YPMED-04260; No of Pages 5

Preventive Medicine xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Preventive Medicine

journal homepage: www.elsevier.com/locate/ypmed

1Q4 Parenting style and obesity risk in children

2Q5 Lisa Kakinami a,b,c,⁎, Tracie A. Barnett d,e,f, Louise Séguin g,h, Gilles Paradis a

F
3 a
Department of Epidemiology, Biostatistics and Occupational Health, McGill University, Montréal, Québec, Canada
4 b
Department of Mathematics and Statistics, Concordia University, Montréal, Québec, Canada

O
5 c
PERFORM Centre, Concordia University, Montréal, Québec, Canada
6 d
Epidemiology and Biostatistics Unit, INRS-Institut Armand-Frappier, Université du Québec, Laval, Québec, Canada
7 e
Department of Exercise Science, Concordia University, Montréal, Québec, Canada

O
8 f
Centre de recherche du Centre Hospitalier Universitaire (CHU) Sainte-Justine, Canada
9 g
Département de médecine sociale et préventive, Université de Montréal, Canada
10
Q6 h
Institut de recherche en santé publique l'université de Montréal (IRSPUM), Canada

R
1 1 a r t i c l e i n f o a b s t r a c t

P
12 19
13 Available online xxxx Background. Parents play a critical role in their children's lifestyle habits. The objective was to assess the effect 20
of parenting style on the risk of childhood obesity, and to determine whether poverty was a moderator of the
D 21
14 Keywords: association. 22
15 Children and adolescents
Methods. Participants were from the 1994–2008 cross-sectional samples of the National Longitudinal Survey 23
16 Poverty
of Children and Youth (NLSCY), a nationally representative survey of Canadian youth. Factor and cluster analyses 24
17
E
Obesity
18 Parenting style
identified four parenting styles consistent with Baumrind's parenting style prototypes. Multivariable logistic 25
regression assessed the risk of obesity based on parenting style after adjusting for covariates. Analyses were 26
27
T

stratified by age (preschool: 2–5 years of age, n = 19,026; school-age: 6–11 years of age, n = 18,551) and the
moderating effect of poverty (household income b low income cut-offs adjusted for household size and 28
geographic region) was assessed. Analyses used sampling and bootstrap weights. 29
C

Results. In multivariable analyses, compared to authoritative parenting, preschool- and school-age children 30
with authoritarian parents were 35% (95% CI: 1.2–1.5) and 41% (CI: 1.1–1.8) more likely to be obese, respectively. 31
In preschool children, poverty moderated this association: authoritarian and negligent parenting was associated 32
E

with 44% (CI: 1.3–1.7) and 26% (CI: 1.1–1.4) increased likelihood of obesity, respectively, but only among the 33
children not living in poverty. In school-age children, poverty was not a moderator. 34
R

Conclusions. Parenting style is associated with childhood obesity, but may be moderated by poverty. Successful 35
strategies to combat childhood obesity should reflect the independent and interactive associations of 36
sociodemographic and social–familial influences on health especially in early childhood. 37
R

© 2015 Elsevier Inc. All rights reserved. 38

42
40
39
O

41

43 Background practices”) and has found that a child's obesity risk is influenced by 54
C

parental control over specific behaviors (such as if the parent controls 55

44 Obesity prevalence in Canadian youth has doubled since 1980 the types and amount of snacks a child is allowed) (Berge, 2009). 56
45 (Tremblay and Willms, 2000). In addition to biologic factors, General parenting style has been shown to affect adolescent risk 57
N

46 sociodemographic and environmental factors are associated with obesity; behaviors, but its effect on childhood obesity is poorly understood 58
47 lower socioeconomic status (SES) (Wang and Beydoun, 2007), living in (Newman et al., 2008; Berge, 2009). While some studies report an asso- 59
U

