Infant Behavior & Development 39 (2015) 136–147
Contents lists available at ScienceDirect
Infant Behavior and Development
Full length article
Finnish mothers’ and fathers’ reports of their boys and girls by using the Brief Infant-Toddler Social and Emotional Assessment (BITSEA) Jaana Alakortes a,∗ , Jenni Fyrstén a,b , Alice S. Carter c , Irma K. Moilanen a,b , Hanna E. Ebeling a,b a b c
Department of Child Psychiatry, Institute of Clinical Medicine, University of Oulu, Box 26, 90029 OYS, Oulu, Finland Clinic of Child Psychiatry, Oulu University Hospital, Oulu, Finland Department of Psychology, University of Massachusetts Boston, Boston, MA, United States
a r t i c l e
i n f o
Article history: Received 14 August 2014 Received in revised form 1 February 2015 Accepted 10 February 2015 Available online 28 March 2015 Keywords: BITSEA Infant Toddler Social–emotional Behavior problems Competence
a b s t r a c t This study investigated maternal and paternal reports about their very young boys and girls on the Brief Infant-Toddler Social and Emotional Assessment (BITSEA). Two samples were recruited through child health centers in Northern Finland. The infant sample consisted of 227 children (112 boys and 115 girls) (mean age 13.0 ± 1.1 months) and the toddler sample consisted of 208 children (94 boys and 114 girls) (mean age 19.3 ± 1.4 months). Among the infants, girls obtained higher paternal competence total scores than boys, whereas among the toddlers, both maternal and paternal competence total scores were higher for girls compared to boys. In the problem total scale, boys were scored higher than girls by mothers, but not by fathers, in both age groups. In the externalizing problem domain, maternal scores were higher for boys compared to girls among both samples, whereas paternal scores were significantly higher for boys than for girls only among the infants. Also maternal internalizing problem scores were higher for boys than for girls among the toddlers. Compared to fathers, mothers perceived more social–emotional competencies in toddler boys and girls, as well as more total, externalizing and dysregulation problems in toddler boys. However, significant differences between the maternal and paternal BITSEA ratings were not found among the infants of either sex. The results suggest that sex differences in the social–emotional/behavior domain may be observed by the parents among children as young as 11 to 24 months of age. Our findings highlight the importance of paying attention to probable sex differences when assessing and treating early social–emotional/behavior problems. © 2015 Elsevier Inc. All rights reserved.
1. Introduction During the last decades researchers and clinicians have become convinced that clinically significant social–emotional and behavioral (SEB) problems exist in early childhood, even among infants and toddlers younger than 3 years of age. These
Abbreviations: BITSEA, Brief Infant–Toddler Social and Emotional Assessment; CBCL, Child Behavior Checklist; ITSEA, Infant–Toddler Social and Emotional Assessment; SD, standard deviation; SEB, social–emotional and behavior. ∗ Corresponding author. Tel.: +358 40 5538306. E-mail address: [email protected] (J. Alakortes). http://dx.doi.org/10.1016/j.infbeh.2015.02.016 0163-6383/© 2015 Elsevier Inc. All rights reserved.
J. Alakortes et al. / Infant Behavior & Development 39 (2015) 136–147
137
observations have contributed to the creation of age appropriate, developmentally sensitive assessment tools and measures for systematic and broad evaluation of these problems (DelCarmen-Wiggins & Carter, 2001), such as The Brief Infant-Toddler Social and Emotional Assessment (BITSEA) (Briggs-Gowan, Carter, Irwin, Wachtel, & Cicchetti, 2004; Briggs-Gowan & Carter, 2006). Longitudinal studies have increased the body of evidence that a substantial proportion of early SEB problems are stable over time (Baillargeon, Keenan, & Cao, 2012; Briggs-Gowan, Carter, Bosson-Heenan, Guyer, & Horwitz, 2006; Lavigne et al., 1998; Mathiesen & Sanson, 2000; Mesman & Koot, 2001; Mäntymaa et al., 2012; Pihlakoski et al., 2006; Van Zeijl et al., 2006). For example, in a community sample of 1 to 3-year-old children (n = 1082) from the USA, approximately half of the children who had high parent-reported SEB problems continued to have such psychopathology one year later, even among the youngest, 1-year-old children (Briggs-Gowan et al., 2006). In a general population study from the Netherlands (n = 332), Mesman and Koot (2001) found that parent-reported internalizing and externalizing problems in 2 to 3-year-old children were predictive of their corresponding DSM-IV diagnoses (Shaffer, Fisher, & Lucas, 1998) 8 years later, even independent of the influence of early parent-reported adverse parenting characteristics (e.g., negative maternal attitude) and general family risk factors (e.g., family psychopathology and low socioeconomic status). The importance of early identification and interventions of SEB problems in infancy and toddlerhood has been highlighted also by associations with delayed social–emotional competence and disruptions in family life (Briggs-Gowan, Carter, Skuban, & Horwitz, 2001). Prevalence rates of parent-reported SEB problems in 2 to 3-year-old children have been found to range from approximately 5% to 24%, usually settling around 10% to 15% (Briggs-Gowan et al., 2001; Earls, 1980; Erol, Simsek, Oner, & Munir, 2005; Koot & Verhulst, 1991; Larson, Pless, & Miettinen, 1988; Lavigne et al., 1996; Sourander, 2001; Stallard, 1993). Studies reporting prevalence rates of mental health problems for children younger than 2 years are very scarce and vary in methods. In studies relying on parent-report measures, the prevalence rates of these problems have varied from approximately 4% to 14% (Baillargeon, Sward, Keenan, & Cao, 2011; Bayer, Hiscock, Ukoumunne, Price, & Wake, 2008; Briggs-Gowan et al., 2001; Mathiesen & Sanson, 2000). For example, Mathiesen and Sanson (2000) examined early behavior problems in a population based sample of 18-month-old Norwegian children (n = 750) and found prevalence rates ranging from approximately 6% to 14% for 4 dimensions of maladjustment (social adjustment, emotional adjustment, overactive-inattentive and regulation) by applying the Behavior Checklist (BCL) (Richman & Graham, 1971). In Denmark, Skovgaard et al. (2007) investigated a random sample of 1.5-year-old children (n = 211) from the Copenhagen Child Cohort 2000 and diagnosed mental health problems in 16–18% of these toddlers. Compared to the large body of evidence regarding older children’s SEB problems (e.g., Rescorla et al., 2007), parallel trends of significant sex differences have been found in some studies among 2 to 3-year-olds (Erol et al., 2005; Koot & Verhulst, 1991; Lavigne et al., 1996; Sourander, 2001; Stallard, 1993). According to these toddler studies, overall mental health problems and/or externalizing problems have been more common in boys than in girls, whereas internalizing and/or dysregulation problems may be more common in girls than in boys. Among children younger than 2 years, significant sex differences in the prevalence rates of SEB problems have not usually been found (e.g., Baillargeon et al., 2011; Briggs-Gowan et al., 2001; Skovgaard et al., 2007). However, Carter, Briggs-Gowan, Jones, and Little (2003) reported significant sex differences in parental ratings on the Infant-Toddler Social and Emotional Assessment (ITSEA) for a representative birth cohort sample of 12 to 36-month-old children (n = 1235) from the USA; boys were rated higher than girls in activity/impulsivity, whereas girls were rated higher than boys on anxiety and most competence scales. There are few studies addressing paternal reports about their infant/toddler-age children’s SEB problems. In their metaanalysis, Achenbach, McConaughy, and Howell (1987) found a moderate correlation between mothers’ and fathers’ ratings of their 1.5 to 19-year-old children’s emotional/behavior problems (mean r = 0.59), with no significant difference between the mean correlations for overcontrolled (mean r = 0.59) and undercontrolled (mean r = 0.62) problems. However, results of a more recent meta-analysis suggested that the correspondence (effect sizes) between maternal and paternal ratings was moderate for internalizing and high for externalizing and total behavior problems among 3 to 19-year-old children (Duhig, Renk, Epstein, & Phares, 2000). According to this later