Dietary behaviors of adults born prematurely may explain future risk for cardiovascular disease

Accepted Manuscript Dietary Behaviors of Adults Born Prematurely May Explain Future Risk for Cardiovascular Disease Mastaneh Sharafi, Valerie B. Duffy, Robin J. Miller, Suzy B. Winchester, Tania B. Huedo-Medina, Mary C. Sullivan PII:

S0195-6663(16)30007-1

DOI:

10.1016/j.appet.2016.01.007

Reference:

APPET 2823

To appear in:

Appetite

Received Date: 9 October 2015 Revised Date:

15 December 2015

Accepted Date: 7 January 2016

Please cite this article as: Sharafi M., Duffy V.B, Miller R.J, Winchester S.B, Huedo-Medina T.B. & Sullivan M.C, Dietary Behaviors of Adults Born Prematurely May Explain Future Risk for Cardiovascular Disease, Appetite (2016), doi: 10.1016/j.appet.2016.01.007. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

1

Title Page

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

Dietary Behaviors of Adults Born Prematurely May Explain Future Risk for Cardiovascular Disease

RI PT

Mastaneh Sharafi a,Valerie B Duffy a*, Robin J Miller b, Suzy B Winchester c, d, Tania B. Huedo-Medina a, Mary C Sullivan d,

AC C

EP

TE D

Phone: (860) 486-1997 Fax: (860) 486-5375 E-Mail: [email protected]

M AN U

* Address correspondence to: Valerie B Duffy University of Connecticut Department of Allied Health Sciences College of Agriculture and Natural Resources 358 Mansfield Road, Unit 2101 Storrs, CT 06269-2101

SC

Affiliations: a Department of Allied Health Sciences, University of Connecticut, Storrs, CT ; b School of Nursing, University of Connecticut, Storrs, CT; c Brown Center for Study of Children at Risk Women & Infants Hospital, Providence, RI; d College of Nursing, University of Rhode Island, Kingston, RI

Abbreviations: ANCOVA – Analysis of Covariance; BMI – Body Mass Index; CVD – Cardiovascular Disease; FT-adults – Full term born adults; HDL – High Density Lipoprotein; HEPI – Healthy Eating Preference Index; HPTs – Healthy Preterm Infants; LDL – Low Density Lipoprotein; MPTs – Medical Preterm Infants; NPTs – Neurological Preterm Infants; PT-adults – Preterm born adults; SGA – Small for Gestational Age Preterms; TC – Total Cholesterol.

ACCEPTED MANUSCRIPT

Abstract

TE D

M AN U

SC

RI PT

Being born prematurely associates with greater cardiovascular disease (CVD) risk in adulthood. Less understood are the unique and joint associations of dietary patterns and behaviors to this elevated risk among adults who are born prematurely. We aimed to model the associations between term status, dietary and lifestyle behaviors with CVD risk factors while accounting for the longitudinal effects of family protection, and medical or environmental risks. In wave-VIII of a longitudinal study, 23-year olds born prematurely (PT-adults, n=129) and full term (FT-adults, n=38) survey-reported liking for foods/beverages and activities, constructed into indexes of dietary quality and sensationseeking, dietary restraint and physical activity. Measured CVD risk factors included fasting serum lipids and glucose, blood pressure and adiposity. In bivariate relationships, PT-adults reported lower dietary quality (including less affinity for protein-rich foods and higher affinity for sweets), less liking for sensation-seeking foods/activities, and less restrained eating did FT-adults. In comparison to nationally-representative values and the FT-adults, PT-adults showed greater level of CVD factors for blood pressure and serum lipids. In structural equation modeling, dietary quality completely mediated the association between term status and HDL-cholesterol (higher quality, lower HDLcholesterol) yet joined term status to explain variability in systolic blood pressure (PTadults with lowest dietary quality had highest blood pressures). Through lower dietary quality, being born prematurely was indirectly linked to higher cholesterol/HDL, higher LDL/HDL and elevated waist/hip ratios. The relationship between dietary quality and CVD risk was strongest for PT-adults who had developed greater cumulative medical risk. Protective environments failed to attenuate relationships between dietary quality and elevated CVD risk among PT-adults. In summary, less healthy dietary behaviors contribute to elevated CVD risk among young adults who are born prematurely.