48 neighborhoods with poor walkability, and poor access to fresh fruits ciation between parenting style and BMI z-scores (Berge et al., 2010; 60
49 and vegetables increase the risk for obesity (Powell et al., 2006; Nilsen Chen and Kennedy, 2005; Olvera and Power, 2010), few studies report 61
Q7
50 et al., 2010). an association with obesity (Agras et al., 2004; Wake et al., 2007). The 62
51 In children, environmental factors specific to family life may play an inconsistent findings might suggest that small studies were underpow- 63
52 additional role (Berge, 2009). Previous research in the social–familial do- ered to detect associations (Agras et al., 2004), or that the study was 64
53 main has primarily focused on behavior-specific parenting (“parenting assessed in young children who had not yet undergone their adiposity 65
rebound (Wake et al., 2007), or that other variables may be moderating 66
the association. 67
⁎ Corresponding author at: Concordia University, Department of Mathematics and
In particular, despite the mounting evidence that poverty is an inde- 68
Statistics and the PERFORM Centre, 1455 de Maisonneuve West, LB-901 Montréal,
Québec H3G 1M8 Canada. pendent predictor of obesity, sociodemographic characteristics have 69
E-mail address: lisa.kakinami@concordia.ca (L. Kakinami). been inadequately investigated in the existing literature (Berge, 2009; 70

http://dx.doi.org/10.1016/j.ypmed.2015.03.005
0091-7435/© 2015 Elsevier Inc. All rights reserved.

Please cite this article as: Kakinami, L., et al., Parenting style and obesity risk in children, Prev. Med. (2015), http://dx.doi.org/10.1016/
j.ypmed.2015.03.005
 2 L. Kakinami et al. / Preventive Medicine xxx (2015) xxx–xxx

71 Sleddens et al., 2011). It has been suggested that the relationship Weight status and health 136
72 between parenting styles and children's obesity risk may be affected The child's height and weight were reported by the PMK and were converted 137
73 by the larger social environments and contexts in which children live. to body mass index (BMI) values and compared to the World Health Organiza- 138
tion (WHO) growth curves (De Onis et al., 2007). The literature indicates parents 139
74 For instance, neighborhood safety, or parenting norms based on
tend to underestimate height (thus artificially inflating their child's BMI). Thus 140
Q8
75 different SES may warrant different parenting styles to be effective
the main outcome of interest was whether the child was obese, rather than 141
76 (Patrick et al., 2013). But research is limited and further investigation
whether the child was overweight or obese. While this analytic decision was 142
77 of potential moderation is needed (Sleddens et al., 2011). not expected to fully eliminate misclassification, because larger errors in height 143
78 Because obesity tracks from childhood to adulthood (Juhola et al., and weight are required to erroneously misclassify a normal-weight child as 144
79 2011) and is associated with several health risks (Freedman et al., obese, it was likely to diminish misclassification. A child was identified as obese 145
80 2001; Moriarty-Kelsey and Daniels, 2010), reducing childhood obesity if their BMI percentile was N the 97.7th percentile for their age and sex as defined 146
81 is a top public health priority (World Health Organization, 2012). Thus by the WHO growth curves. BMI is not always used for children less than 2 years 147
82 the aim of this study is to investigate the association between parenting of age, thus only the participants between the ages of 2–11 (n = 43,437) com- 148
83 style and obesity in a large representative sample of Canadian youth, prised the eligible sample (Daniels, 2009). An additional 5860 participants 149
84 and to assess if poverty moderates this association. Specifically, we were excluded due to missing data. 150

F
85 hypothesized that 1) compared to authoritative parenting, authoritarian,
Poverty 151
86 permissive and negligent parenting would be associated with a higher
The past year's household income (before taxes and deductions) was 152

O
87 risk of childhood obesity and 2) the strength of this association would
reported by the PMK and compared to the low income cut-offs established by 153
88 differ between children living and not living in poverty. Statistics Canada which are adjusted for household size and geographic region 154
(Giles, 2004). Poverty was defined as annual household income below the 155

O
cut-off. A subanalysis maintaining household income as a low income cut-off 156
89 Methods
ratio did not affect the results; thus poverty as defined above was used in all 157
analyses. 158

R
90 A cross-sectional sample of participants from the National Longitudinal Sur-
91 vey of Children and Youth (NLSCY) database comprised the study population.
Covariates 159
92 The study's sampling design and study protocol have been published previously

P
Sociodemographic characteristics known to be associated with parenting 160
93 (Statistics Canada and Human Resources Development Canada, 1995). Briefly,
style or health were included in multivariable analyses. These included contin- 161
94 samples representative of the non-institutionalized civilian youth (ages 0–11)
uous covariates (child's age, birth weight, birth order of the child, maternal age, 162
95 throughout Canada was collected bi-annually since 1994. The cross-sectional
family functioning score, neighborhood cohesion, neighborhood safety) and 163
96 samples of interest for this study (n = 66,813) were those obtained from
D
binary covariates (child's sex, whether the PMK was an immigrant, whether 164
97 1994–2008 (8 cycles).
the PMK had at least a high-school education, whether the family was a two- 165
98 Data collection consisted of computer-assisted questionnaires completed
parent or a single-parent household). The family functioning score measured 166
E
99 by the person most knowledgeable (PMK) of the child and included
the family's ability to deal with stressors as a unit and was based on the validated 167
100 sociodemographic and SES characteristics, family characteristics, neighborhood
McMaster Family Assessment Device (Byles et al., 1988). A sensitivity analysis 168
101 characteristics, and child's characteristics. Statistics Canada obtained informed
T

