ACCIÓN PSICOLÓGICA, junio 2019, vol. 16, nº. 1, 1-12. ISSN: 2255-1271 https://doi.org/10.5944/ap.16.1.22098 1 RISK FACTORS FOR OVERWEIGHT IN EARLY AGES: LONGITUDINAL DESIGN WITH SPANISH PRESCHOOLERS FACTORES DE RIESGO PARA EL SOBREPESO EN EDADES TEMPRANAS: DISEÑO LONGITUDINAL CON PREESCOLARES ESPAÑOLES IRIS PÉREZ-BONAVENTURA1,2, ROSER GRANERO1,3, AND LOURDES EZPELETA1,2 Cómo referenciar este artículo/How to reference this article: Pérez-Bonaventura, I., Granero, R., & Ezpeleta, L. (2019). Risk Factors for Overweight in Early Ages: Longitudinal Design with Spanish Preschoolers [ F a c t o r e s d e r i e s g o p a r a e l s o b r e p e s o e n e d a d e s t e m p r a n a s : d i s e ñ o l o n g i t u d i n a l c o n p r e e s c o l a r e s e s p a ñ o l e s ] . Acción Psicológica, 16(1), 1- 12. https://doi.org/10.5944/ap.16.1.22098 focused on school-aged children. The aim of this study is Abstract to identify risk factors in early life for overweight in a community sample of Spanish preschoolers. A sample of Previous studies have shown that there are several 622 three-year-olds was monitored until the age of 5, and significant early-life risk factors associated with their height and weight were registered annually. childhood overweight. However, research has mainly Overweight status was defined by World Health Organization standards. A large set of risk factors Correspondence address [Dirección para correspondencia]: Lourdes Ezpeleta. Departament de Psicologia Clínica i de la Salut, Universitat Autònoma de Barcelona, Bellaterra, Barcelona, Spain. Email: Lourdes.Ezpeleta@uab.cat ORCID: Iris Pérez-Bonaventura (http://orcid.org/0000-0001-7100-6205), Roser Granero (http://orcid.org/0000-0001- 6308-3198) y Lourdes Ezpeleta (http://orcid.org/0000-0002-8957-083X). 1 Unitat d’Epidemiologia i de Diagnòstic en Psicopatologia del Desenvolupament, Universitat Autònoma de Barcelona, Bellaterra, Barcelona, Spain. 2 Departament de Psicologia Clínica i de la Salut, Universitat Autònoma de Barcelona, Bellaterra, Barcelona, Spain. 3 Departament de Psicobiologia i Metodologia de las Ciències de la Salut, Universitat Autònoma de Barcelona, Bellaterra, Barcelona, Spain. Retrieved: June 4, 2018. Accepted: September 5, 2018. Retrieved: 22 mayo de 2018. Aceptado: 24 de julio de 2018. 2 ACCIÓN PSICOLÓGICA, junio 2019, vol. 16, nº. 1, 1-12. ISSN: 2255-1271 https://doi.org/10.5944/ap.16.1.22098 including sociodemographic variables, family structure, para cada grupo de edad. El presente estudio es el primer pregnancy, birth, postnatal period, school and estudio longitudinal en examinar diferentes factores de neighborhood were measured through semi-structured riesgo para el sobrepeso en niños españoles con edad interviews. Stepwise logistic regressions created preescolar. Diferentes factores de riesgo están asociados a predictive models with the best predictors of overweight un posterior sobrepeso: mayor peso al nacer, etnia, uso ex- for each age group. This is the first longitudinal study to cesivo de pantallas, capacidad de atención baja, madre fu- examine a large set of risk factors among Spanish madora durante el embarazo, disciplina inconsistente, cas- preschoolers. Several potential early risk factors in tigo físico y psicopatología parental. De entre ellos, el fac- preschool children were associated with later overweight: tor más importante fue: mayor peso al nacer (razón de pro- high birth weight, ethnicity, excess screen time, low babilidades = 1.89 a los 3 años, 1.87 a los 4 años y 2.35 a attention span, maternal smoking during pregnancy, los 5 años), subrayando la importancia de monitorizar el inconsistent discipline, fewer rules, corporal punishment peso del niño desde el post-parto y realizar cribados ruti- and parents psychopathology. Among them, the most narios de sobrepeso en aquellos niños que nacieron con un powerful factor was high birth-weight (odds ratio = 1.89 peso elevado. La prevalencia cada vez mayor de sobrepeso at age 3, 1.87 at age 4, and 2.35 at age 5), underlining the en niños cada vez más pequeños hace que la comprensión importance of weight monitoring from postpartum and e identificación de los factores de riesgo relacionados con routine screening for overweight in children who have el sobrepeso sea crucial para que los profesionales públi- high birth weight. With the increasing prevalence of cos de salud y los responsables políticos implementen pro- overweight in children at early ages, understanding the gramas de prevención e intervención efectivos. determinants for overweight risk becomes crucial for public health professionals and policy-makers in order to Palabras clave: Factor de riesgo; IMC; Preescolares; implement effective prevention and intervention Sobrepeso. programs. Keywords: BMI; Overweight; Preschool children; Risk factor; Spain. Introduction Resumen Childhood overweight continues to be a major chal- lenge worldwide, associated with serious health conse- Diferentes estudios han mostrado una asociación clara en- quences and psycho-social outcomes (Lakshman, Elks, & tre determinados factores de riesgo en edad temprana y so- Ong, 2012; Reilly et al., 2003). Alarmingly, one out of brepeso en la infancia. Sin embargo, hasta ahora la inves- four Spanish children is already overweight or obese by tigación solo se ha centrado en niños en edad escolar. El age five (Miqueleiz et al., 2014). This prevalence is of objetivo de este estudio es el de examinar en niños prees- considerable concern owing to the adverse consequences colares los factores de riesgo para el sobrepeso en una of childhood overweight