EP

Key words: preterm; dietary quality; structural equation modeling; food preference; restraint, food neophobia; cardiovascular disease

AC C

27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72

2

ACCEPTED MANUSCRIPT

1. Introduction

74

Adults who were born prematurely can have increased rates of risk factors for

75

cardiovascular disease (CVD) (Rogers & Velten, 2011; Sipola-Leppanen et al., 2014),

76

including elevated blood pressure (de Jong, Monuteaux, van Elburg, Gillman, & Belfort,

77

2012; Parkinson, Hyde, Gale, Santhakumaran, & Modi, 2013), impaired lipid metabolism

78

(Parkinson et al., 2013), insulin resistance (Finken et al., 2006; Hofman et al., 2004) and

79

greater central adiposity (Barbieri et al., 2009). These risk factors may result from the

80

prematurity, including adverse prenatal environments with maternal health complications

81

(fetal programming), and/or challenges in developing healthy diets, including food

82

sensitivities and intolerances (Barbieri et al., 2009; Behrman & Butler, 2007; Kaseva et

83

al., 2013; Silveira et al., 2012), food neophobia (Migraine et al., 2013) and fussy eating

84

(Samara, Johnson, Lamberts, Marlow, & Wolke, 2010), which parallels avoidance of risk

85

taking (Hack et al., 2012; Roberts et al., 2013) and sensation seeking (Alley & Potter,

86

2011; Allin et al., 2006; Pliner & Melo, 1997). Intrauterine growth restriction with or

87

without prematurity is linked to greater preference for sweets (Ayres et al., 2012; Barbieri

88

et al., 2009; Silveira et al., 2012), salty taste (Stein, Cowart, & Beauchamp, 2006) and

89

lower affinity/intakes of protein-rich foods and fruits (Kaseva et al., 2013; Migraine et

90

al., 2013) in children and higher intakes of carbohydrates (Barbieri et al., 2009) and lower

91

intakes of fruits, vegetables (Kaseva et al., 2013) and alcohol (Cooke, 2004; Roberts et

92

al., 2013) in adults. A recent analysis of a large prospective cohort study suggested the

93

additive effects of unhealthy lifestyle (scored from diet, physical activity, smoking,

94

alcohol consumption and BMI) and low birth weight on greater relative risk of type 2

95

diabetes through multivariate logistic modeling (Li et al., 2015), yet the study sample had

AC C

EP

TE D

M AN U

SC

RI PT

73

3

ACCEPTED MANUSCRIPT

lower representation of low-birth weight, relied on self-reported birth weight and did not

97

control for the effect of being born prematurely. Appropriate self-regulation or restraint

98

could counteract the negative effects of less healthy dietary behaviors (Johnson, Pratt, &

99

Wardle, 2012; Tomiyama, Moskovich, Haltom, Ju, & Mann, 2009; Wardle & Beales,

100

RI PT

96

1987).

Linking diet with risk of chronic disease in a longitudinal study requires

102

assessment instruments that capture habitual dietary behaviors and are low burden to the

103

participants. Multiple dietary records and/or food frequency surveys may or may not

104

reflect habitual dietary intake, are time and labor intensive, and challenged by

105

misreporting (Thompson & Subar, 2013). The present study used reported

106

liking/disliking of foods and beverages as a rapid proxy of dietary intake that may capture

107

more habitual behaviors. Survey reporting of food liking links genetic variation in oral

108

sensation with dietary intake as adults (Duffy, Hayes, Sullivan & Faghri, 2009; Törnwall

109

et al, 2014), assuming that genetic variation influences development of food preferences

110

(Mennella, Pepino, & Reed, 2005; Pallister et al., 2015) and that, overtime, we eat what is

111

liked and avoid what is not. Survey liking correlates with reported intake (Erinosho et al.,

112

2015; Tuorila et al., 2008), nutritional biomarkers, including carotenoid (Scarmo et al.,

113

2012) and fatty acid (Hutchins-Wiese, Duffy, Watkins, Li, & Kenny, 2013) status, variety

114

of metabolites (Pallister et al., 2015) and adiposity (Duffy, Hayes, Sullivan, & Faghri,

115

2009; Duffy et al., 2007; Peracchio, Henebery, Sharafi, Hayes, & Duffy, 2012). Inclusion