additionally adjusted for neighborhood cohesion and safety. Because the study 169
102 consent and assent from parents and youth respectively, and protects against
sample was reduced due to missing neighborhood data but results were unaf- 170
103 the identity disclosure of any individual or organization (Canadian Institutes
fected, neighborhood characteristics were omitted from the final models. 171
C

104 Of Health Research et al., 2010). Thus, as a secondary data analysis with no iden-
105 tifiable information, ethics approval from the Institutional Review Board was
Data analysis 172
106 not required.
E

All analyses were conducted with SAS 9.3. Due to the complex sampling 173
107 Measures strategy used in the NLSCY, sampling and bootstrap weights were used. The as- 174
R

sociation between parenting style and obesity risk was assessed in multivariable 175
108 Parenting style logistic regression adjusting for all the covariates previously described. Authori- 176
177
R

109 Parenting style describes the general characteristics of the interactions tative parenting style served as the referent. Whether poverty was a moderator
110 between parent and child. Initially described by Baumrind (1966) and expanded in this association was also assessed. Due to evidence of differences in parenting 178
111 upon by Maccoby and Martin (1983) four archetypes of parenting style occur style between preschool (2–5 years of age) and school-age children (6–11 years 179
O

112 along two dimensions of responsiveness (nurturing or warmth) and demanding- of age) in this study sample, analyses were stratified by age category. 180
113 ness (establishing boundaries and enforcing them): authoritative (responsive
114 and demanding), authoritarian (not responsive but demanding), permissive Results 181
C

115 (responsive but not demanding) and negligent (not responsive and not
116 demanding). Twenty-five questions encompassing different interactions between
The analytic sample (n = 37,577) did not significantly differ from 182
117 the PMK and child were collected on 5-point Likert scales. Parenting styles were
N

those excluded due to missing data (n = 5860) in sex or age of the 183
118 identified according to the methodology outlined by Chao and Willms (2002)
child, but was less likely to be from an immigrant family or poor, and 184
119 and have previously been shown to correspond well with Baumrind's parenting
120 styles. Briefly, a factor analysis on the PMK–child interactions was conducted. the PMK was more likely to have at least a high-school education 185
U

121 The number of factors was based on the least number of factors with a cumulative (data not shown). The implications of this are described in the discus- 186
122 eigenvalue of approximately 1. Five variables had low loadings (b0.40) on all sion. Most of the missing data occurred with the covariates; the propor- 187
123 factors and were excluded from further analysis. Due to non-unique factor tion missing parenting style was b5%. A sensitivity analysis wherein 188
124 loadings, a varimax rotation was used and the remaining 20 variables had high multivariable models maximized sample size by minimizing the num- 189
125 loadings on a single factor. The four factors that were identified were consistent ber of covariates did not differ from the results presented here. 190
126 with those previously reported (Chao and Willms, 2002) and were labeled as Many of the characteristics were significantly different between the 191
127 ‘reasons with child’ (5 variables), ‘difficulty managing the child’ (5 variables), younger and older samples (Table 1). Approximately one-third of the 192
128 ‘responsive’ (5 variables), and ‘uses a firm approach’ (5 variables) (eTable 1).
younger children were obese (29.8%), and 18.8% of the school-age 193
129 Any variable with negative factor loadings was reverse coded; higher values indi-
children were obese. Nearly half of the parents of younger children 194
130 cated higher cumulative scores of the variables comprising that factor. A scaled
were authoritative (41.9%), and the rest were nearly evenly divided 195
131 score was calculated for each factor, and a cluster analysis was conducted to
132 group similar observations together using k-means clustering. The parenting into authoritarian, permissive, and negligent (15.7%, 20.1%, and 22.3%, 196
133 style that best corresponded to a cluster was identified based on the characteris- respectively). Parents of school-age children were primarily authorita- 197
134 tics provided by Baumrind (1966) and were consistent with the literature (Chao tive (31.9%), or negligent (31.1%), followed by permissive (21.7%) or 198
135 and Willms, 2002). authoritarian (15.3%). 199