and the fact that weight status at muestra comunitaria de niños españoles. En el presente es- 5 years of age is a good indicator of the future health of a tudio se seleccionó una muestra de 622 niños de 3 años de child (Gardner et al., 2009). Furthermore, overweight edad a los que se siguió anualmente hasta los 5 años. El during childhood tends to persist into adulthood peso y la talla fueron registrados anualmente y el sobre- (Freedman Mei, Srinivasan, Berenson, & Dietz, 2004) peso fue definido siguiendo los criterios definidos por la with resultant long-term consequences that are difficult to Organización Mundial de la Salud. A través de una entre- treat successfully in the long term. Therefore, prevention vista semi-estructurada, se analizaron diversos factores de of the onset of overweight is a public health priority and riesgo tales como estructura familiar, embarazo, parto, pe- should be a primary target (Cominato et al., 2018; Gordon- ríodo post-natal, escuela y vecindario. Las regresiones lo- Larsen, Adair, Nelson, & Popkin, 2004). Indeed, the key gísticas múltiples permitieron crear modelos predictivos to prevention is to examine and fully characterize the que seleccionaron los mejores predictores de sobrepeso factors involved in the development of overweight. ACCIÓN PSICOLÓGICA, junio 2019, vol. 16, nº. 1, 1-12. ISSN: 2255-1271 https://doi.org/10.5944/ap.16.1.22098 3 Overweight is a multi-factorial condition influenced by who declined to take part (p = .951). However, the propor- diverse factors operating across the lifespan: genetic, neu- tion of refusals was statistically higher in low socioeco- roendocrine, metabolic, psychological, environmental and nomic status families (SES; Hollingshead index; Hollings- socio-cultural. To develop evidence-based policy and pro- head, 1975) groups (p < .001). Screening of children with gram decision-making, it is crucial to identify the risk fac- possible behavioral problems was carried out by adminis- tors associated with pediatric overweight. In a recent sys- tering, to parents, four items from the Strengths and Diffi- tematic review of thirty prospective studies from devel- culties Questionnaire (SDQ; Ezpeleta et al., 2014) conduct oped countries, several significant early-life risk factors problems scale plus four questions derived from the were identified for childhood overweight: maternal pre- Diagnostic and Statistical Manual of Mental Disorders pregnancy overweight, high infant birth-weight, early in- diagnostic criteria for Oppositional Defiant Disorder. fant rapid weight gain and maternal smoking during preg- Based on the results of the screening process, two groups nancy (Weng, Redsell, Swift, Yang, & Glazebrook, 2012). were formed. The positive screen group (n = 205, 51.2 % However, research has mainly focused on school-aged boys) included all children who reached the threshold for children and adolescents, although the preschool years are behavioral problems while the negative screen group a critical period for obesity prevention (Baker, Olsen, & (n = 417, 49.4 % boys) included a random sample from Sørensen, 2007). Given the high prevalence of overweight the 30 % of children who did not reach the threshold (the across Southern Europe (van Stralen et al., 2012), research percentage of negative screen participants guaranteeing is particularly needed on preschoolers in these countries, adequate statistical power for the analysis planned in the including Spain. research). In the second phase, n = 622 children and their parents agreed to take part although 11 children were This study explores the main risk factors for the devel- excluded due to missing height or weight measurement opment of childhood overweight in a large community data. Consequently, the sample for this study was n = 611 sample of Spanish preschoolers longitudinally assessed three-year-old children and their parents. Children were from ages 3 to 5 years-old. The set of potential risk factors followed-up at age 4 (N = 596) and 5 (N = 564). Drop-outs includes variables of the main areas associated with over- were statistically equal by gender (p = .188) and SES weight during childhood and adolescence, including fam- (p = .062). Excluded were children with intellectual ily structure, pregnancy, birth, postnatal period, medical disability or pervasive developmental disorder, no parent history, school and neighborhood. to provide reports on the child, parental lack of fluency in Spanish or Catalan, or family relocating outside study area within 12 months. Method Measures Participants Sociodemographic and health charac- The data were derived from a large-scale longitudinal teristics project on psychopathology and risk factors specifically focused on behavioral problems in preschool children. A The Schedule of Risk Factors (SRF; Unitat d'Epidemi- full description of the inclusion criteria, recruitment, and ologia i de Diagnòstic en Psicopatologia del Desenvo- retention strategies is provided elsewhere (Ezpeleta, de la lupament, 2009) is a semistructured interview answered Osa, & Doménech, 2014). In the first phase of two-phase by parents and conceived of as a compendium of the main sampling, a community sample of 2,283 children in the areas of psychopathology risk in children. It provides in- first year of preschool education was randomly selected formation on the child (e.g., gender and ethnicity), the from the census of preschool center facilities in Barcelona, family (e.g., family structure and maternal smoking during Spain, in 2009. The families were contacted and 1,341 pregnancy) and