116

of non-food items on a liking survey supports the ability to compare food liking across

117

individuals by generalizing the scale (Bartoshuk et al, 2006) and can capture affinity for

118

sensation seeking foods and activities (Terasaki & Imada, 1988). Responses from a liking

AC C

EP

TE D

M AN U

SC

101

4

ACCEPTED MANUSCRIPT

119

survey can be formed into a dietary quality index that has acceptable internal consistency

120

as well as construct and criterion-related validity with the ability to predict adiposity and

121

nutritional biomarker, carotenoid status (Sharafi, Peracchio, et al., 2015). The present study aimed to extend previous research by assessing the direct and

RI PT

122

indirect relationships between dietary behaviors and measures of CVD risk in adults who

124

were born prematurely (PT-adults) versus full-term (FT-adults) via Structural Equation

125

Modeling (SEM). SEM is a relatively novel multivariate technique that is proposed to

126

overcome limitations of the traditional statistical methodologies by allowing testing direct

127

and indirect (e.g. mediation) relationships among multiple variables (A. F. Hayes, 2009).

128

CVD is a complex, multifactorial disease process and although SEM seems very

129

promising in the study of complex pathways, it has been rarely been used in the field of

130

diet related CVD risks (Antonogeorgos et al., 2012). Specifically, we hypothesized that

131

PT-adults would report greater preference for high fat/sweet/salty foods, lower preference

132

for healthy foods, and less dietary restraint on eating. Consistent with risk avoidance

133

(Allin et al., 2006; Hack et al., 2012), we hypothesized that PT-adults would report

134

lowest affinity for sensation-seeking activities and foods. Finally, since environment

135

influences early development (Bradley et al., 1994; M. M. McGrath & Sullivan, 2003;

136

Pridham, Brown, Sondel, Clark, & Green, 2001), we examined how measures of risk and

137

protective environments during childhood influenced relationships between dietary

138

quality and CVD risk factors in adults who were born prematurely.

M AN U

TE D

EP

AC C

139

SC

123

140

2. Material and Methods

141

2.1. Design and Subjects

5

ACCEPTED MANUSCRIPT

142

This study was part of a wave VIII of a prospective longitudinal study of premature infants. The 215 premature and full-term infants were recruited at birth with follow-up

144

assessments at ages 18m, 30m, 4y, 8y, 12y, 17y and now at age 23. The recruitment at

145

birth occurred between 1985-1989 from a Level-III neonatal intensive care unit in a

146

specialty hospital in southern New England. Parents were recruited by research nurses

147

and provided written consent. The recruitment criteria, identified a priori, included

148

neonatal diagnoses, birth weight, absence of maternal mental illness, and English as a

149

primary language. Five groups were classified by neonatal morbidity. Full-term infants (n

150

= 55) were born from mothers with uncomplicated labor and delivery, absent of neonatal

151

diagnoses, >2,500 grams birth weight, >38 weeks gestational age, and were recruited

152

from the same medical center at the same time as the preterm infants. All preterm infants

153

were <37 weeks gestational age, birth weight <1850 grams, and classified into four

154

groups based on their neonatal illness. Healthy preterm infants (HPTs; n = 33) had no

155

medical or neurological illness. Preterm infants with medical illness (MPTs; n = 60) had

156

bronchopulmonary dysplasia (defined as oxygen requirement at 28 days of life),

157

respiratory distress syndrome, necrotizing enterocolitis (Bell, 1978), and/or sepsis.

158

Preterm infants with neurological illness (NPTs; n = 36) had Grade III & IV IVH (Papile,

159

Burstein, Burstein, & Koffler, 1978), meningitis, and/or shunted hydrocephalus. Small

160

for Gestational Age (SGA; n = 31) preterm infants had birth weights below the 10th

161

percentile for length with/without neonatal illness (Lubchenco, Hansman, & Boyd, 1966).

162

Socioeconomic status (SES) was equally distributed within and across neonatal groups at

163

recruitment. Fewer than 10% of the parent(s) declined participation in the study.