meta-analysis, interparental agreement was also higher for adolescents than for children in early (3–5-year-olds) and middle (6–12-year-olds) childhood (Duhig et al., 2000). Concerning 1 to 3-yearold children, moderate to high interparental agreement was reported for the ITSEA with intraclass correlation coefficient (ICC) ranging from 0.43 to 0.79 for scales and domains (Carter et al., 2003). Correspondingly, for the BITSEA ICC ranged from 0.70 for boys to 0.78 for girls for the problem total score and from 0.58 for girls to 0.67 for boys for the competence total score (Briggs-Gowan & Carter, 2006). In terms of discrepancy, mothers tended to report slightly more behavior problems about their offspring (aged 3–19 years) than fathers (Duhig et al., 2000). Some later studies have reported similar findings (Luoma, Koivisto, & Tamminen, 2004b; Van der Valk, van den Oord, Verhulst, & Boomsma, 2001), particularly concerning boys (Luoma et al., 2004b). Parental reports generally play an invaluable role in young children’s mental health assessments, because many challenges exist concerning these assessments compared to older children and adults. For example, young children have very limited verbal and cognitive abilities to express their thoughts and feelings (Carter, Godoy, Marakovitz, & Briggs-Gowan, 2009). In addition, young child’s behavior during a short office visit in an unfamiliar setting may not be representative of behavior in day-to-day settings (Briggs-Gowan et al., 2004), and professionals miss a substantial number of serious cases with early SEB problems (Klein Velderman, Crone, Wiefferink, & Reijneveld, 2010). Parental worry has been shown to play a central role in help-seeking for children with behavior problems (Ellingson, Briggs-Gowan, Carter, & Horwitz, 2004). However, many parents have insufficient developmental knowledge to distinguish between normative misbehaviors and clinically
138
J. Alakortes et al. / Infant Behavior & Development 39 (2015) 136–147
concerning SEB problems in young children (Ellingson et al., 2004). For example, parents’ educational or occupational level does not necessarily increase parental perception of children’s problem behaviors and utilization of child mental health services (Verhulst & van der Ende, 1997). In turn, parental psychopathology may lower a parent’s threshold for assessing his/her child’s behavior as problematic (Verhulst & van der Ende, 1997), but even if the parent’s evaluation was not consistent with the child’s actual behavior (Müller, Achtergarde, & Furniss, 2011), parental early negative expectations and perceptions may set the child at elevated risk for future SEB problems (Mesman & Koot, 2001; Luoma et al., 2004a). Consequently, utilization of standardized, norm-referenced parent-report screening questionnaires may improve early identification of mental health problems among young children, and do so in a time-efficient and cost-effective way (Briggs-Gowan et al., 2004). The BITSEA is a screening tool that was developed for identifying possible SEB problems and/or delays or deficits in social–emotional competence in children ages 12 to 35 months (Briggs-Gowan et al., 2004; Briggs-Gowan & Carter, 2006). (For brevity, “deficits” will be used to refer to “delays and deficits”). The BITSEA was derived from a longer and more comprehensive questionnaire, the ITSEA (Carter & Briggs-Gowan, 2006), especially for use in settings requiring a more timeefficient and concise screening tool. The BITSEA may be used e.g., as a part of routine examinations at preventive child-health visits (Kruizinga, Jansen, Mieloo, Carter, & Raat, 2013) or in early intervention settings (Briggs-Gowan & Carter, 2007), and most parents are capable of completing it independently in a few minutes (Briggs-Gowan & Carter, 2006). The BITSEA has some special strengths. It is brief and easy to administer, score and interpret (Briggs-Gowan & Carter, 2006). Besides behavioral problems, the BITSEA addresses deficits in acquisition of social–emotional competencies and behaviors that may be indicative of autism spectrum disorders. The BITSEA has evidenced acceptable psychometric properties (e.g., internal consistency, test–retest and inter-rater reliability, one-year stability, concurrent validity relative to the Child Behavior Checklist (CBCL)/1.5–5 (Achenbach & Rescorla, 2000) scores and evaluator ratings) (Briggs-Gowan et al., 2004; Briggs-Gowan & Carter, 2006). Results of a recent study from the USA concerning clinical validity of the BITSEA indicated that the BITSEA problem index has fair to excellent sensitivity and specificity relative to diagnosis by “gold standard” interview and clinical-range CBCL scores (Briggs-Gowan et al., 2013). The BITSEA has also demonstrated excellent sensitivity and good specificity in detecting autistic disorder (Briggs-Gowan & Carter, 2006). Finally, in a large birth cohort sample (n = 1004 children) from the USA, the BITSEA screen status at 12 to 36 months of age was found to predict parent- and teacher-reported psychiatric problems and disorders at early elementary school age (Briggs-Gowan & Carter, 2008). Recent studies have supported the reliability and validity of the BITSEA as a screening measure of young children’s SEB problems also in Turkey (Karabekiroglu et al., 2009; Karabekiroglu, Briggs-Gowan, Carter, Rodopman-Arman, & Akbas, 2010) and in the Netherlands (Kruizinga et al., 2012). In a Finnish pilot study concerning 18-month-old children (n = 50), the BITSEA problem index had good internal consistency, and correlations between the BITSEA problem total and CBCL/1.5–5 internalizing, externalizing and total problem scores were evident (Haapsamo et al., 2009). When significant sex differences have been found in the earlier BITSEA studies, girls have been rated higher than boys in competence scores, whereas problem total scores have been higher for boys than for girls (Briggs-Gowan et al., 2004; Briggs-Gowan & Carter, 2006; Karabekiroglu et al., 2009, 2013; Kruizinga et al., 2012, 2013). To our knowledge, only Turkish researchers have evaluated possible differences between the maternal and paternal BITSEA ratings so far, and they found no significant differences between the maternal and paternal scores in any age or sex groups (Karabekiroglu et al., 2009). As described above, there are only few studies focusing on possible sex differences in the occurrence and appearance of SEB problems among children younger than 2 years of age and comparing mothers’ and fathers’ reports about their young children’s SEB problems and competencies. The main purpose of the current study was to compare boys’ and girls’ BITSEA problem and competence scores among 12-month-old infants and 18-month-old toddlers recruited through public child health centers in the city of Oulu, Finland. The second aim was to investigate probable differences in parents’ perceptions of their young children’s SEB problems and skills related to parental and child sex. 2. Methods 2.1. Setting The study was carried out in Oulu which is the capital of northern Finland, in Scandinavia. Between 2006 and 2011 the population of Oulu increased from 129 000 to 142 000 and the birth rate was about 2000 live born children per annum (www.ouka.fi/Statistical Yearbook of the City of Oulu, 2011). The Finnish public child health center organization covers all municipalities in Finland and it is free of charge for families. Almost all children in Oulu attend regular well-child visits at child health centers during their first six years of life, before the school age. Thus, the data collection was carried out in collaboration with nurses at all child health centers of Oulu. 2.2. Procedure The BITSEA data collection was a part of a research project concerning early identification of young children’s mental health problems in Northern Finland launched in 2008. The current paper is the first publication regarding the project results. The research was approved by the Ethical Committee of the University Hospital of Oulu and the Municipal Boards of the Social and Health Care Units of the City of Oulu. The active BITSEA data collection took place from February 2008 to March 2009
J. Alakortes et al. / Infant Behavior & Development 39 (2015) 136–147
139
