents with chronic conditions and parents agreed on adolescents QoL, and as adolescents are expected to become partners in their own health care, the perceptions of their own QoL are important (Sattoe, van Staa, & Moll, 2012). The QoL concept focuses on subjective feelings, and physical and mental health providers are increasingly regarding QoL as one of the main aspects of health (World Health Organization, 1995). Even though improved QoL is associated with reduction of symptoms, long-term effects of treatment remain relatively unexplored (Buitelaar & Medori, 2010).
The findings of the present study are limited by the rather small sample sizes for some of the diagnostic groups, and by a low response rate, which could lead to imprecise results. Nevertheless, the reason for referral did not differ from the population of patients treated in the clinic during the study period. Response rate for parent reports differed between ADHD groups, as shown in Table 1. Thus, parent reports of the ADHD + COND group could be biased, and results should be interpreted with caution. Among the parents represented in our study, parental educational level had overlapping CI for all four groups. Family structure and clinical interview/information about the parents were not obtained, so we could not adjust for parental ADHD, or other chronic conditions in the analysis. Another limitation of the study is the cross-sectional design, which does not allow for causal inferences based on the data. A longitudinal study would allow for assessment of reciprocal relationships between the variables, and development of family functioning and QoL. Finally, the diagnostic groups were based on clinical ICD-10 diagnoses rather than on standardized semistructured child psychiatric interviews. Interrater reliability scores are not available; however, all diagnoses were set by an experienced child and adolescent psychiatrist or a clinical psychologist based on standard national and international guidelines.
In conclusion, a higher level of coexisting psychiatric problems had a significant impact on the QoL and family functioning reports by adolescents with ADHD. Furthermore, by including self-reports in the assessment, new information became available. A large percentage of the adolescents treated in child and adolescent psychiatric clinics are diagnosed with ADHD, and as patients, their perspectives of emotional and conduct problems, family functioning, and QoL should be acknowledged. In a time when clinical practices are diversifying, providing health care for adolescents with complex, chronic disorders is challenging. The findings of the present study suggest that multidimensional and multi-informant perspectives generate useful information that may help to optimize clinical assessment.
Footnotes
Declaration of Conflicting Interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Torunn Stene Nøvik served in an advisory board for Eli Lilly. She received research support from Eli Lilly, speaker’s fees from Eli Lilly, GlaxoSmithKline, and Lundbeck and conference attendance support from Eli Lilly, Janssen-Cilag, and Bristol-Myers Squibb. The present work is unrelated to the above grants or relationships. The other authors reported no conflicts of interest.
Funding
The author(s) declared receipt of the following financial support for the research, authorship, and/or publication of this article: This study was financed by a PhD grant awarded to the first author by the Department of Research and Development (AFFU), Division of Mental Health Care, St. Olav's University Hospital Trondheim.