AC C

EP

TE D

M AN U

SC

RI PT

143

6

ACCEPTED MANUSCRIPT

164

In wave VIII, participants were contacted within 6 months of their 23rd birthday. The sample at age 23 years consisted of 180 of the original 215 participants recruited at birth

166

(84% retention rate from birth). Reasons for attrition (n = 35) included travel/scheduling

167

issues, military deployment, or inability to locate/contact. Those who dropped from the

168

study between the birth time point and age 23 did not favor a specific neonatal group and

169

there were no differences between those who participated and those who dropped out at

170

age 23 in neonatal illness, SES, parent education, marital status, or race.

SC

171

RI PT

165

Of the 180 subjects, 13 did not complete the dietary surveys, leaving a final sample of 167 (129 PT-adults and 38 FT-adults) described in Table 1. Informed consent was

173

obtained during the home visit, followed by collection of fasting capillary blood by the

174

research nurse. The laboratory visit included a physical assessment, health interviews and

175

dietary questionnaires. Informed consent was obtained from parents at recruitment, and at

176

each time point (with the exception of age 23 years). Child assent was obtained at ages 8

177

years, 12 years, and 17 years, and informed consent by study participants at age 23 years.

178

The reported health information was verified by medical record. Trained research staff

179

collected all data, with 90%+ inter-rater reliability (Sullivan, 2008) and were blinded to

180

term group. The university and hospital Institutional Review Boards approved the study

181

at each time point. Study participants received $100 for participating at age 23 years.

183 184

TE D

EP

AC C

182

M AN U

172

185 186

7

ACCEPTED MANUSCRIPT

Table 1. Descriptive characteristics of sample (N=167) at age 23 years. All

Term status

Preterm subgroups

(n =167)

Fullterm (n = 38)

Preterm All (n = 129)

HPTs (n = 24)

MPTs (n = 52)

NPTs (n = 28)

SGA (n = 25)

88

20

68

14

24

10

20

79

18

61

10

28

18

5f

White

144

33

111

19

43

25

24

African-American

15

4

11

2

6

2

1

Other

8

1

7

3

3

1

0

40±13.8

41.1±14. 0

39.7±13.8

37.5±14. 8

39.3±13. 9

40.0±14. 7

42.3±11. 8

7.6±2.9

7.4±3.0

7.6±2.9

7.9±3.1

7.7±3.2

7.8±2.7

7.0±2.4

1.6±0.7

1.6±0.1

1.7±0.1

1.6±0.1

1.7±0.1

1.8±0.1

1.6±0.1

1.4±0.6

1.4±0.7

1.4±0.6

1.6±0.7

1.5±0.6

1.4±0.5

1.3±0.5

Female Male

Physical activityc Moderate (days/week) Vigorous (days/week) Smoked

9

29

3

14

6

6

6

16

8

3

2

3

7

23

11

7

2

3

3433±40 4

1264±323 f

1470±21 8g

39.9±0.7

30.0±2.5f

31.0±2.3

1179±28 6 28.6±2.1

1152±37 6

32.3±4.7

1278±32 6 29.3±2.3 h

h

32.2±2.6

26.0±5.8

26.6±6.4

25.8±5.7

28.4±10. 9i

25.4±5.1

25.0±5.0

26.1±5.9

5

2

3

0

0

2

1

73

14

59

11

27

12

9

56

10

46

8

17

10

10

33

12

11

5

8

4

4

EP

Never (0 times/day) Moderate (1-3 cigarettes/day) Frequent (≥ 4cigarettes/day) Birth weight (grams)a Gestational age (weeks)a Adult BMIa,e

TE D

SES

a, b

1755±96 9

AC C

Underweight (<18.5) Normal (18.5-<25) Overweight (25-<30) Obesity (≥ 30)

188 189

M AN U

Race

SC

Gender

RI PT

187

HPTs=Healthy Preterm infants; MPTs=Medical Preterm infants; NPTs=Neurological Preterm infants; SGA=Small for Gestational Age infants.

8

ACCEPTED MANUSCRIPT

a

190 191 192 193 194 195 196 197 198 199 200 201 202

Mean±SD. SES (socioeconomical status defined by Hollingshead Four-Factor Index at age 23). c Physical activity score is sum of weighted moderate and vigorous physical activities. d Missing data is due to participants opting not to answer to the question. There were no differences between those who completed smoking questions versus those who did not on demographic and health-related variables. e BMI (body mass index) is weight (kg)/height squared (m2). f Significant mean difference from full-term (P < .001). g Among preterm subgroups, significant difference from full-term (MPTs (P < .05), NPTs (P < .01), and SGA (P = .001). h Among preterm subgroups, significant difference from HPTs and SGA (Ps <0.001). i Significant variance difference from NPTs and MPTs (Ps < .05).