for 18-month-old toddlers and from February 2010 to February 2011 for 12-month-old infants. After training, health care nurses delivered the BITSEA (a red copy to mothers and a blue copy to fathers) and background information questionnaires to parents during routine well-child visits. Parents mailed completed questionnaires and a written informed consent form to researchers in a pre-paid envelope. Parents could also leave their informed consent form with contact information with the nurse; in these cases, when questionnaires were not returned, researchers telephoned parents and remailed materials to encourage participation. The BITSEA data was used only for research purposes and nurses were not aware of the children’s BITSEA scores during the well-child visits. Participation for both nurses and parents was completely voluntary. 2.3. Participants In 2010–2011 the target population was all 12-month-old children and their parents who participated in the child’s well-child visit in the city of Oulu. Unfortunately, it was not possible to get reliable information about the actual number of parents who were invited in the study by nurses. However, given the facts that almost all children in Oulu attend these well-child visits and the data collection was performed during about one year, the number of 1-year-old population in Oulu at the end of the year 2010 (n = 1964) was used as an estimated number of the target population. Similarly, in 2008–2009 the target population was all 18-month-old children and their parents who participated in the child’s well-child visit in Oulu, and the number of 1-year-old population in Oulu at the end of the year 2008 (n = 1757) was used as an estimation of the target population. The numbers of the 1-year-olds were obtained from the Statistics Finland (www.stat.fi). An initial infant sample consisted of 230 children and toddler sample consisted of 209 children. Thus, both samples represented ∼10% of the estimated target populations. Three subjects were excluded from the initial infant sample due to out of range ages (two adjusted ages were under 11 months and one child was over 18 months). One child who was over 24 months was excluded from the initial toddler sample. Thus, the final infant sample consisted of 227 children, 112 boys (49.3%) and 115 girls (50.7%), with a mean age of 13.0 months. The final toddler sample consisted of 208 children, 94 boys (45.2%) and 114 girls (54.8%), with a mean age of 19.3 months. The maternal BITSEA data was available for all children in both samples. The paternal BITSEA data was available for 205 of 227 (90.3%) infants, 104 of 112 (92.9%) boys and 101 of 115 (87.8%) girls, and for 181 of 208 (87.0%) toddlers, 82 of 94 (87.2%) boys and 99 of 114 (86.8%) girls. Among the final samples, child age was adjusted for 4 infants (2 boys and 2 girls) and 5 toddlers (3 boys and 2 girls) born at 36 weeks or less gestation. None of the children had diagnoses of severe primary diseases or developmental delays. Every child whose Apgar score status was available to the researchers had the last given Apgar score ≥6. More detailed information concerning child age, weight at birth, Apgar scores, as well as parental age and educational level is shown in Table 1. There were no significant differences between the infant sample boys and girls, or between the toddler sample boys and girls, concerning child age, parental age and educational level. Moreover, no significant differences were found between the infant and toddler boys, or between the infant and toddler girls, in relation to weight at birth, parental age and educational level. Based on the background information data, no significant differences were found between the infants with and without the paternal BITSEA data concerning child sex and age, or parental age and educational level. Significant differences were neither found between the toddlers with and without the paternal BITSEA data concerning child sex, age and weight at birth, or parental age and maternal educational level. However, the infant girls with the paternal BITSEA data were somewhat lighter at birth than the infant girls without the paternal BITSEA data (p = 0.001), and the fathers of toddler girls with the paternal BITSEA data were somewhat higher educated than the fathers of toddler girls without the paternal BITSEA data (p = 0.015). Although both samples represented only ∼10% of the target populations (18-month-old inhabitants of Oulu in 2008–2009 and 12-month-old inhabitants of Oulu in 2010–2011), no significant differences were found between the final study samples and the corresponding target populations in relation to the two available child characteristics: (1) children’s mean weight at birth and (2) mothers’ mean age at the child’s birth, except for the infant sample boys who were on average slightly heavier at birth than the target population boys born in 2009 or 2010 (p = 0.032 or 0.027, respectively, but the average effect size was small, d = 0.209). Background characteristics of the target populations were obtained from the Medical Birth Register at The National Institute for Health and Welfare (THL). Additionally, there were no significant differences between the educational levels of the mothers in the study samples and the target females, or between the educational levels of the fathers of the toddler girls and the target males, based on data on the target populations (30–34-year-old citizens of Oulu in 2008–2009 or 2010–2011), by applying the Statistics Finland (www.stat.fi). However, fathers of infant boys and girls, and of toddler boys, were somewhat higher educated than the target males (with p-values of 0.019–0.029, 0.003–0.010, and 0.045–0.048, respectively). The corresponding average effect sizes were at the medium level (w = 0.262, 0.309, and 0.267, respectively). Finally, 2.7% (12/446) of the infant sample parents and 2.1% (8/390) of the toddler sample parents were immigrants, which corresponded percentages of the non-native population in Oulu in the data collection years (www.ouka.fi/ Statistical Yearbook of the City of Oulu 2008–2011). 2.4. Measures 2.4.1. The BITSEA The BITSEA questionnaire (Briggs-Gowan & Carter, 2006) consists of 42 items; 31 items concern SEB problems and 11 items concern deficits in social–emotional competence. The problem items address problematic behavior in externalizing
140
J. Alakortes et al. / Infant Behavior & Development 39 (2015) 136–147
Table 1 Background characteristics of the study samples. Child variables
Infant boys (n = 112) Mean (SD; range)
Infant girls (n = 115) Mean (SD; range)
Infant total (n = 227) Mean (SD; range)
Toddler boys (n = 94) Mean (SD; range)
Toddler girls (n = 114) Mean (SD; range)
Toddler total (n = 208) Mean (SD; range)
Age when BITSEA completed (months)
12.9 (1.1; 11.5–17.6) 3655 (520; 2150–5200) 1 n (%) 106 (94.6)
13.0 (1.2; 11.2–17.9) 3476 (488; 2100–4490) 0 n (%) 111 (96.5)
13.0 (1.1; 11.2–17.9) 3564 (511; 2100–5200) 1 n (%) 217 (95.6)
19.2 (1.3; 16.4–24.0) 3657 (515; 2185–5220) 1 n (%) 85 (90.4)
19.4 (1.4; 16.5–23.7) 3462 (482; 2160–4615) 0 n (%) 103 (90.4)
19.3 (1.4; 16.4–24.0) 3550 (505; 2160–5220) 1 n (%) 188 (90.4)
Parent variables
Infant boys (n = 112) Mean (SD; range)
Infant girls (n = 115) Mean (SD; range)
Infant total (n = 227) Mean (SD; range)
Toddler boys (n = 94) Mean (SD; range)
Toddler girls (n = 114) Mean (SD; range)
Toddler total (n = 208) Mean (SD; range)
Mother’s age at the child’s birth (years)
29.4 (5.7; 17–41) 0 31.4 (6.5; 19–47) 0 n (%) 4 (3.6) 45 (40.2) 63 (56.3) 0 (0) n (%) 4 (3.6) 56 (50.0) 50 (44.6) 2 (1.8)
29.0 (4.6; 17–40) 1 30.9 (5.6; 17–48) 1 n (%) 5 (4.3) 33 (28.7) 77 (67.0) 0 (0) n (%) 4 (3.5) 42 (36.5) 65 (56.5) 4 (3.5)
29.2 (5.2; 17–41) 1 31.2 (6.0; 17–48) 1 n (%) 9 (4.0) 78 (34.4) 140 (61.7) 0 (0) n (%) 8 (3.5) 98 (43.2) 115 (50.7) 6 (2.6)
29.0 (5.1; 20–42) 1 30.9 (6.3; 19–49) 4 n (%) 5 (5.3) 31 (33.0) 47 (50.0) 11 (11.7) n (%) 2 (2.1) 42 (44.7) 43 (45.7) 7 (7.4)
29.8 (5.5; 17–43) 1 31.4 (6.2; 21–47) 2 n (%) 7 (6.1) 31 (27.2) 68 (59.6) 8 (7.0) n (%) 5 (4.4) 43 (37.7) 56 (49.1) 10 (8.8)
29.4 (5.3; 17–43) 2 31.1 (6.2; 19–49) 6 n (%) 12 (5.8) 62 (29.8) 115 (55.3) 19 (9.1) n (%) 7 (3.4) 85 (40.9) 99 (47.6) 17 (8.2)
Weight at birth (g) - Missing data (n) Last given Apgar score (≥6)a
- Missing data (n) Father’s age at the child’s birth (years) - Missing data (n) Mother’s educational level (a) Elementary school (b) Vocational/senior high school (c) College/university (d) Undefined Father’s educational level (a) Elementary school (b) Vocational/senior high school (c) College/university (d) Undefined
Note: SD = standard deviation. a Every child whose Apgar score status was available to the researchers had the last given Apgar score ≥6.