Please cite this article as: Kakinami, L., et al., Parenting style and obesity risk in children, Prev. Med. (2015), http://dx.doi.org/10.1016/
j.ypmed.2015.03.005
 L. Kakinami et al. / Preventive Medicine xxx (2015) xxx–xxx 3

Q1
t1:1 Table 1 for this association. Furthermore, the inclusion of poverty in the multi- 217
t1:2 Weighted frequencies and demographic characteristics of the National Longitudinal Sur- variable model did not affect our results, and poverty was not indepen- 218
t1:3 vey of Children and Youth (NLSCY) 1994–2008 Canadian study sample.
dently associated with obesity after adjusting for parenting style in the 219
t1:4 2–5 year olds 6–11 year olds pa model. 220
(n = 19,026) (n = 18,551)

t1:5 Child's characteristics Discussion 221

t1:6 Male 50.9% 51.% 0.35
Q2
t1:7 Age, mean (SD) 3.8 (0.4) 9.2 (0.5) 0.0001
In this large sample of Canadian youth, authoritarian parenting was 222
t1:8 Obese 29.8% 18.8% 0.0001
t1:9 Birth order 1.8 (0.3) 1.7 (0.4) 0.06 consistently associated with an increased risk of childhood obesity 223
t1:10 when compared to authoritative parenting. However, further analysis 224
t1:11 Parental characteristics
revealed the poverty moderated this association in younger children. 225
t1:12 PMKab is an immigrant 17.5% 15.1% 0.002
t1:13 PMK N high school education 88.0% 88.7% 0.20 Among younger children not living in poverty, an authoritarian or neg- 226
t1:14 Maternal age 33.3 (2.1) 38.1 (2.3) 0.0001 ligent parenting style was associated with an increased likelihood of 227
t1:15 obesity compared to an authoritative parenting style. Parenting style 228

F
t1:16 Household characteristics
t1:17 Family functioning score 8.1 (2.0) 8.1 (2.3) 0.60 was not associated with obesity in younger children from poor house- 229
t1:18 Poverty 15.9% 14.9% 0.11 holds. Compared with authoritative parenting, school-age children of 230

O
t1:19 Parenting stylebc authoritarian parents were more likely to be obese, irrespective of 231
t1:20 Authoritative 41.9% 31.9% 0.0001 household income. 232
t1:21 Authoritarian 15.7% 15.3%
The results are consistent with the literature reporting the positive 233

O
t1:22 Permissive 20.1% 21.7%
t1:23 Negligent 22.3% 31.1% effects of authoritative parenting on decreasing BMI (Berge et al., 2010; 234
a Sleddens et al., 2011; Rodenburg et al., 2013), and the negative effects 235
t1:24 P-value comparing the distribution between 2–5 year olds and 6–11 year olds.

R
t1:25 ab
PMK: person most knowledgeable about the child. of negligent or permissive parenting on an increased risk of elevated 236
t1:26 bc
Last year's annual household income b low income cut-off established by Statistics BMI (Olvera and Power, 2010; Lissau and Sorensen, 1994). However, 237
Canada. the association between parenting style and obesity has been less 238

P
clear (Sleddens et al., 2011). While this study had nearly 40,000 239
200 In multivariable logistic regression, compared to younger children respondents, previous studies averaged a few hundred respondents. 240
201 with authoritative parents, younger children with authoritarian parents By contrast to our study, several cross-sectional studies reported no as-
D 241
202 were 35% more likely to be obese (95% Confidence Interval [CI]: 1.2–1.5, sociation between authoritarian parenting and elevated BMI. However, 242
Q9
203 p = 0.0001) (Table 2). Children with permissive or negligent parents these other studies were conducted among children of different ethnic 243
204 were not more likely to be obese, and poverty was an independent pre- backgrounds, and cultural heritage and ethnicity may play a role in 244
E
205 dictor of obesity (Adjusted Odds Ratio [AOR]: 1.2, CI: 1.0–1.4, p = 0.03). parental interactions with the child and should be further assessed 245
206 However, our moderation analysis revealed a significant interaction be- (Park and Walton-Moss, 2012). 246
T

207 tween poverty and parenting style; parenting style was only associated A limitation of the existing literature is the role that SES may play in 247
208 with obesity among children not living in poverty. In non-poor house- the association between parenting style and childhood obesity. While 248
C