203

2.2. Measures

204

Table 2 shows when the dietary and lifestyle (smoking, physical activity) behaviors, risk

205

and protective indices as well as CVD risk factors were collected across the longitudinal

206

study.

M AN U

SC

RI PT

b

Cumulative Medical Risk Cumulative Protection Index

Dietary restraint

X

X

X

X

X

X

X

X

X

Age 23

X

X

X X X

AC C

Food neophobia

X

Age 17

EP

Environmental Risk Dietary intake (liking survey)

TE D

Table 2. Timing of data collection across the prospective study of premature infants 18 30 Age Age Age Birth months months 4 8 12

Lifestyle behaviors

X

CVD risk factors

X

207 208

2.3 Diet and Healthy Eating Preference Index

209

Using a general Visual Analog Scale (Byrnes & Hayes, 2013), participants reported

210

liking/disliking for 47 foods/beverages and 19 non-food items. Liking scores were treated

211

continuously (±100 points) for conceptual categorizing into statistically-reliable

9

ACCEPTED MANUSCRIPT

preference groups with latent variable analysis (Figure 1). A “variety” score was formed

213

from the number of liked (scale score ≥ 35) nutrient-dense items (e.g., fruits, vegetables,

214

whole grains) and mathematically converted to ±100 score. Consistent with diets for

215

CVD prevention (Eckel et al., 2013) and other dietary indices [e.g., Healthy Eating Index,

216

Diet Quality Index (Kourlaba & Panagiotakos, 2009)], conceptual weights were assigned

217

prior to averaging into a dietary quality index, Healthy Eating Preference Index (HEPI):

218

fruits/vegetables (+3), protein (+1), fat (-3), sweets (-3), salty food (-3), wine (+2), and

219

variety score (+2). Specific to red wine, moderate consumption is part of Mediterranean

220

diet that associates with favorable effects on cardiovascular risk factors (Rees, et al.,

221

2013), possibly through improved glucose metabolism (Chiva-Blanch et al., 2013) and

222

lipopolysaccharide concentrations (Clemente-Postigo et al., 2013). The averaging liking

223

for wine in the current study suggested that most participants were moderate drinkers.

224

From previous work in our laboratory, the HEPI among preschoolers showed good

225

construct and criterion related validity and adequate reliability, comparable to healthy

226

eating indexes formed from food frequency or dietary records (Sharafi, Peracchio, et al.,

227

2015) and was a significant predictor of carotenoid status and adiposity.

228

2.4. Other Dietary and Lifestyle Behaviors

229

The Revised Restraint Scale (10-item self-report questionnaire) was used to identify

230

dietary restraint from two subscales: concern for dieting, weight fluctuation (Herman &

231

Polivy, 1980). A Revised Food Neophobia Survey (Ritchey, Frank, Hursti, & Tuorila,

232

2003) assessed the avoidance of new and/or spicy foods (Alley & Potter, 2011; Alley,

233

Willet, & Muth, 2006)—two items formed a single question about liking multi-cultural

234

foods with responses on 5-point frequency versus original Likert scale (Bartoshuk, Duffy,

AC C

EP

TE D

M AN U

SC

RI PT

212

10

ACCEPTED MANUSCRIPT

235

Hayes, Moskowitz, & Snyder, 2006). The neophobia score was the average frequency of

236

neophobia (3 of 7) minus of neophilia (4 of 7) items (Ritchey et al., 2003). The dietary

237

restraint and food neophobia scores were statistically reliable (α > 0.7). A sensation seeking latent variable was formed from liking (Duffy et al., 2009;

RI PT

238

Duffy et al., 2007) for spicy foods (Byrnes & Hayes, 2013) (spicy rib, jalapeno, spicy

240

chips; α = 0.76) and risk-taking behaviors (driving fast, smoking cigarettes, drinking

241

vodka/gin, walking barefoot; α = 0.60) and food neophobia score (reverse coded).