(6 items; e.g., impulsivity, defiance, peer aggression), internalizing (8 items; e.g., fearfulness, worry, anxiety, sadness) and dysregulation (8 items; e.g., sleep and eating problems, negative emotionality, sensory sensitivities) domains, as well as rare behaviors that may be early markers of autism spectrum disorders (e.g., repeating of same action or phrase over and over without enjoyment) or other significant psychopathology (14 red flag items; e.g., hurts self on purpose). Some of the items address problematic behavior overlapping in two or three domains (e.g., avoids physical contact). Besides 9 problem items, the autism domain consists of 8 competence items (deficits in those). The competence items rate the child’s skills of attention, compliance, mastery motivation, prosocial peer relations, empathy, imitation/play, and social relatedness. Response alternatives for the BITSEA items are Not true/rarely (=0), Somewhat true/sometimes (=1), and Very true/often (=2), providing problem total score (range 0–62) and competence total score (range 0–22). High problem total score and/or low competence total score indicates that the assessed child may have SEB problems and/or deficits in competence, and follow-up or more comprehensive evaluation is needed to determine whether or not difficulties are clinically significant. The BITSEA cut score values are based on a national USA standardization sample for 12 to 36-month-olds (by sex and age in 6-month age bands) and set at the highest 25th percentile for problem total scores and at the lowest 15th percentile for competence total scores. In addition, the BITSEA form includes two questions addressing the parental level of worry about the child’s (1) psychosocial and (2) language development but not counting toward the BITSEA scores. The current study utilized the Finnish translation of the BITSEA developed for a previous Finnish BITSEA pilot study (Haapsamo et al., 2009). Because there are no standardized Finnish cut score values for the BITSEA, the USA standardized cut scores (Briggs-Gowan & Carter, 2006) were applied; for 12–17-month-olds competence total score ≤12 (for both sexes) and problem total score ≥13 (for both sexes), and for 18–23-month-olds competence total score ≤14 (for both sexes) and problem total score ≥15 (for boys) or ≥13 (for girls). If the child’s competence total score was at or below the competence cut score, it was in the possible deficit range. Correspondingly, if the child’s problem total score was at or above the problem cut score, it was in the possible problem range. All the infants, and the toddler sample subjects who were slightly under 18 months of age (11 boys and 11 girls rated by mothers, 11 boys and 9 girls rated by fathers), were screened by using the cut score values for 12–17-month-old children. The toddlers who were at minimum 18-month-olds were screened by using the cut score values for 18–23-month-old children. Consistent with scoring guidelines, child age was adjusted for children born at 36 weeks or less gestation. If ≥2 competence and/or ≥5 problem items per a rater were unanswered, the corresponding total score/-s could not be calculated and used in the analyses. Only the infants and toddlers whose both competence and problem total
J. Alakortes et al. / Infant Behavior & Development 39 (2015) 136–147
141
Table 2 Maternal and paternal mean scores and standard deviations of the BITSEA scales and domains for the infant and toddler sample boys and girls. Infant sample
Maternal
Paternal
(n mat/n pat/n both)
Mean (SD)
Mean (SD)
Total (224/198/197) Boys (110/100/99) Girls (114/98/98) pa /Effect size (d)c Total (223/200/198) Boys (109/103/101) Girls (114/97/97) pa /Effect size (d)c Total (224/202/201) Boys (111/103/103) Girls (113/99/98) pa /Effect size (d)c Total (223/203/200) Boys (109/104/102) Girls (114/99/98) pa /Effect size (d)c Total (225/202/202) Boys (111/103/103) Girls (114/99/99) pa /Effect size (d)c
Competence 15.30 (2.79) 14.95 (2.75) 15.64 (2.79) 0.062/– Problem total 8.15 (4.13) 8.63 (4.27) 7.69 (3.96) 0.046* /0.228 Externalizing 1.91 (1.58) 2.26 (1.65) 1.57 (1.43) 0.001* /0.447 Internalizing 1.17 (1.14) 1.17 (1.08) 1.18 (1.20) 0.767/– Dysregulation 3.10 (2.05) 3.04 (1.90) 3.16 (2.20) 0.657/–
pb /Effect size (d)c
Toddler sample
Maternal
Paternal
(n mat/n pat/n both)
Mean (SD)
Mean (SD)
15.07 (3.05) 14.58 (3.10) 15.57 (2.94) 0.022* /0.328
0.525/– 0.383/– 0.975/–
Total (207/181/180) Boys (93/82/81) Girls (114/99/99) pa /Effect size (d)c
7.67 (4.51) 7.95 (4.56) 7.36 (4.46) 0.313/–
0.277/– 0.252/– 0.701/–
Total (207/181/180) Boys (93/82/81) Girls (114/99/99) pa /Effect size (d)c
1.81 (1.56) 2.08 (1.67) 1.53 (1.39) 0.011* /0.358
0.737/– 0.388/– 0.595/–
Total (207/181/180) Boys (93/82/81) Girls (114/99/99) pa /Effect size (d)c
0.93 (1.07) 0.93 (1.18) 0.93 (0.96) 0.653/–
0.024* /0.157 0.173/– 0.071/–
Total (208/181/181) Boys (94/82/82) Girls (114/99/99) pa /Effect size (d)c
2.85 (2.10) 2.73 (1.97) 2.97 (2.24) 0.416/–
0.140/– 0.125/– 0.506/–
Total (207/181/180) Boys (93/82/81) Girls (114/99/99) pa /Effect size (d)c
Competence 18.02 (2.36) 17.28 (2.52) 18.63 (2.04)
Contents lists available at ScienceDirect
Infant Behavior and Development
Full length article
Finnish mothers’ and fathers’ reports of their boys and girls by using the Brief Infant-Toddler Social and Emotional Assessment (BITSEA) Jaana Alakortes a,∗ , Jenni Fyrstén a,b , Alice S. Carter c , Irma K. Moilanen a,b , Hanna E. Ebeling a,b a b c
Department of Child Psychiatry, Institute of Clinical Medicine, University of Oulu, Box 26, 90029 OYS, Oulu, Finland Clinic of Child Psychiatry, Oulu University Hospital, Oulu, Finland Department of Psychology, University of Massachusetts Boston, Boston, MA, United States
a r t i c l e
i n f o
Article history: Received 14 August 2014 Received in revised form 1 February 2015 Accepted 10 February 2015 Available online 28 March 2015 Keywords: BITSEA Infant Toddler Social–emotional Behavior problems Competence
a b s t r a c t This study investigated maternal and paternal reports about their very young boys and girls on the Brief Infant-Toddler Social and Emotional Assessment (BITSEA). Two samples were recruited through child health centers in Northern Finland. The infant sample consisted of 227 children (112 boys and 115 girls) (mean age 13.0 ± 1.1 months) and the toddler sample consisted of 208 children (94 boys and 114 girls) (mean age 19.3 ± 1.4 months). Among the infants, girls obtained higher paternal competence total scores than boys, whereas among the toddlers, both maternal and paternal competence total scores were higher for girls compared to boys. In the problem total scale, boys were scored higher than girls by mothers, but not by fathers, in both age groups. In the externalizing problem domain, maternal scores were higher for boys compared to girls among both samples, whereas paternal scores were significantly higher for boys than for girls only among the infants. Also maternal internalizing problem scores were higher for boys than for girls among the toddlers. Compared to fathers, mothers perceived more social–emotional competencies in toddler boys and girls, as well as more total, externalizing and dysregulation problems in toddler boys. However, significant differences between the maternal and paternal BITSEA ratings were not found among the infants of either sex. The results suggest that sex differences in the social–emotional/behavior domain may be observed by the parents among children as young as 11 to 24 months of age. Our findings highlight the importance of paying attention to probable sex differences when assessing and treating early social–emotional/behavior problems. © 2015 Elsevier Inc. All rights reserved.