209 holds, compared to children with authoritative parents, children with the evidence for low SES and the risk of childhood obesity is mounting 249
210 authoritarian or negligent parents were 44% (CI: 1.3–1.7, p = 0.0001) (Shrewsbury and Wardle, 2008), studies of parenting style as an inde- 250
211 and 26% (CI: 1.1–1.4, p = 0.01) more likely to be obese, respectively. pendent predictor of obesity risk in children have been mixed, often- 251
E

212 There was no association between parenting style and obesity among times hampered by small sample sizes, or multivariable models which 252
213 the children living in poverty. did not adjust for SES (Olvera and Power, 2010; Sleddens et al., 2011). 253
R

214 Compared to authoritative parenting, authoritarian parenting was The moderation of poverty on the risk of obesity has only been assessed 254
215 associated with a 41% higher likelihood of the school-age child being by Topham et al. (2010) who reported that poverty was a moderator for 255
216 obese (CI: 1.1–1.8, p = 0.007). However, poverty was not a moderator the association between permissive parenting and obesity, but no 256
R

t2:1 Table 2
O

t2:2 Adjusted odds ratios and 95% Confidence Intervals from logistic regression assessing the association of obesity based on parenting style and poverty from the National Longitudinal Survey
t2:3 of Children and Youth (NLSCY) 1994–2008 Canadian study sample.

t2:4 Full sample Poor households Non-poor households

C

t2:5 AORa p AORb p AORb p

t2:6 (95% CI) (95% CI) (95% CI)
N

t2:7 Younger children

t2:8 Parenting style
t2:9 Authoritative 1.0 1.0 1.0
U

Q3
t2:10 Authoritarian 1.35 (1.2–1.5) 0.0001 1.0 (0.7–1.5) 1.0 1.44 (1.3–1.7) 0.0001
t2:11 Permissive 1.07 (0.9–1.2) 0.40 1.1 (0.7–1.6) 0.7 1.05 (0.9–1.2) 0.36
t2:12 Negligent 1.13 (1.0–1.4) 0.13 0.6 (0.4–1.0) 0.05 1.26 (1.1–1.4) 0.01
t2:13 Poverty (yes v no) 1.20 (1.0–1.4) 0.03
t2:14
t2:15 School-age childrenc
t2:16 Parenting style
t2:17 Authoritative 1.0
t2:18 Authoritarian 1.41 (1.1–1.8) 0.007
t2:19 Permissive 1.09 (0.9–1.3) 0.35
t2:20 Negligent 0.86 (0.7–1.0) 0.12
t2:21 Poverty (yes v no) 1.08 (0.8–1.3) 0.51
a
t2:22 Adjusted for age, sex, poverty, whether the person most knowledgeable about the child was an immigrant, whether the person most knowledgeable about the child had at least a high-
t2:23 school education, whether the household was a single or two-parent household, family functioning score, birth order of the child, birth weight of the child, and maternal age.
b
t2:24 Adjusted for age, sex, whether the person most knowledgeable about the child was an immigrant, whether the person most knowledgeable about the child had at least a high-school
t2:25 education, whether the household was a single or two-parent household, family functioning score, birth order of the child, birth weight of the child, and maternal age.
c
t2:26 Poverty was not a moderator for the association between parenting style and obesity among older children, thus no stratified results are presented.

Please cite this article as: Kakinami, L., et al., Parenting style and obesity risk in children, Prev. Med. (2015), http://dx.doi.org/10.1016/
j.ypmed.2015.03.005
 4 L. Kakinami et al. / Preventive Medicine xxx (2015) xxx–xxx

257 association for authoritarian parenting was found, and negligent par- obesity when compared to authoritative parenting, but this association 323
258 enting was not assessed. In addition, the study had a limited sample was moderated by household income among the sample of preschool- 324
259 size, and used parental education and occupation for SES, which have age children. Thus results suggest that both poverty and parenting 325
260 been shown to be distinctive from household income (Geyer et al., style are important predictors of childhood health and successful strat- 326
261 2006). egies to combat childhood obesity need to reflect their independent and 327
262 Study results suggest that compared to authoritative parenting, interactive associations on health. However, in order to successfully 328
263 authoritarian parenting is associated with obesity risk throughout implement a strategy that combines individual-level targets with larger 329
264 early and middle-childhood. The child's ability to self-regulate their societal targets, more research on the qualitative differences of social– 330
265 energy intake has been suggested as one possible mechanism. familial characteristics based on sociodemographic characteristics is 331
266 Authoritarian parenting (characterized by low-responsiveness and needed. 332
267 high-demandingness) may translate to parents a) not responding to Supplementary data to this article can be found online at http://dx. 333
268 children's cues of hunger and/or satiety, and b) demanding or control- doi.org/10.1016/j.ypmed.2015.03.005. 334
269 ling the child's energy intake (Frankel et al., 2012). Thus, children's
270 ability to regulate their own energy intake is underdeveloped and Conflict of interest statement 335