The Personal and Social Development Questionnaire (Jessor, Costa, & Turbin,

M AN U

242

SC

239

2003) was used to assess smoking and physical activity. The subjects also responded to

244

the number of cigarettes per day within the past 6 months. Responses ranged from never

245

(0/day), moderate smoker (1-3 cigarettes/day), and frequent smoker (4 cigarettes/day to 2

246

packs or more). The associations between reported liking and frequency of smoking was

247

tested to assess whether reported liking could be used as a proxy for smoking.

248

Participants responded to the number of hours per week of moderate physical activity

249

(e.g., brisk walking, you still sweat, but may carry on a conversation), vigorous physical

250

activity (e.g., running, you still sweat, have a rapid heart rate, and can only talk in short

251

phrases) within a typical week within the past 6 months. Moderate and vigorous physical

252

activity ranged from minimal (0-2 hours/week, coded as 1), moderate (3-5 hours/week,

253

coded as 2) and intense (6 or more hours/week, coded as 3). A physical activity score was

254

formed from the sum of weighted codes of moderate (weight = +2) and vigorous (weight

255

= + 3) physical activity (Sharafi, Faghri, Huedo-Medina, Minski, & Duffy, 2015).

256

2.5. Risk and Protective Indexes

AC C

EP

TE D

243

11

ACCEPTED MANUSCRIPT

A cumulative medical risk index was comprised of neonatal variables, and medical and

258

neurological factors at birth, toddler, preschool, school-age, and adolescence, obtained

259

from medical records and physical health assessments administered by research nurse at

260

each time point, and developed using confirmatory factor analysis with appropriate fit

261

statistics (Winchester, Sullivan, & Msall, 2014). The neonatal factors were the Hobel

262

neonatal illness (Hobel, Hyvarinen, Okada, & Oh, 1973), length of hospital stay, birth

263

weight, gestational age. Medical neonatal factors were medical health status, necrotizing

264

enterocolitis, bronchopulmonary dysplasia, and hours of oxygen. Neurological neonatal

265

factors were neurological health status, intraventricular hemorrhage, and shunted

266

hydrocephalus. Medical health status was classified as normal (no abnormalities), suspect

267

(continued chronic respiratory, cardiac murmurs, referral for hearing, orthopedic), and

268

abnormal (asthma, allergies, diabetes, and/or autoimmune deficiencies). Neurological

269

health status was classified as normal (no abnormalities), suspect (fine motor weakness,

270

unilateral sensorineural hearing loss, uncorrected vision problems, atypical neurologic

271

findings in tone, reflexes, gait or movement with no specific diagnosis), and abnormal

272

(cerebral palsy, blindness, deafness, shunted hydrocephalus, uncontrolled seizures, and

273

attention-deficit/hyperactivity disorder (Niswander & Gordon, 1972; Prechtel & Beitema,

274

1967).

SC

M AN U

TE D

EP

AC C

275

RI PT

257

Environmental risk was assessed at birth using the Hollingshead Four-Factor

276

Index (Hollingshead, 1975), an indicator of social status and comprised of maternal and

277

paternal education and occupation levels.

278 279

A cumulative protection index was formed from variables in the environment proximal to the child at birth, 4 years, 8 years, and age 12 years (Winchester et al., 2014).

12

ACCEPTED MANUSCRIPT

Included in the index was stimulation within the home environment [Home Observation

281

for Measurement of the Environment; (Caldwell & Bradley, 2001)], maternal perception

282

of child vulnerability (M. M. McGrath, 1989b), and maternal self-esteem (M. M.

283

McGrath, 1989a), and maternal involvement as well as maternal control style measured at

284

4, 8 and 12 years from laboratory paradigms. The cumulative protect index was

285

developed using confirmatory factor analysis with appropriate fit statistics (Winchester et

286

al., 2014).

SC

RI PT

280

Since breastfeeding data only were available for 59% of PT-adults and 74% of

288

FT-adults (61 of 99 PT-adults were breastfed versus 21 of 28 FT-adults), sub-analyses

289

were not conducted. Mothers reported to either breastfeed exclusively for 5.6 months

290

(M=5.6 months, SD=6.2, Median=4.0, 0-28 months) or a combination of formula and

291

breastmilk for 5.4 months (M=5.4, SD=8.3, Median=3.0, range=0-48 months).

292

2.6. Cardiovascular risk factor assessment

293

Blood pressure, adiposity, serum lipids and blood glucose served as CVD risk factors.