1. Introduction During the last decades researchers and clinicians have become convinced that clinically significant social–emotional and behavioral (SEB) problems exist in early childhood, even among infants and toddlers younger than 3 years of age. These
Abbreviations: BITSEA, Brief Infant–Toddler Social and Emotional Assessment; CBCL, Child Behavior Checklist; ITSEA, Infant–Toddler Social and Emotional Assessment; SD, standard deviation; SEB, social–emotional and behavior. ∗ Corresponding author. Tel.: +358 40 5538306. E-mail address: [email protected] (J. Alakortes). http://dx.doi.org/10.1016/j.infbeh.2015.02.016 0163-6383/© 2015 Elsevier Inc. All rights reserved.
J. Alakortes et al. / Infant Behavior & Development 39 (2015) 136–147
137
observations have contributed to the creation of age appropriate, developmentally sensitive assessment tools and measures for systematic and broad evaluation of these problems (DelCarmen-Wiggins & Carter, 2001), such as The Brief Infant-Toddler Social and Emotional Assessment (BITSEA) (Briggs-Gowan, Carter, Irwin, Wachtel, & Cicchetti, 2004; Briggs-Gowan & Carter, 2006). Longitudinal studies have increased the body of evidence that a substantial proportion of early SEB problems are stable over time (Baillargeon, Keenan, & Cao, 2012; Briggs-Gowan, Carter, Bosson-Heenan, Guyer, & Horwitz, 2006; Lavigne et al., 1998; Mathiesen & Sanson, 2000; Mesman & Koot, 2001; Mäntymaa et al., 2012; Pihlakoski et al., 2006; Van Zeijl et al., 2006). For example, in a community sample of 1 to 3-year-old children (n = 1082) from the USA, approximately half of the children who had high parent-reported SEB problems continued to have such psychopathology one year later, even among the youngest, 1-year-old children (Briggs-Gowan et al., 2006). In a general population study from the Netherlands (n = 332), Mesman and Koot (2001) found that parent-reported internalizing and externalizing problems in 2 to 3-year-old children were predictive of their corresponding DSM-IV diagnoses (Shaffer, Fisher, & Lucas, 1998) 8 years later, even independent of the influence of early parent-reported adverse parenting characteristics (e.g., negative maternal attitude) and general family risk factors (e.g., family psychopathology and low socioeconomic status). The importance of early identification and interventions of SEB problems in infancy and toddlerhood has been highlighted also by associations with delayed social–emotional competence and disruptions in family life (Briggs-Gowan, Carter, Skuban, & Horwitz, 2001). Prevalence rates of parent-reported SEB problems in 2 to 3-year-old children have been found to range from approximately 5% to 24%, usually settling around 10% to 15% (Briggs-Gowan et al., 2001; Earls, 1980; Erol, Simsek, Oner, & Munir, 2005; Koot & Verhulst, 1991; Larson, Pless, & Miettinen, 1988; Lavigne et al., 1996; Sourander, 2001; Stallard, 1993). Studies reporting prevalence rates of mental health problems for children younger than 2 years are very scarce and vary in methods. In studies relying on parent-report measures, the prevalence rates of these problems have varied from approximately 4% to 14% (Baillargeon, Sward, Keenan, & Cao, 2011; Bayer, Hiscock, Ukoumunne, Price, & Wake, 2008; Briggs-Gowan et al., 2001; Mathiesen & Sanson, 2000). For example, Mathiesen and Sanson (2000) examined early behavior problems in a population based sample of 18-month-old Norwegian children (n = 750) and found prevalence rates ranging from approximately 6% to 14% for 4 dimensions of maladjustment (social adjustment, emotional adjustment, overactive-inattentive and regulation) by applying the Behavior Checklist (BCL) (Richman & Graham, 1971). In Denmark, Skovgaard et al. (2007) investigated a random sample of 1.5-year-old children (n = 211) from the Copenhagen Child Cohort 2000 and diagnosed mental health problems in 16–18% of these toddlers. Compared to the large body of evidence regarding older children’s SEB problems (e.g., Rescorla et al., 2007), parallel trends of significant sex differences have been found in some studies among 2 to 3-year-olds (Erol et al., 2005; Koot & Verhulst, 1991; Lavigne et al., 1996; Sourander, 2001; Stallard, 1993). According to these toddler studies, overall mental health problems and/or externalizing problems have been more common in boys than in girls, whereas internalizing and/or dysregulation problems may be more common in girls than in boys. Among children younger than 2 years, significant sex differences in the prevalence rates of SEB problems have not usually been found (e.g., Baillargeon et al., 2011; Briggs-Gowan et al., 2001; Skovgaard et al., 2007). However, Carter, Briggs-Gowan, Jones, and Little (2003) reported significant sex differences in parental ratings on the Infant-Toddler Social and Emotional Assessment (ITSEA) for a representative birth cohort sample of 12 to 36-month-old children (n = 1235) from the USA; boys were rated higher than girls in activity/impulsivity, whereas girls were rated higher than boys on anxiety and most competence scales. There are few studies addressing paternal reports about their infant/toddler-age children’s SEB problems. In their metaanalysis, Achenbach, McConaughy, and Howell (1987) found a moderate correlation between mothers’ and fathers’ ratings of their 1.5 to 19-year-old children’s emotional/behavior problems (mean r = 0.59), with no significant difference between the mean correlations for overcontrolled (mean r = 0.59) and undercontrolled (mean r = 0.62) problems. However, results of a more recent meta-analysis suggested that the correspondence (effect sizes) between maternal and paternal ratings was moderate for internalizing and high for externalizing and total behavior problems among 3 to 19-year-old children (Duhig, Renk, Epstein, & Phares, 2000). According to this later meta-analysis, interparental agreement was also higher for adolescents than for children in early (3–5-year-olds) and middle (6–12-year-olds) childhood (Duhig et al., 2000). Concerning 1 to 3-yearold children, moderate to high interparental agreement was reported for the ITSEA with intraclass correlation coefficient (ICC) ranging from 0.43 to 0.79 for scales and domains (Carter et al., 2003). Correspondingly, for the BITSEA ICC ranged from 0.70 for boys to 0.78 for girls for the problem total score and from 0.58 for girls to 0.67 for boys for the competence total score (Briggs-Gowan & Carter, 2006). In terms of discrepancy, mothers tended to report slightly more behavior problems about their offspring (aged 3–19 years) than fathers (Duhig et al., 2000). Some later studies have reported similar findings (Luoma, Koivisto, & Tamminen, 2004b; Van der Valk, van den Oord, Verhulst, & Boomsma, 2001), particularly concerning boys (Luoma et al., 2004b). Parental reports generally play an invaluable role in young children’s mental health assessments, because many challenges exist concerning these assessments compared to older children and adults. For example, young children have very limited verbal and cognitive abilities to express their thoughts and feelings (Carter, Godoy, Marakovitz, & Briggs-Gowan, 2009). In addition, young child’s behavior during a short office visit in an unfamiliar setting may not be representative of behavior in day-to-day settings (Briggs-Gowan et al., 2004), and professionals miss a substantial number of serious cases with early SEB problems (Klein Velderman, Crone, Wiefferink, & Reijneveld, 2010). Parental worry has been shown to play a central role in help-seeking for children with behavior problems (Ellingson, Briggs-Gowan, Carter, & Horwitz, 2004). However, many parents have insufficient developmental knowledge to distinguish between normative misbehaviors and clinically