the community (e.g., distance to a traffic (58.7 %) agreed to participate. There were no significant dense route and proximity to school) that could have an differences by gender among the participants and those 4 ACCIÓN PSICOLÓGICA, junio 2019, vol. 16, nº. 1, 1-12. ISSN: 2255-1271 https://doi.org/10.5944/ap.16.1.22098 impact on the children’s psychological status. Inter-rater (DICA-PPYC; Ezpeleta, de la Osa, Granero, Doménech, reliability and concurrent validity are acceptable in Span- & Reich, 2011) is a semi-structured interview designed to ish populations. assess the most common psychological disorders at ages 3–7 years according to DSM-IV-TR criteria. Parents were The Children’s Behaviour Questionnaire (CBQ; Roth- interviewed about the presence of feeding problem symp- bart, Ahadi, Hershey, & Fisher, 2001) is designed to assess toms. temperament in 3 to 7-year-old children. Completed by parents, it contains 94 items over 15 scales clustered in 3 Anthropometric variables broad dimensions of temperament: negative affectivity (anger-frustration, discomfort, fear, sadness, falling reac- Children were weighed and their height was measured tivity-soothability), effortful control (attention focusing, while barefoot and wearing light clothes. Measurements inhibitory control, low-intensity pleasure, perceptual sen- were taken and noted in the child’s medical record by sitivity), and surgency-extraversion (activity level, high- nurses at outpatient pediatric primary care units during intensity pleasure, impulsivity, shyness). The study sam- each child’s annual visit. These values were recorded in ple alpha values ranged between α = .50 for sadness and the child’s medical records and were recovered annually α = .79 for shyness. during the course of this study. BMI (Body Mass Index) was calculated as the ratio between weight (kg) and height The Adult Self Report (ASR; Achenbach & Rescorla, (m2). Children were classified as non-overweight or over- 2003) is a self-report questionnaire that assesses behav- weight (obesity included) according to the World Health ioral problems in adults aged 18 to 59. The emotional and Organization (WHO) international age- and sex-specific behavioral problems section includes 8 empirical syn- child growth standards specifically for preschool-aged dromes that can be classified as internalizing problems, children (World Health Organization, 2006). Based on the externalizing problems and total problems. Internal con- WHO recommendations specifically for preschool-aged sistency for mothers’ reports in this sample ranged be- children, overweight was defined as having a BMI greater tween .83 for externalizing problems and .92 for the total than two standard deviations from the mean weight in the score while, for fathers, it ranged between .86 for exter- WHO reference population (de Onis, Blössner, & Borghi, nalizing problems and .93 for the total score. 2010). However, categorizing children based on BMI cri- teria can produce different classifications with varying The Alabama Parenting Questionnaire-Pr (APQ-Pr; clinical and public health implications (Jansen, Mensah, Osa, Granero, Penelo, Domènech, & Ezpeleta, 2014) as- Nicholson, & Wake, 2013). Therefore, the BMI z-scores sesses the behavior of parents of children with externaliz- were also calculated from WHO reference curves (World ing disorders. It contains 42 items, clustered in 5 sub- Health Organization, 2006). As such, throughout the scales: parental involvement (α = .56), positive parenting study, two measures were analyzed: (a) a binary variable (α = .66), poor monitoring/supervision (α = .20), incon- (non-overweight status vs overweight) extensively used in sistent discipline (α = .43) and corporal punishment clinical settings; and (b) a quantitative variable (BMI z- (α = .42). In the Spanish version of the APQ-Pr, two scores) mainly used for research purposes. subscales were added: norms (α = .88) and autonomy (α = .81), this version shows moderate to very good Procedure internal consistency in our study sample. This longitudinal project was approved by the ethics The Four-Factor Hollingshead Index of Social Status review committee at the authors’ institution. Heads of the (Hollingshead, 1975) based on parents’ educational level participating schools and parents were provided with a full and occupation was used to code the SES into five levels. description of the study. Families were recruited at the schools, and parental consent was obtained. Parents of Finally, the Diagnostic Interview of Children and Ad- children in Grade P3 (three-year-olds) at the participating olescents for Parents of Preschool and Young Children schools were invited to answer the SDQ at home and re- ACCIÓN PSICOLÓGICA, junio 2019, vol. 16, nº. 1, 1-12. ISSN: 2255-1271 https://doi.org/10.5944/ap.16.1.22098 5 turn it to the schools. Families who agreed and met the We longitudinally assessed associations between early screening criteria were contacted by telephone and inter- risk factors (independent variables) before age 3 and viewed at the school. Interviewers were all experienced weight status (presence of overweight and z-BMI scores; graduate students who received training on the develop- dependent variables) at 3, 4, and 5 years-old. This was mental psychology and psychopathology of preschool achieved by using binary logistic regressions through children and on the use of the DICA-PPYC and the Sched- stepwise procedures to select the best discriminative ule of Risk Factors. To ensure the reliability and validity models of the children’s weight group (non-overweight of the tests, all the interviews were audio-recorded and su- vs. overweight) and stepwise multiple linear regressions pervised. Researchers conducted a face-to-face interview to select the best predictors of the child’s z-BMI score. with the children’s parents before the parents answered the Due the large set of risk factors considered, seven questionnaires. Participants and their families followed independent models were built for groups of variables: the same procedures at each annual assessment. Data were child’s characteristics, temperament CBQ-scales, family collected annually between November, 2009 and July, characteristics, mother psychopathology (ASR), father 2012. psychopathology (ASR), parenting style (APQ-Pr) and community. Additionally, due to the strong association Statistical analysis between children’s weight status and SES (Stamatakis, Wardle, & Cole, 2010), all the analyses were controlled by The statistical analysis was carried out with SPSS20 family SES. The analyses were also controlled for child’s for Windows. Due to the multi-sampling design (the data gender. This was carried out entering in a first block of the analyzed in this work were selected through a sampling multiple and logistic regressions the covariates family procedure including a screening tool), the Complex Sam- SES and child’s gender. The predictive capacity of the risk ples module (CS) was used, assigning to each child a factors selected in the final models was measured through weight inversely proportional to the probability of the par- the change in the R2 coefficient (linear regressions) and ticipant’s selection in the second phase of the design (after the change in Nagelkerke’s-R2 (logistic regressions) the screening). (DR2), comparing the first block with the covariates and the second block with the selected risk factors. The association of overweight with demographic vari- ables was analyzed using Chi square tests for categorical measures and t-test for quantitative measures. Table 1. Descriptives for the sample at baseline (age 3 years-old). Non-overweight Overweight (N= 558) (N= 53) p Gender (male); n (%) 278.0 (49.8%) 27.0 (50.9%) .876 Ethnicity; n (%) Caucasian 505.0 (90.5%) 40.0 (75.5%) .006 American-Hispanic 37.0 (6.6%) 10.0 (18.9%) Other 16.0 (2.9%) 3.0 (5.7%) SES; n (%) High 194.0 (34.8%) 9.0 (17.0%) .005 Mean-High 178.0 (31.9%) 14.0 (26.4%) Mean 77.0 (13.8%) 9.0 (17.0%) Mean-Low 81.0 (14.5%) 15.0 (28.3%) Low 28.0 (5.0%) 6.0 (11.3%) BMI; mean (standard deviation) 15.5 (1.59) 20.7 (2.79) < .001 Note. p-value obtained through c2 for categorical measures and T-TEST for quantitative measures. BMI = body mass index (kg/m2), SES = Socioeconomic status. 6 ACCIÓN PSICOLÓGICA, junio 2019, vol. 16, nº. 1, 1-12. ISSN: 2255-1271 https://doi.org/10.5944/ap.16.1.22098 Results p = .619 and p = .143 at ages 3-4-5). Statistical differences related to overweight emerged for ethnicity and SES; Sample characteristics prevalence of overweight was higher for Hispanics (Latin American origin), other ethnic minority groups and Table 1 includes the characteristics of the sample at families with lower SES. baseline (age 3 years-old) stratified by children’s weight group. Overall, 7.2 % of children at age 3 were overweight Risk Factors for child’s overweight (7.2 % boys and 7.2 % girls, p = 1.00), 8.2 % at age 4 (7.9 % boys and 8.5 % girls, p = .814) and 8.7 % at age 5 Table 2 presents the results for the second block/step (6.7 % boys and 10.8 % girls, p = .083). No statistical as- of the final stepwise logistic regressions selecting the main sociation emerged for gender and z-BMI (p = .754, risk factors for child’s overweight at ages 3, 4, and 5. Table 2. Final predictive models for children’s group of weight (0=non-overweight vs. 1=overweight). Significant predictors B SE(B) Sig OR 95%CI for OR DR2 Overweight at age 3 Child characteristics Birth-weight (kg) 0.639** 0.324 .049 1.895 1.004 3.576 0.04 Number delays in development 0.464** 0.165 .005 1.591 1.150 2.201 CBQ-Temperament Attention focusing -0.347** 0.141 .014 0.707 0.536 0.932 0.02 Smoking during pregnancy 0.770** 0.367 .036 2.161 1.051 4.440 ASR-Mother None ASR-Father None APQ Parenting style Inconsistence 0.135** 0.062 .029 1.144 1.014 1.291 0.02 Community None Overweight at age 4 Child characteristics Ethnic (non- Caucasian) 0.856** 0.430 0.046 2.354 1.014 5.464 0.10 Television viewing (minutes) 0.009** 0.003 0.004 1.009 1.003 1.016 Presence of feeding symptoms -1.129** 0.518 0.029 0.323 0.117 0.893 Birth-weight (kg) 0.624** 0.319 0.050 1.866 1.000 3.487 Number delays in development 0.361** 0.170 0.034 1.435 1.027 2.004 CBQ None ASR-Mother None ASR-Father Total score 0.023** 0.008 0.003 1.023 1.008 1.038 0.04 APQ Parenting style Norms -0.149** 0.047 0.002 0.862 0.786 0.945 0.04 Community None Overweight at age 5 Child characteristics Ethnic (non-Caucasian) 0.933** 0.453 .040 2.543 1.046 6.186 0.09 Birth-weight (kg) 0.853** 0.327 .009 2.347 1.237 4.454 Frequent doctor appointments 0.859** 0.424 .043 2.360 1.027 5.421 Daily medication 0.730** 0.376 .050 2.076 1.000 4.338 CBQ None ASR-Mother None ASR-Father Total score 0.020** 0.007 .007 1.020 1.005 1.035 0.03 APQ Parenting style None Community None Note. Models obtained through binary logistic regressions (stepwise procedure) adjusted for socioeconomic status and child’s sex. APQ = Alabama Parenting Questionnaire, ASR = Adult Self Report, B = Regression coefficient, CBQ = Children’s Behaviour Questionnaire, SE = standard error, SES = Socioeconomic status, Sig = Statistical significance, OR = Odds Ratio, DR2 = Change in R2. *p < .05, ** p < .01, ** p < .001. ACCIÓN PSICOLÓGICA, junio 2019, vol. 16, nº. 1, 1-12. ISSN: 2255-1271 https://doi.org/10.5944/ap.16.1.22098 7 These results are adjusted for family’s SES and child’s sex The predictive capacity of the final logistic regressions covariates. The presence of overweight at age 3 was selecting the best overweight risk factors was low to related to lower CBQ-attention focusing score, higher moderate. The lowest DR2 corresponded to the models; birth weight, more developmental delays, smoking during parenting style at age 3 (1.9%), CBQ at age 3 (2.4%) and pregnancy and higher scores in the APQ-inconsistent ASR-father at age 5 (2.9%). The highest DR2 was for the discipline scale. child characteristics models at age 4 (10.1%) and 5 (8.5%). Being over-weight at age 4 was predicted by non- Risk Factors for child’s z-BMI Caucasian ethnicity, higher birth weight, more hours watching TV, the absence of feeding symptoms, more Table 3 shows the results for the second block/step of developmental delays and lower scores in the parenting the final stepwise multiple regressions selecting the main style norms scale before age 3. variables related to the z-BMI score, adjusted for the fam- ily’s SES and child’s sex covariates. The set of predictors Overweight at age 5 was predicted by non-Caucasian selected was similar to that obtained with the logistic re- ethnicity, higher birth weight, more visits to the doctor, gressions for the dependent variable child’s overweight. daily medication and higher ASR-father-total score before However, for the dependent variable z-BMI score, at age age 3. 3 the variable “living near to a dense street” was selected, at age 4 the variable “teasing for appearance” was a sig- Table 3. Final predictive models for children’s z-BMI. Z-BMI at age 3 Significant predictors B SE (B) Beta Sig 95%CI for B DR2 Child characteristics Ethnic (non-Caucasian) 0.763 0.248 0.128 .002 0.275 1.250 .04 Presence of feeding symptoms -0.334 0.161 -0.083 .038 -0.650 -0.019 Birth-weight (kg) 0.395 0.115 0.140 .001 0.170 0.621 CBQ Attention focusing -0.139 0.061 -0.092 .024 -0.260 -0.019 .01 Family characteristics SES (1=high to 5=low) 0.175 0.056 0.127 .002 0.066 0.285 .02 ASR-mother None ASR-father Externalizing 0.024 0.012 0.086 .048 0.000 0.048 .01 APQ-Parenting style Corporal punishment 0.144 0.063 0.093 .023 0.020 0.269 .01 Community Distance dense route (< 300 m) 0.289 0.141 0.086 .041 0.012 0.566 .01 Z-BMI at age 4 Child characteristics Ethnic (non-Spanish) 0.767 0.230 0.139 .001 0.315 1.220 .08 Birth-weight (kg) 0.415 0.124 0.163 .001 0.172 0.658 Prematurity 0.651 0.260 0.120 .012 0.141 1.161 Breastfeeding 0.316 0.147 0.087 .032 0.027 0.606 Television viewing (minutes) 0.003 0.001 0.108 .013 0.001 0.006 Teased for appearance 1.527 0.637 0.096 .017 0.276 2.777 Presence of feeding symptoms -0.289 0.146 -0.080 .048 -0.576 -0.002 CBQ None Family characteristics SES (1=high to 5=low) 0.209 0.051 0.166 <.001 0.109 0.310 .03 ASR-Mother None ASR-Father Total score 0.008 0.003 0.104 .018 0.001 0.014 .01 Parenting style Norms -0.047 0.023 -0.082 .044 -0.093 -0.001 .01 Community None Note. Models obtained through binary logistic regressions (stepwise procedure) adjusted for socioeconomic status and child’s sex. APQ = Alabama Parenting Questionnaire, ASR = Adult Self Report, B = Regression coefficient , BMI = body mass index, CBQ = Children’s Behaviour Questionnaire, SE = standard error, SES = Socioeconomic status, Sig = Statistical significance, OR = Odds Ratio, DR2 = Change in R2. *p < .05, ** p < .01, ** p < .001. 8 ACCIÓN PSICOLÓGICA, junio 2019, vol. 16, nº. 1, 1-12. ISSN: 2255-1271 https://doi.org/10.5944/ap.16.1.22098 Table 3 (continued). Final predictive models for children’s z-BMI. Z-BMI at age 5 Significant predictors B SE (B) Beta Sig 95%CI for B DR2 Child characteristics Ethnic (non-Spanish) 0.666*** 0.234 0.120 .005 0.206 1.126 .06 Birth-weight (kg) 0.366*** 0.104 0.147 <.001 0.162 0.570 Clumsy 0.545*** 0.183 0.123 .003 0.186 0.904 Frequent doctor appointments 0.397*** 0.202 0.082 .049 0.001 0.794 CBQ None ASR-Mother None ASR-Father Total 0.010*** 0.004 0.131 .003 0.004 0.017 .02 Parenting style None Community None Note. Models obtained through binary logistic regressions (stepwise procedure) adjusted for socioeconomic status and child’s sex. APQ = Alabama Parenting Questionnaire, ASR = Adult Self Report, B = Regression coefficient , BMI = body mass index, CBQ = Children’s Behaviour Questionnaire, SE = standard error, SES = Socioeconomic status, Sig = Statistical significance, OR = Odds Ratio, DR2 = Change in R2. *p < .05, ** p < .01, ** p < .001. nificant predictor, and at ages 4 and 5 “being clumsy” was overweight in children who had high birth weight as a an additional predictor. means of preventing adverse health outcomes. In this study, American-Hispanic children had tripled the prevalence of overweight found in Caucasian children Discussion and being an ethnic minority was associated with over- weight at age 4 and 5. Ethnic minorities have been shown To the best of our knowledge, this is the first commu- to be more prone to overweight in many European coun- nity-based study to examine a variety of risk factors for tries (Veldhuis et al., 2013). Members of minority ethnic overweight in a sample of Spanish preschoolers. groups often have lower socioeconomic status, which is in turn associated with a greater risk of obesity in children. Socioeconomic status was independently associated Inequities in a range of factors (such us income, stable and with