294

For blood pressure, participants rested in a sitting position for approximately 5 minutes

295

where a registered nurse determined their cuff size (where bladder width is 40% of the

296

arm circumference) and bladder length (80% of the circumference of the arm). Blood

297

pressure was recorded for three times in 5-minute intervals using a calibrated automated

298

(Dinamap) sphygmomanometer. The average of three systolic blood pressure (SBP)

299

measurements was analyzed.

TE D

EP

AC C

300

M AN U

287

Adiposity was measured by waist/hip ratio and body mass index. Using a 72”

301

paper tape measure, waist was the circumference at the umbilicus and hip at the widest

302

area of the hips. Participants were instructed to take a breath, exhale, and then take the

13

ACCEPTED MANUSCRIPT

waist measurement to avoid a falsely lower reading. Height was measured while standing

304

against a wall-mounted stadiometer and weight with a medical scale (calibrated before

305

each weighing), both without shoes, for BMI calculation. For capillary blood analysis,

306

participants were asked to fast for 8-10 hours before the assessment using the

307

CardioChek/PA meter (Polymer Technology Systems, Inc, Indianapolis, IN) (Panz, Raal,

308

Paiker, Immelman, & Miles, 2005) for serum glucose, total cholesterol (TC), High

309

Density Lipoprotein (HDL), and triglycerides. Low Density Lipoprotein (LDL) was

310

derived from the Friedewald formula: LDL= TC - HDL - (TG / 5) (Friedewald, Levy, &

311

Fredrickson, 1972). Total cholesterol/HDL (TC/HDL) and LDL/HDL ratios also were

312

analyzed as CVD risk factors.

313 314

2.7. Statistical analysis

M AN U

SC

RI PT

303

Data were analyzed using SPSS (version 17.0; SPSS Inc, Chicago) and M plus

316

(version 7.0; Mplus Inc, Los Angeles) software. Significance criterion was P ≤ .05.

317

Descriptive statistics characterized participants by term status and preterm subgroups,

318

comparing the values to published norms where appropriate using the standardized mean

319

difference (d) the effect size index (Hedges, 1981). Latent variable analysis formed HEPI

320

and sensation seeking. For the analysis, preterm subgroups were combined in one group

321

of PT-adults as preliminary analysis showed no significant differences in their diet and

322

CVD risk factors. The power analysis with medium effect size (d = 0.5) determined that

323

110 PT-adults and 32 FT-adults were adequate to capture mean differences between

324

adults born preterm versus full term (α = 0.05, 1-β = 0.80).

AC C

EP

TE D

315

14

ACCEPTED MANUSCRIPT

325

Analysis of covariance (ANCOVA) initially assessed for differences between PTadults and FT-adults regarding dietary, risk taking, and index variables. Multiple

327

regression analysis was conducted to describe the relationship among the dietary and

328

sensation seeking variables as well between birth weight and these variables in PT-adults.

329

Based on the results of the ANCOVA and regression analysis, a model was developed

330

and tested with SEM in order to identify the direct and indirect influences of term status

331

on CVD risk factors with potential effects of dietary (HEPI, dietary restraint, sensation

332

seeking latent variable), other lifestyle factors (physical activity, liking for smoking as

333

individual variable), demographic (gender, SES) variables, and adiposity. To detect a

334

medium effect size for a SEM including criteria for adequate model fit were non-

335

significant chi-square [χ2 (p > .05)] or χ2 to df ratio (χ2/df) < 2, with Comparative Fit

336

Index (CFI) ≥ 0.92 and Root Mean Square Error of Approximation (RMSEA) ≤ 0.05

337

(Ullman, 2001). The power analysis with a medium effect size (d = 0.5) for the main path

338

and small effects for each other paths and the mediation effects (d = 0.10), showed that

339

100 subjects were required to achieve 98% power with a minimum R2 = 0.23 and a

340

RMSEA < 0.05.

343

3. Results

AC C

342

EP

341

TE D

M AN U

SC

RI PT

326

As shown in Table 1, FT-adults and PT-adults (or subgroups) did not differ by SES,

344

smoking, exercise level or average BMI. However, HPTs had significantly greater

345

variances in BMI than NPTs and MPTs (Table 1), yet no significant differences in the

346

distributions of overweight and obese individuals. All groups were gender matched

15

ACCEPTED MANUSCRIPT

347

except for significantly fewer males in SGAs than FT-adults (χ2 = 4.87, P < .05). As

348

expected, FT-adults had higher birth weights than PT-adults (Table 1).