F
271 these children may be more likely to overindulge when given the The authors declare that there are no conflicts of interest. 336
272 opportunity (Patrick et al., 2013).

O
273 Consistent with the literature, poverty was a strong independent Acknowledgments 337
274 predictor of obesity and moderated the association among younger chil-
275 dren (Wells et al., 2010). Among school-age children, however, poverty LK was supported by a postdoctoral fellowship from the Fonds de 338

O
276 was no longer an independent predictor of obesity risk after adjusting recherche du Québec -Santé. TAB is a Fonds de la Recherche en Santé 339
277 for parenting style and other SES characteristics such as maternal immi- du Québec Junior 2 Scholar. GP holds a Canadian Institutes of Health 340
278 gration status and parental education. Thus it is unclear whether the in-

R
Research (CIHR) Applied Public Health Research Chair in Chronic Diseases 341
279 dependent effect of poverty on obesity risk may overshadow the effect Prevention (CPP 85705). Data were collected by Statistics Canada. These 342
280 of parenting style in younger children. Although parenting style may agencies were not involved in the study design, data analyses, data inter- 343

P
281 not impact obesity risk during early childhood among poor children in pretation or manuscript writing and submission processes. LK had full ac- 344
282 this study, parenting style has been associated with affecting poverty's cess to all of the data in the study and takes responsibility for the integrity 345
283 impact on cognitive development, risky behaviors in adolescence, and of the data and the accuracy of the data analysis.
D 346
Q10
284 other health and wellness factors (Linver et al., 2002; Wen and
285 Shenassa, 2012).
References 347
286 Due to the cross-sectional nature of this study, we cannot assess
E
287 if parenting style adapted to different needs of the children based Agras, W., Hammer, L.D., Mcnicholas, F., Kraemer, H.C., 2004. Risk factors for childhoood 348
288 on their different ages. In addition, parent–child interactions were overweight: a prospective study from birth to 9.5 years. J. Pediatr. 145, 20–25. 349
Banach, A., Wade, T.J., Cairney, J., Hay, J.A., Faught, B.E., O'Leary, D.D., 2007. Comparison of 350
T

289 based on parental report, and variations due to cultural differences, anthropometry and parent-reported height and weight among nine year olds. Can. 351
290 sociodemographic characteristics and parental perceptions may exist J. Public Health 98, 251–253. 352
353
C

291 (Baumrind, 1966, 1967). Thus while the Baumrind parenting styles Chen, J.-L., Kennedy, C., 2005. Factors associated with obesity in Chinese-American
children. Pediatr. Nurs. 31, 110–115. 354
292 may be able to generally group parenting practices, direct observations Baumrind, D., 1966. Effects of authoritative parental control on child behavior. Child Dev. 355
293 of parent–child interactions are strongly needed. Parenting styles were 37, 887–907. 356
E

294 identified based on cluster analysis across multiple cross-sectional sam- Baumrind, D., 1967. Child care practices anteceding three patterns of preschool behavior. 357
Genet. Psychol. Monogr. 75, 43–88. 358
295 ples between the years 1994–2008. Thus although parenting styles Berge, J.M., 2009. A review of familial correlates of child and adolescent obesity: what has 359
R

296 were identified based on their common features, it is possible that the the 21st century taught us so far? Int. J. Adolesc. Med. Health 21, 457–483. 360
297 association between parenting styles with children's obesity risk may Berge, J.M., Wall, M., Loth, K., Neumark-Sztainer, D., 2010. Parenting style as a predictor of 361
adolescent weight and weight-related behaviors. J. Adolesc. Health 46, 331–338. 362
298 have changed during this time period. Longitudinal studies assessing
R

Byles, J., Byrne, C., Boyle, M.H., Offord, D.R., 1988. Ontario Child Health Study: reliability 363
299 the stability of parenting styles over time and across ages are needed. and validity of the general functioning subscale of the McMaster Family Assessment 364
300 As previously mentioned, parental report of height and weight likely ar- Device. Fam. Process 27, 97–104. 365
366
O