overweight, and this pattern remained for both gen- affordable housing and access to quality education), dis- ders over the 3-year assessment. As in other studies (Sta- parate access to affordable, healthy food or safe places to matakis et al., 2010), we observed a gradient whereby be physically active and the fact that less nutritious, calo- children of socioeconomically disadvantaged families rie-dense foods are often less expensive than healthier were more likely to be overweight. That this is seen in foods contribute to higher rates of obesity in ethnic minor- preschool years suggests that differences in adiposity by ities. Addressing these disparities requires making health- socioeconomic status may emerge early in life. ier choices easier in children’s and families’ daily lives by removing obstacles that make healthy, affordable food The stepwise regression analysis carried out revealed less accessible, and at the same time ensuring communi- several risk factors for overweight. Among them, infant ties have more safe and accessible places for kids to be birth weight was the most relevant predictor for over- physically active. However, it is important to mention that weight. Therefore, the risk of overweight escalates with in this study all the analyses were controlled for socioeco- increasing birth weight in both genders. Several studies nomic status, highlighting the fact that the reasons for the have demonstrated an association between high birth differences in prevalence of childhood obesity among dif- weight and overweight in children (Mitchell et al., 2016; ferent ethnic groups are complex, likely involving genet- Weng et al., 2012;), underlining the importance of weight ics, physiology, culture and environment. monitoring from postpartum and routine screening for Lower scores in the CBQ attention focusing scale were associated with overweight at age 3 supporting studies ACCIÓN PSICOLÓGICA, junio 2019, vol. 16, nº. 1, 1-12. ISSN: 2255-1271 https://doi.org/10.5944/ap.16.1.22098 9 linking attention deficit with excess weight in childhood. Philips, Orzol, and Burdette (2007) reported that the odds Faith and Hittner (2010) found that among boys, poorer of obesity for 3-year-old children were 50 % higher for attention span at the age of 1 year-old predicted greater those who experienced neglect in the previous year. In weight gain at 6 years of age. Results obtained in a previ- another study, those adults who reported experiencing ous study using the same sample showed that ADHD harsh physical punishment during childhood had a higher symptoms were associated with overweight as early as the likelihood of obesity (Afifi, Mota, MacMillan, & Sareen, preschool years (Perez-Bonaventura, Granero, & 2013). Moreover, lack of effective rules in families with Ezpeleta, 2015). It might be that reduced attention span obese children often interferes with the development of interferes in the development of self-regulatory eating and regulated eating patterns. satiety recognition, leaving the child relatively inattentive to hunger and satiety cues. Interestingly, in our study, the Among fathers, some psychological symptoms scales, scale attention focusing was only associated negatively such as externalizing symptoms and total number of symp- with overweight at age 3, but not at age 4 and 5, suggesting toms, were associated with the child’s weight. Adults with that attention focusing is only one of the clinical features binge eating disorders describe their childhood environ- of ADHD and that boarder tests are needed to provide ment and the parenting they received more negatively than more insight into the relationship between ADHD and healthy controls (Hodges, Cochrane, & Brewerton, 1998). overweight. During childhood, they might have used eating as a strat- egy to cope with negative feelings. The effect of father’s Maternal smoking during pregnancy was associated psychopathology on the child’s weight may also be medi- with overweight at age 3. The relationship between mater- ated through parenting. Knowledge of the mechanisms in- nal smoking and childhood obesity has been described as volved may help us to modify dysfunctional associations catch-up growth following low birth weight caused by and improve the child’s adjustment. smoking during pregnancy or the poorer eating habits of smoking mothers (Hui, Nelson, Yu, Li, & Fok, 2003). On a community level, the only variable that showed a relationship with weight status was living near a traffic In line with other reports (Kuhle, Allen, & Veugelers, dense route (< 300 m), which was only associated with z- 2010), excess television screen time, before the age of 3 BMI at age 3. Growing evidence suggests air pollution as years-old was associated with overweight at 4 years-old. a risk factor for the development of obesity and over- Various mechanisms may explain this association; pro- weight. Several epidemiological studies have reported an longed periods of inactivity, the opportunity to eat snacks, association between various ambient and indoor air pollu- a decline in energy expenditure, and the influence of com- tants and obesity (Limaye & Salvi, 2014). Evidence-based mercials on food choices and nutrition. Consequently the research to link children's health, physiology, and behav- American Academy of Pediatrics (American Academy of ior to atmospheric extremes is an important area of future Pediatrics, 2013) has recommended for over a decade that research. parents limit television viewing to no more than 2 h per day. This study indicates that greater understanding of the risk