138
J. Alakortes et al. / Infant Behavior & Development 39 (2015) 136–147
concerning SEB problems in young children (Ellingson et al., 2004). For example, parents’ educational or occupational level does not necessarily increase parental perception of children’s problem behaviors and utilization of child mental health services (Verhulst & van der Ende, 1997). In turn, parental psychopathology may lower a parent’s threshold for assessing his/her child’s behavior as problematic (Verhulst & van der Ende, 1997), but even if the parent’s evaluation was not consistent with the child’s actual behavior (Müller, Achtergarde, & Furniss, 2011), parental early negative expectations and perceptions may set the child at elevated risk for future SEB problems (Mesman & Koot, 2001; Luoma et al., 2004a). Consequently, utilization of standardized, norm-referenced parent-report screening questionnaires may improve early identification of mental health problems among young children, and do so in a time-efficient and cost-effective way (Briggs-Gowan et al., 2004). The BITSEA is a screening tool that was developed for identifying possible SEB problems and/or delays or deficits in social–emotional competence in children ages 12 to 35 months (Briggs-Gowan et al., 2004; Briggs-Gowan & Carter, 2006). (For brevity, “deficits” will be used to refer to “delays and deficits”). The BITSEA was derived from a longer and more comprehensive questionnaire, the ITSEA (Carter & Briggs-Gowan, 2006), especially for use in settings requiring a more timeefficient and concise screening tool. The BITSEA may be used e.g., as a part of routine examinations at preventive child-health visits (Kruizinga, Jansen, Mieloo, Carter, & Raat, 2013) or in early intervention settings (Briggs-Gowan & Carter, 2007), and most parents are capable of completing it independently in a few minutes (Briggs-Gowan & Carter, 2006). The BITSEA has some special strengths. It is brief and easy to administer, score and interpret (Briggs-Gowan & Carter, 2006). Besides behavioral problems, the BITSEA addresses deficits in acquisition of social–emotional competencies and behaviors that may be indicative of autism spectrum disorders. The BITSEA has evidenced acceptable psychometric properties (e.g., internal consistency, test–retest and inter-rater reliability, one-year stability, concurrent validity relative to the Child Behavior Checklist (CBCL)/1.5–5 (Achenbach & Rescorla, 2000) scores and evaluator ratings) (Briggs-Gowan et al., 2004; Briggs-Gowan & Carter, 2006). Results of a recent study from the USA concerning clinical validity of the BITSEA indicated that the BITSEA problem index has fair to excellent sensitivity and specificity relative to diagnosis by “gold standard” interview and clinical-range CBCL scores (Briggs-Gowan et al., 2013). The BITSEA has also demonstrated excellent sensitivity and good specificity in detecting autistic disorder (Briggs-Gowan & Carter, 2006). Finally, in a large birth cohort sample (n = 1004 children) from the USA, the BITSEA screen status at 12 to 36 months of age was found to predict parent- and teacher-reported psychiatric problems and disorders at early elementary school age (Briggs-Gowan & Carter, 2008). Recent studies have supported the reliability and validity of the BITSEA as a screening measure of young children’s SEB problems also in Turkey (Karabekiroglu et al., 2009; Karabekiroglu, Briggs-Gowan, Carter, Rodopman-Arman, & Akbas, 2010) and in the Netherlands (Kruizinga et al., 2012). In a Finnish pilot study concerning 18-month-old children (n = 50), the BITSEA problem index had good internal consistency, and correlations between the BITSEA problem total and CBCL/1.5–5 internalizing, externalizing and total problem scores were evident (Haapsamo et al., 2009). When significant sex differences have been found in the earlier BITSEA studies, girls have been rated higher than boys in competence scores, whereas problem total scores have been higher for boys than for girls (Briggs-Gowan et al., 2004; Briggs-Gowan & Carter, 2006; Karabekiroglu et al., 2009, 2013; Kruizinga et al., 2012, 2013). To our knowledge, only Turkish researchers have evaluated possible differences between the maternal and paternal BITSEA ratings so far, and they found no significant differences between the maternal and paternal scores in any age or sex groups (Karabekiroglu et al., 2009). As described above, there are only few studies focusing on possible sex differences in the occurrence and appearance of SEB problems among children younger than 2 years of age and comparing mothers’ and fathers’ reports about their young children’s SEB problems and competencies. The main purpose of the current study was to compare boys’ and girls’ BITSEA problem and competence scores among 12-month-old infants and 18-month-old toddlers recruited through public child health centers in the city of Oulu, Finland. The second aim was to investigate probable differences in parents’ perceptions of their young children’s SEB problems and skills related to parental and child sex. 2. Methods 2.1. Setting The study was carried out in Oulu which is the capital of northern Finland, in Scandinavia. Between 2006 and 2011 the population of Oulu increased from 129 000 to 142 000 and the birth rate was about 2000 live born children per annum (www.ouka.fi/Statistical Yearbook of the City of Oulu, 2011). The Finnish public child health center organization covers all municipalities in Finland and it is free of charge for families. Almost all children in Oulu attend regular well-child visits at child health centers during their first six years of life, before the school age. Thus, the data collection was carried out in collaboration with nurses at all child health centers of Oulu. 2.2. Procedure The BITSEA data collection was a part of a research project concerning early identification of young children’s mental health problems in Northern Finland launched in 2008. The current paper is the first publication regarding the project results. The research was approved by the Ethical Committee of the University Hospital of Oulu and the Municipal Boards of the Social and Health Care Units of the City of Oulu. The active BITSEA data collection took place from February 2008 to March 2009
J. Alakortes et al. / Infant Behavior & Development 39 (2015) 136–147
139