Canadian Institutes of Health Research, Natural Sciences And Engineering Research

301 tificially inflates their child's BMI. However, while the analytic decision
Council Of Canada, Social Sciences And Humanities Research Council Of Canada, 367
302 to assess obesity likely minimizes misclassification of normal-weight 2010. Tri-Council Policy Statement: Ethical conduct for research involving humans 368
303 children in this study sample, misclassification of overweight children (December). 369
C

304 as obese is still a concern as it is improbable that one-third of the Chao, R., Willms, J.D., 2002. The effects of parenting practices on children's outcomes. 370
Vulnerable Children: Findings from Canada's National Longitudinal Study of Children 371
305 2–5 year olds in this sample are truly obese. Although prevalence and Youth. University of Alberta Press. 372
N

306 estimates of obesity based on parental report among 9-year olds in Daniels, S.R., 2009. The use of BMI in the clinical setting. Pediatrics 124 (Suppl. 1), 373
307 the NLSCY sample was found to be similar to those obtained using S35–S41. 374
De Onis, M., Onyango, A.W., Borghi, E., Siyam, A., Nishida, C., Siekmann, J., 2007. Develop- 375
308 measured BMI (Banach et al., 2007), larger studies measuring height ment of a WHO growth reference for school-aged children and adolescents. Bull. 376
U

309 and weight in representative samples of children are needed. World Health Organ. 85, 660–667. 377
310 Residual confounding by unmeasured variables (e.g., physical activ- Frankel, L.A., Hughes, S.O., O'Connor, T.M., Power, T.G., Fisher, J.O., Hazen, N.L., 2012. 378
Parental influences on children's self-regulation of energy intake: insights from 379
311 ity and dietary intake) may also affect our results and should be further developmental literature on emotion regulation. J. Obes. (March 28). 380
Q11
312 assessed. Maternal depression has been suggested to also be a modera- Freedman, D.S., Khan, L.K., Dietz, W.H., Srinivasan, S.R., Berenson, G.S., 2001. Relationship 381
313 tor (Topham et al., 2010), but we were unable to assess this. Lastly, anal- of childhood obesity to coronary heart disease risk factors in adulthood: the Bogalusa 382
Heart Study. Pediatrics 108, 712–718. 383
314 yses were restricted to households with children between the ages of
Geyer, S., Hemstrom, O., Peter, R., Vagero, D., 2006. Education, income, and occupational 384
315 2–11 and with data on parenting style and may no longer be represen- class cannot be used interchangeably in social epidemiology. Empirical evidence 385
316 tative of Canadian youth. However, the proportion lost due to missing against a common practice. J. Epidemiol. Community Health 60, 804–810. 386
317 data was relatively low (14%) and to the best of our knowledge the Giles, P., 2004. Low Income Measurement in Canada. [Online], Minister of Industry, 387
Ottawa, Ontario, Canada (Available: http://www.statcan.gc.ca/pub/75f0002m/ 388
318 sample is the largest to date and may still be informative of general 75f0002m2004011-eng.pdf). 389
319 parenting practices and their associations with obesity in Canadian Juhola, J., Magnussen, C.G., Viikari, J.S., et al., 2011. Tracking of serum lipid levels, blood 390
320 youth. pressure, and body mass index from childhood to adulthood: the Cardiovascular 391
Risk in Young Finns Study. J. Pediatr. 159, 584–590. 392
321 In this large cross-sectional sample of Canadian youth, authoritarian Linver, M.R., Brooks-Gunn, J., Kohen, D.E., 2002. Family processes as pathways from 393
322 parenting was associated with an increased likelihood of childhood income to young children's development. Dev. Psychol. 38, 719–734. 394

Please cite this article as: Kakinami, L., et al., Parenting style and obesity risk in children, Prev. Med. (2015), http://dx.doi.org/10.1016/
j.ypmed.2015.03.005
 L. Kakinami et al. / Preventive Medicine xxx (2015) xxx–xxx 5