factors identified might help with early prevention and This study showed that when a child presented symp- intervention programs. The fact that we included a com- toms of DSM-IV feeding disorder of infancy before the munity population of young children within a small age age of 3 (a child's refusal to eat certain food groups, tex- range specific to preschool age group adds to existing tures, solids or liquids or refusal to accept new foods), knowledge. there was a negative association with overweight. The main study strengths are the longitudinal assess- Some aspects of early parenting style were associated ment of risk factors of overweight in preschool children, with later excess of body weight such as inconsistent dis- the inclusion of a large set of potential predictors including cipline, fewer rules and corporal punishment. Whitaker, 10 ACCIÓN PSICOLÓGICA, junio 2019, vol. 16, nº. 1, 1-12. ISSN: 2255-1271 https://doi.org/10.5944/ap.16.1.22098 individual, family and community level factors and a large References community sample (from the general population). Achenbach, T. M. & Rescorla, L. A. (2003). Manual for Some of the study’s limitations should also be consid- the ASEBA Adult Forms & Profiles. Burlington, ered when interpreting the results. Had the data been avail- VT: University of Vermont, Research Center for able, additional relevant risk factors potentially contrib- Children. uting to weight gain in childhood could have been ex- plored, such as, parental overweight, physical activity or American Academy of Pediatrics. (2013). Children, energy expenditure. However, this study is the first to ex- Adolescents, and the Media. Pediatrics, 132(5), amine a broad variety of risk factors for overweight in 957–961 https://doi.org/10.1542/peds.2013-2656 Spanish preschoolers. Future studies should explore more detailed information on these risk factors. Another study Afifi, T. O., Mota, N., MacMillan, H. L., & Sareen, J. limitation is that measurements of the children’s weight (2013). Harsh Physical Punishment in Childhood and height were taken from medical records, although rou- and Adult Physical Health. Pediatrics, 132(2), tine weight measurements have been found to be highly 333–340. https://doi.org/10.1542/peds.2012-4021 accurate, which supports their usefulness in both clinical practice and research (Howe, Tilling, & Lawlor, 2009). Baker, J. L., Olsen, L. W., & Sørensen, T. I. (2007). Another limitation is that few families of low socioeco- Childhood Body-mass Index and the Risk of nomic status participated in our study, possibly families of Coronary Heart Disease in Adulthood. New higher socioeconomic status are overrepresented in the England Journal of Medicine, 357, 2329–2337. study. Finally, some scales of the Spanish version of the https://doi.org/10.1056/NEJMoa072515 CBQ had low alphas in the current sample, although this must be interpreted with caution due to the small number Cominato, L., Di Biagio, G. F., Lellis, D., Franco, R. R., of items and the CBQ scale Attention Focusing (which Mancini, M. C., & de Melo M., E. (2018). Obesity was the only scale that was statistically significant) had an Prevention: Strategies and Challenges in Latin alpha value of α = .69 (almost .70 which is considered America. Current Obesity Reports, 7(2), 97–104 acceptable for a measurement scale). https://doi.org/10.1007/s13679-018-0311-1. The preschool years are a period of rapid growth and De Onis, M., Blössner, M., & Borghi, E. (2010). Global habit formation. Knowledge of the age at which the asso- Prevalence and Trends of Overweight and Obesity ciation between early specific risk factors and weight sta- among Preschool Children. American Journal of tus becomes significant can provide avenues for early in- Clinical Nutrition, 92, 1257–1264. https://doi.org/ terventions. 10.3945/ajcn.2010.29786 Acknowledgements Ezpeleta, L., de la Osa, N., & Doménech, J. M. (2014). Prevalence of DSM-IV Disorders, Comorbidity and Funding was from Spanish Ministry of Science and In- Impairment in 3-year-old Spanish Preschoolers. novation grant PSI2009-07542 and Ministry of Economy Social Psychiatry and Psychiatric Epidemiology, and Competitiveness grant PSI2012-32695. Thanks to the 49, 145–155. https://doi.org/10.1007/s00127-013- Comissionat per a Universitats I Recerca, Departament 0683-1 d’Economia I Coneixement de la Generalitat de Catalunya (2014 SGR 312). The authors are grateful to the partici- Ezpeleta, L., de la Osa, N., Granero, R., Doménech, J. pating schools and families for their involvement. M., & Reich, W. (2011). The Diagnostic Interview for Children and Adolescents for Parents of Preschool Children: Psychometric Properties in the General Population. Psychiatry Research, 190, ACCIÓN PSICOLÓGICA, junio 2019, vol. 16, nº. 1, 1-12. ISSN: 2255-1271 https://doi.org/10.5944/ap.16.1.22098 11 137–144. https://doi.org/10.1016/j.psychres.2011. 94, 950–954. https://doi.org/10.1136/adc.2009. 