for 18-month-old toddlers and from February 2010 to February 2011 for 12-month-old infants. After training, health care nurses delivered the BITSEA (a red copy to mothers and a blue copy to fathers) and background information questionnaires to parents during routine well-child visits. Parents mailed completed questionnaires and a written informed consent form to researchers in a pre-paid envelope. Parents could also leave their informed consent form with contact information with the nurse; in these cases, when questionnaires were not returned, researchers telephoned parents and remailed materials to encourage participation. The BITSEA data was used only for research purposes and nurses were not aware of the children’s BITSEA scores during the well-child visits. Participation for both nurses and parents was completely voluntary. 2.3. Participants In 2010–2011 the target population was all 12-month-old children and their parents who participated in the child’s well-child visit in the city of Oulu. Unfortunately, it was not possible to get reliable information about the actual number of parents who were invited in the study by nurses. However, given the facts that almost all children in Oulu attend these well-child visits and the data collection was performed during about one year, the number of 1-year-old population in Oulu at the end of the year 2010 (n = 1964) was used as an estimated number of the target population. Similarly, in 2008–2009 the target population was all 18-month-old children and their parents who participated in the child’s well-child visit in Oulu, and the number of 1-year-old population in Oulu at the end of the year 2008 (n = 1757) was used as an estimation of the target population. The numbers of the 1-year-olds were obtained from the Statistics Finland (www.stat.fi). An initial infant sample consisted of 230 children and toddler sample consisted of 209 children. Thus, both samples represented ∼10% of the estimated target populations. Three subjects were excluded from the initial infant sample due to out of range ages (two adjusted ages were under 11 months and one child was over 18 months). One child who was over 24 months was excluded from the initial toddler sample. Thus, the final infant sample consisted of 227 children, 112 boys (49.3%) and 115 girls (50.7%), with a mean age of 13.0 months. The final toddler sample consisted of 208 children, 94 boys (45.2%) and 114 girls (54.8%), with a mean age of 19.3 months. The maternal BITSEA data was available for all children in both samples. The paternal BITSEA data was available for 205 of 227 (90.3%) infants, 104 of 112 (92.9%) boys and 101 of 115 (87.8%) girls, and for 181 of 208 (87.0%) toddlers, 82 of 94 (87.2%) boys and 99 of 114 (86.8%) girls. Among the final samples, child age was adjusted for 4 infants (2 boys and 2 girls) and 5 toddlers (3 boys and 2 girls) born at 36 weeks or less gestation. None of the children had diagnoses of severe primary diseases or developmental delays. Every child whose Apgar score status was available to the researchers had the last given Apgar score ≥6. More detailed information concerning child age, weight at birth, Apgar scores, as well as parental age and educational level is shown in Table 1. There were no significant differences between the infant sample boys and girls, or between the toddler sample boys and girls, concerning child age, parental age and educational level. Moreover, no significant differences were found between the infant and toddler boys, or between the infant and toddler girls, in relation to weight at birth, parental age and educational level. Based on the background information data, no significant differences were found between the infants with and without the paternal BITSEA data concerning child sex and age, or parental age and educational level. Significant differences were neither found between the toddlers with and without the paternal BITSEA data concerning child sex, age and weight at birth, or parental age and maternal educational level. However, the infant girls with the paternal BITSEA data were somewhat lighter at birth than the infant girls without the paternal BITSEA data (p = 0.001), and the fathers of toddler girls with the paternal BITSEA data were somewhat higher educated than the fathers of toddler girls without the paternal BITSEA data (p = 0.015). Although both samples represented only ∼10% of the target populations (18-month-old inhabitants of Oulu in 2008–2009 and 12-month-old inhabitants of Oulu in 2010–2011), no significant differences were found between the final study samples and the corresponding target populations in relation to the two available child characteristics: (1) children’s mean weight at birth and (2) mothers’ mean age at the child’s birth, except for the infant sample boys who were on average slightly heavier at birth than the target population boys born in 2009 or 2010 (p = 0.032 or 0.027, respectively, but the average effect size was small, d = 0.209). Background characteristics of the target populations were obtained from the Medical Birth Register at The National Institute for Health and Welfare (THL). Additionally, there were no significant differences between the educational levels of the mothers in the study samples and the target females, or between the educational levels of the fathers of the toddler girls and the target males, based on data on the target populations (30–34-year-old citizens of Oulu in 2008–2009 or 2010–2011), by applying the Statistics Finland (www.stat.fi). However, fathers of infant boys and girls, and of toddler boys, were somewhat higher educated than the target males (with p-values of 0.019–0.029, 0.003–0.010, and 0.045–0.048, respectively). The corresponding average effect sizes were at the medium level (w = 0.262, 0.309, and 0.267, respectively). Finally, 2.7% (12/446) of the infant sample parents and 2.1% (8/390) of the toddler sample parents were immigrants, which corresponded percentages of the non-native population in Oulu in the data collection years (www.ouka.fi/ Statistical Yearbook of the City of Oulu 2008–2011). 2.4. Measures 2.4.1. The BITSEA The BITSEA questionnaire (Briggs-Gowan & Carter, 2006) consists of 42 items; 31 items concern SEB problems and 11 items concern deficits in social–emotional competence. The problem items address problematic behavior in externalizing
140
J. Alakortes et al. / Infant Behavior & Development 39 (2015) 136–147
Table 1 Background characteristics of the study samples. Child variables
Infant boys (n = 112) Mean (SD; range)
Infant girls (n = 115) Mean (SD; range)
Infant total (n = 227) Mean (SD; range)
Toddler boys (n = 94) Mean (SD; range)
Toddler girls (n = 114) Mean (SD; range)
Toddler total (n = 208) Mean (SD; range)
Age when BITSEA completed (months)
12.9 (1.1; 11.5–17.6) 3655 (520; 2150–5200) 1 n (%) 106 (94.6)
13.0 (1.2; 11.2–17.9) 3476 (488; 2100–4490) 0 n (%) 111 (96.5)
13.0 (1.1; 11.2–17.9) 3564 (511; 2100–5200) 1 n (%) 217 (95.6)
19.2 (1.3; 16.4–24.0) 3657 (515; 2185–5220) 1 n (%) 85 (90.4)
19.4 (1.4; 16.5–23.7) 3462 (482; 2160–4615) 0 n (%) 103 (90.4)
19.3 (1.4; 16.4–24.0) 3550 (505; 2160–5220) 1 n (%) 188 (90.4)
Parent variables
Infant boys (n = 112) Mean (SD; range)
Infant girls (n = 115) Mean (SD; range)
Infant total (n = 227) Mean (SD; range)
Toddler boys (n = 94) Mean (SD; range)
Toddler girls (n = 114) Mean (SD; range)
Toddler total (n = 208) Mean (SD; range)
Mother’s age at the child’s birth (years)
29.4 (5.7; 17–41) 0 31.4 (6.5; 19–47) 0 n (%) 4 (3.6) 45 (40.2) 63 (56.3) 0 (0) n (%) 4 (3.6) 56 (50.0) 50 (44.6) 2 (1.8)
29.0 (4.6; 17–40) 1 30.9 (5.6; 17–48) 1 n (%) 5 (4.3) 33 (28.7) 77 (67.0) 0 (0) n (%) 4 (3.5) 42 (36.5) 65 (56.5) 4 (3.5)
29.2 (5.2; 17–41) 1 31.2 (6.0; 17–48) 1 n (%) 9 (4.0) 78 (34.4) 140 (61.7) 0 (0) n (%) 8 (3.5) 98 (43.2) 115 (50.7) 6 (2.6)
29.0 (5.1; 20–42) 1 30.9 (6.3; 19–49) 4 n (%) 5 (5.3) 31 (33.0) 47 (50.0) 11 (11.7) n (%) 2 (2.1) 42 (44.7) 43 (45.7) 7 (7.4)
29.8 (5.5; 17–43) 1 31.4 (6.2; 21–47) 2 n (%) 7 (6.1) 31 (27.2) 68 (59.6) 8 (7.0) n (%) 5 (4.4) 43 (37.7) 56 (49.1) 10 (8.8)
29.4 (5.3; 17–43) 2 31.1 (6.2; 19–49) 6 n (%) 12 (5.8) 62 (29.8) 115 (55.3) 19 (9.1) n (%) 7 (3.4) 85 (40.9) 99 (47.6) 17 (8.2)
Weight at birth (g) - Missing data (n) Last given Apgar score (≥6)a
- Missing data (n) Father’s age at the child’s birth (years) - Missing data (n) Mother’s educational level (a) Elementary school (b) Vocational/senior high school (c) College/university (d) Undefined Father’s educational level (a) Elementary school (b) Vocational/senior high school (c) College/university (d) Undefined
Note: SD = standard deviation. a Every child whose Apgar score status was available to the researchers had the last given Apgar score ≥6.