395 Lissau, I., Sorensen, T.I., 1994. Parental neglect during childhood and increased risk of Sleddens, E.F., Gerards, S.M., Thijs, C., De Vries, N.K., Kremers, S.P., 2011. General parenting, 422
396 obesity in young adulthood. Lancet 343, 324–327. childhood overweight and obesity-inducing behaviors: a review. Int. J. Pediatr. Obes. 423
397 Maccoby, E.E., Martin, J.A., 1983. Socialization in the context of the family: parent–child 6, e12–e27. 424
398 interaction. In: Hetherington, M. (Ed.), Handbook of child psychology. Willey. Statistics Canada & Human Resources Development Canada, 1995. National Longitudinal 425
399 Moriarty-Kelsey, M., Daniels, S., 2010. Childhood obesity is the fuel that fires Survey of Children: overview of survey instruments for 1994–95 data collection, 426
400 adult metabolic abnormalities and cardiovascular disease. Child Obes. 6, 250–256. Cycle 1 (Ottawa, ON). 427
401 Newman, K., Harrison, L., Dashiff, C., Davies, S., 2008. Relationships between parenting Topham, G.L., Page, M.C., Hubbs-Tait, L., et al., 2010. Maternal depression and socio- 428
402 styles and risk behaviors in adolescent health: an integrative literature review. Rev. economic status moderate the parenting style/child obesity association. Public Health 429
403 Lat. Am. Enfermagem 16, 142–150. Nutr. 13, 1237–1244. 430
404 Nilsen, S.M., Krokstad, S., Holmen, T.L., Westin, S., 2010. Adolescents' health-related die- Tremblay, M.S., Willms, J.D., 2000. Secular trends in the body mass index of Canadian 431
405 tary patterns by parental socio-economic position, the Nord-Trondelag Health children. CMAJ 163, 1429–1433. 432
406 Study (HUNT). Eur. J. Pub. Health 20, 299–305. Wake, M., Nicholson, J.M., Hardy, P., Smith, K., 2007. Preschooler obesity and parenting 433
407 Olvera, N., Power, T.G., 2010. Brief report: parenting styles and obesity in Mexican styles of mothers and fathers: Australian National Population Study. Pediatrics 120, 434
408 American children: a longitudinal study. J. Pediatr. Psychol. 35, 243–249. e1520–e1527. 435
409 Park, H., Walton-Moss, B., 2012. Parenting style, parenting stress, and children's health- Wang, Y., Beydoun, M.A., 2007. The obesity epidemic in the United States–gender, age, 436
410 related behaviors. J. Dev. Behav. Pediatr. 33, 495–503. socioeconomic, racial/ethnic, and geographic characteristics: a systematic review 437
411 Patrick, H., Hennessy, E., Mcspadden, K., Oh, A., 2013. Parenting styles and practices in and meta-regression analysis. Epidemiol. Rev. 29, 6–28. 438
412 children's obesogenic behaviors: Scientific gaps cand future research directions. Wells, N.M., Evans, G.W., Beavis, A., Ong, A.D., 2010. Early childhood poverty, cumulative 439
413 440

F
Child Obes. S73-86 (Suppl. 9). risk exposure, and body mass index trajectories through young adulthood. Am.
414 Powell, L.M., Slater, S., Chaloupka, F.J., Harper, D., 2006. Availability of physical activity- J. Public Health 100, 2507–2512. 441
415 related facilities and neighborhood demographic and socioeconomic characteristics: Wen, X., Shenassa, E.D., 2012. Interaction between parenting and neighborhood quality 442

O
416 a national study. Am. J. Public Health 96, 1676–1680. on the risk of adolescent regular smoking. Nicotine Tob. Res. 14, 313–322. 443
417 Rodenburg, G., Kremers, S.P., Oenema, A., Van De Mheen, D., 2013. Associations of paren- World Health Organization, 2012. Prioritizing Areas for Action in the Field of 444
418 tal feeding styles with child snacking behaviour and weight in the context of general Population-Based Prevention of Childhood Obesity. [Online], World Health Organiza- 445
419 parenting. Public Health Nutr. 1–10. tion, Geneva, Switzerland (Available: http://apps.who.int/iris/bitstream/10665/ 446

O
420 Shrewsbury, V., Wardle, J., 2008. Socioeconomic status and adiposity in childhood: a system- 80147/1/9789241503273_eng.pdf?ua=1, Accessed August 20 2014). 447
421 atic review of cross-sectional studies 1990–2005. Obesity (Silver Spring) 16, 275–284.

R
P
D
E
T
C
E
R
R
O
C
N
U

Please cite this article as: Kakinami, L., et al., Parenting style and obesity risk in children, Prev. Med. (2015), http://dx.doi.org/10.1016/
j.ypmed.2015.03.005