04.034 162552 Faith, M. S. & Hittner, J. B. (2010). Infant Temperament Hui, L. L., Nelson, E. A., Yu, L. M., Li, A. M., & Fok, and Eating Style Predict Change in Standardised T. F. (2003). Risk Factors for Childhood Weight Status and Obesity Risk at 6 years of AGE. Overweight in 6- to 7-y-old Hong Kong Children. International Journal of Obesity, 34, 1515–1523. International Journal of Obesity and Related https://doi.org/10.1038/ijo.2010 Metabolic Disorders, 27(11), 1411–1418. Freedman, D. S., Mei, Z., Srinivasan, S. R., Berenson, Jansen, P. W., Mensah, F. K., Nicholson, J. M., & Wake, G. S., & Dietz, W. H. (2007). Cardiovascular risk M. (2013). Family and Neighbourhood Factors and Excess Adiposity among Overweight Socioeconomic Inequalities in Childhood Children and Adolescents: The Bogalusa Heart Trajectories of BMI and Overweight: Longitudinal Study. Journal of Pediatrics, 150(1), 12–17. Study of Australian Children. PloS One, 8, e69676. https://doi.org/10.1016/j.jpeds.2006.08.042 https://doi.org/10.1371/journal.pone.0069676 Gardner, D. S., Hosking, J., Metcalf, B. S., Jeffery, A. Kuhle, S., Allen, A. C., & Veugelers, P. J. (2010). N., Voss, L. D., & Wilkin T. J. (2009). Contribution Prevention potential of risk factors for childhood of Early Weight Gain to Childhood Overweight and overweight. Canadian Journal of Public Health, Metabolic Health: A Longitudinal Study 101(5), 365–368. (EarlyBird 36). Pediatrics, 123(1), e67–73. https://doi.org/10.1542/peds.2008-1292. Lakshman, R., Elks, C. E., & Ong, K. K. (2012). Childhood Obesity. Circulation, 126, 1770–1779. Gordon-Larsen, P., Adair, L. S., Nelson, M. C., & https://doi.org/10.1161/CIRCULATIONAHA.111. Popkin, B. M. (2004). Five-year Obesity Incidence 047738 in the Transition Period between Adolescence and Adulthood: The National Longitudinal Study of Limaye, S. & Salvi, S. (2014). Obesity and Asthma: The Adolescent Health. American Journal of Clinical Role of Environmental Pollutants. Immunology and Nutrition, 80, 569–575. https://doi.org/10.1093/ Allergy Clinics of North America, 34(4), 839–855. ajcn/80.3.569 https://doi.org/10.1016/j.iac.2014.07.005 Hodges, E. L., Cochrane, C. E., & Brewerton, T. D. Miqueleiz, E., Lostao, L., Ortega, P., Santos, J. M., (1998). Family Characteristics of Binge Eating Astasio, P., & Regidor, E. (2014). Trends in the Disorder Patients. International Journal of Eating Prevalence of Childhood Overweight and Obesity Disorders, 23(2), 145–151. According to Socioeconomic Status: Spain, 1987- https://doi.org/10.1002/(SICI)1098-108X(199803) 2007. European Journal of Clinical Nutrition, 68, 23:2<145::AID-EAT4>3.0.CO;2-K 209–214. https://doi.org/10.1038/ejcn.2013.255 Hollingshead, A. A. (1975). Four-factor Index of Social Mitchell, E. A., Stewart, A. W., Braithwaite, I., Hancox, Status. New Haven, CT: Yale University, R. J., Murphy, R., Wall, C., ...Beasley, R. (2016). Unpublished manuscript. Birth Weight and Subsequent Body Mass Index in Children: An International Cross-Sectional Study. Howe, L. D., Tilling, K., & Lawlor, D. A. (2009). Pediatric Obesity, 1–6. https://doi.org/10.1111/ Accuracy of Height and Weight Data from Child ijpo.12138 Health Records. Archives of Disease in Childhood, 12 ACCIÓN PSICOLÓGICA, junio 2019, vol. 16, nº. 1, 1-12. ISSN: 2255-1271 https://doi.org/10.5944/ap.16.1.22098 Osa, N. de la, Granero, R., Penelo, E., Domènech, J. M., related Behaviours: A Pooled Analysis of Six & Ezpeleta, L. (2014). Psychometric Properties of European Studies. Obesity Reviews, 13, 29–41. the Alabama Parenting Questionnaire - Preschool https://doi.org/10.1111/j.1467-789X.2011.00959.x revision (APQ-Pr) in 3 year-old Spanish preschoolers. Journal of Child and Family Studies, Veldhuis, L., van Dooremaal, M., Kroeze, W., Renders, 23, 776–784. https://doi.org/ 10.1007/s10826-013- C. M., Hirasing, R. A., & Raat, H. (2013). Ethnic 9730-5 Background and Overweight Among 5-year-old CHILDREN: The "Be Active, Eat Right" study. Perez-Bonaventura, I., Granero, R., & Ezpeleta, L. Pediatrics, 10, 861246. https://doi.org/10.1155/ (2015). The Relationship Between Weight Status 2013/861246 and Emotional and Behavioral Problems in Spanish Preschool. Journal of Pediatric Psychology, 40, Weng, S. F., Redsell, S. A., Swift, J. A., Yang, M., & 455–463. https://doi.org/10.1093/jpepsy/jsu107 Glazebrook, C. P. (2012) Systematic Review and Meta-analyses of Risk Factors for Childhood Reilly, J. J., Methven, E., McDowell, Z. C., Hacking, B., Overweight Identifiable during Infancy. Archives Alexander, D., Stewart, L., …Kelnar, C. J. (2003). of Disease in Childhood, 97(12), 1019–1026. Health Consequences of Obesity. Archives of https://doi.org/10.1136/archdischild-2012-302263 Disease in Childhood, 88, 748–752. https://doi.org/ 10.1136/adc.88.9.748 Whitaker, R. C., Phillips, S. M., Orzol, S. M., & Burdette, H. L. (2007). The Association between Rothbart, M. K., Ahadi, S. A., Hershey, K. L., & Fisher, Maltreatment and Obesity among Preschool P. (2001). Investigations of Temperament at 3–7 Children. Child Abuse & Neglect, 31(11–12), years: The Children’s Behavior Questionnaire. 1187–1199. https://doi.org/10.1016/j.chiabu.2007. Child Development, 72(5), 1394–1408. 04.008 https://doi.org/10.1111/1467-8624.00355 World Health Organization Multicentre Growth Stamatakis, E., Wardie, J., & Cole, T. J. (2010). Reference Study Group. (2006). WHO child Childhood Obesity and Overweight Prevalence Growth Standards Based on Length/Height, Weight Trends in England: Evidence for Growing and Age. Acta Paediatrica, 450, 76–85. Socioeconomic Disparities. International Journal https://doi.org/10.1080/08035320500495548 of Pediatric Obesity, 34(1), 41–47. https://doi.org/10.1038/ijo.2009.217 Unitat d'Epidemiologia i de Diagnòstic en Psicopatologia del Desenvolupament. (2009). Protocolo de Factores de Riesgo. Versión para preescolares [Risk Factors Protocol. Preschool version]. Barcelona, Spain: Departament de Psicologia Clínica i de la Salut, Universitat Autònoma de Barcelona. van Stralen, M. M., te Velde, S. J., van Nassau, F., Brug, J., Grammatikaki, E., Maes, L., …Chinapaw, M. J. (2012). ToyBox-study Group. Weight Status of European Preschool Children and Associations with Family Demographics and Energy Balance-