(6 items; e.g., impulsivity, defiance, peer aggression), internalizing (8 items; e.g., fearfulness, worry, anxiety, sadness) and dysregulation (8 items; e.g., sleep and eating problems, negative emotionality, sensory sensitivities) domains, as well as rare behaviors that may be early markers of autism spectrum disorders (e.g., repeating of same action or phrase over and over without enjoyment) or other significant psychopathology (14 red flag items; e.g., hurts self on purpose). Some of the items address problematic behavior overlapping in two or three domains (e.g., avoids physical contact). Besides 9 problem items, the autism domain consists of 8 competence items (deficits in those). The competence items rate the child’s skills of attention, compliance, mastery motivation, prosocial peer relations, empathy, imitation/play, and social relatedness. Response alternatives for the BITSEA items are Not true/rarely (=0), Somewhat true/sometimes (=1), and Very true/often (=2), providing problem total score (range 0–62) and competence total score (range 0–22). High problem total score and/or low competence total score indicates that the assessed child may have SEB problems and/or deficits in competence, and follow-up or more comprehensive evaluation is needed to determine whether or not difficulties are clinically significant. The BITSEA cut score values are based on a national USA standardization sample for 12 to 36-month-olds (by sex and age in 6-month age bands) and set at the highest 25th percentile for problem total scores and at the lowest 15th percentile for competence total scores. In addition, the BITSEA form includes two questions addressing the parental level of worry about the child’s (1) psychosocial and (2) language development but not counting toward the BITSEA scores. The current study utilized the Finnish translation of the BITSEA developed for a previous Finnish BITSEA pilot study (Haapsamo et al., 2009). Because there are no standardized Finnish cut score values for the BITSEA, the USA standardized cut scores (Briggs-Gowan & Carter, 2006) were applied; for 12–17-month-olds competence total score ≤12 (for both sexes) and problem total score ≥13 (for both sexes), and for 18–23-month-olds competence total score ≤14 (for both sexes) and problem total score ≥15 (for boys) or ≥13 (for girls). If the child’s competence total score was at or below the competence cut score, it was in the possible deficit range. Correspondingly, if the child’s problem total score was at or above the problem cut score, it was in the possible problem range. All the infants, and the toddler sample subjects who were slightly under 18 months of age (11 boys and 11 girls rated by mothers, 11 boys and 9 girls rated by fathers), were screened by using the cut score values for 12–17-month-old children. The toddlers who were at minimum 18-month-olds were screened by using the cut score values for 18–23-month-old children. Consistent with scoring guidelines, child age was adjusted for children born at 36 weeks or less gestation. If ≥2 competence and/or ≥5 problem items per a rater were unanswered, the corresponding total score/-s could not be calculated and used in the analyses. Only the infants and toddlers whose both competence and problem total
J. Alakortes et al. / Infant Behavior & Development 39 (2015) 136–147
141
Table 2 Maternal and paternal mean scores and standard deviations of the BITSEA scales and domains for the infant and toddler sample boys and girls. Infant sample
Maternal
Paternal
(n mat/n pat/n both)
Mean (SD)
Mean (SD)
Total (224/198/197) Boys (110/100/99) Girls (114/98/98) pa /Effect size (d)c Total (223/200/198) Boys (109/103/101) Girls (114/97/97) pa /Effect size (d)c Total (224/202/201) Boys (111/103/103) Girls (113/99/98) pa /Effect size (d)c Total (223/203/200) Boys (109/104/102) Girls (114/99/98) pa /Effect size (d)c Total (225/202/202) Boys (111/103/103) Girls (114/99/99) pa /Effect size (d)c
Competence 15.30 (2.79) 14.95 (2.75) 15.64 (2.79) 0.062/– Problem total 8.15 (4.13) 8.63 (4.27) 7.69 (3.96) 0.046* /0.228 Externalizing 1.91 (1.58) 2.26 (1.65) 1.57 (1.43) 0.001* /0.447 Internalizing 1.17 (1.14) 1.17 (1.08) 1.18 (1.20) 0.767/– Dysregulation 3.10 (2.05) 3.04 (1.90) 3.16 (2.20) 0.657/–
pb /Effect size (d)c
Toddler sample
Maternal
Paternal
(n mat/n pat/n both)
Mean (SD)
Mean (SD)
15.07 (3.05) 14.58 (3.10) 15.57 (2.94) 0.022* /0.328
0.525/– 0.383/– 0.975/–
Total (207/181/180) Boys (93/82/81) Girls (114/99/99) pa /Effect size (d)c
7.67 (4.51) 7.95 (4.56) 7.36 (4.46) 0.313/–
0.277/– 0.252/– 0.701/–
Total (207/181/180) Boys (93/82/81) Girls (114/99/99) pa /Effect size (d)c
1.81 (1.56) 2.08 (1.67) 1.53 (1.39) 0.011* /0.358
0.737/– 0.388/– 0.595/–
Total (207/181/180) Boys (93/82/81) Girls (114/99/99) pa /Effect size (d)c
0.93 (1.07) 0.93 (1.18) 0.93 (0.96) 0.653/–
0.024* /0.157 0.173/– 0.071/–
Total (208/181/181) Boys (94/82/82) Girls (114/99/99) pa /Effect size (d)c
2.85 (2.10) 2.73 (1.97) 2.97 (2.24) 0.416/–
0.140/– 0.125/– 0.506/–
Total (207/181/180) Boys (93/82/81) Girls (114/99/99) pa /Effect size (d)c
Competence 18.02 (2.36) 17.28 (2.52) 18.63 (2.04)