349

3.1. Descriptive finings—dietary quality index and subcomponents In the entire sample, conceptual food groups were formed that varied from most

RI PT

350

to least liked as follows: sweets, salty, fats, protein, fruits/vegetables, and wine. The

352

highest variability in liking ratings were seen in vegetables, fats, and wine (interquartile

353

ranges > 45 in 200 points scale), which indicates that the central 50% of ratings were

354

fairly wide apart. Only 15% of the sample reported liking more than half of the healthier

355

foods/beverages.

M AN U

356

SC

351

The conceptual food groups were then tested by latent variable analysis for internal consistency and overall structure. Figure 1 shows the final items included for

358

each food group and measures of model fit indicating an adequate overall structure for

359

the HEPI. Wine and variety scores added after the latent variable analysis to form the

360

final HEPI score. HEPI score was normally distributed (Shapiro-Wilk = 0.99, P = 0.26),

361

averaging -39.6±38.0 SD.

AC C

EP

TE D

357

16

ACCEPTED MANUSCRIPT

Carrot Brussels sprout

Fruits/ Vegetables

Strawberry Cantaloupe

RI PT

Fried chicken

Fats

Bacon Hotdog Salty French-fries

Salty

SC

Salty popcorn

Fish Hamburger

Steak Cinnamon roll

Chocolate Pancake

363 364

Brownie

Sweets

Figure 1. Latent variable model of food groups used to develop Healthy Eating Preference Index (HEPI). Model fit: χ2/df = 1.4, CFI = 0.94 and RMSEA = 0.05.

TE D

362

Proteins

M AN U

Sausage

365

PT-adults reported significantly lower dietary quality (HEPI scores) than did the

367

FT-adults (F = 7.8, P < .01), driven by less liking for wine and protein-rich foods (Table

368

3). Although non-significant mean differences, PT-adults were distributed to greater

369

liking for sweets—approximately two-third of PT-adults were distributed toward high

370

sweet liking category (> strongly like) while FT-adults more evenly divided across high

371

and low liking categories (χ2 = 4.0, P < .05). The sub-analysis of PT-adults showed that

372

lower birth weight was associated with a greater sweet liking (b = -0.21, P < .05) and

373

lower HEPI scores (b = 0.18, P < .05) with no significant associations for fats, protein

374

and salty foods.

AC C

EP

366

17

ACCEPTED MANUSCRIPT

Table 3. Healthy Eating Preference Index score and sub-components by term statusa,b. Healthy Eating Preference Index Indicators

FT-adults PT-adults mean (standard error) -24.8 (6.1) -44.1 (3.3) 21.1 (5.2) 18.1 (2.8) 29.9 (6.1) 32.4 (3.3) 36.8 (5.3) 42.6 (2.9) 30.1 (5.2) 17.9 (2.8) 45.1 (4.5) 51.0 (2.4) 31.4 (9.3) 5.2 (5.0) 1.0 (6.1) -7.6 (3.3)

377 378 379

a

380

3.2. Descriptive findings—other dietary and lifestyle behaviors

FT-adults = Full term born adults; PT-adults = Preterm born adults. ANCOVA, controlling for influences of gender, BMI and dietary restraint

M AN U

381

b

.02 .72 .77 .45 .01 .25 .02 .12

SC

Total score Fruits/Vegetables Fat Salty Protein Sweets Wine Variety

P value

RI PT

375 376

Across all participants, 13% of males and 35% of females had high dietary restraint [scores > 15 (Allison & Baskin, 2009)]. Controlling for gender and BMI, PT-

383

adults reported lower dietary restraint, including lower weight fluctuation and less

384

concern for dieting, than did FT-adults (Table 4). Dietary restraint scores for PT-adults

385

averaged lower than published norms (Allison, Kalinsky, & Gorman, 1992; Boerner,

386

Spillane, Anderson, & Smith, 2004) (Table 4), while FT-adults did not. In sub-analysis of

387

PT-adults, higher dietary restraint was associated with higher fruits/vegetables liking (b =

388

0.18, P = .05), lower fat liking (b = -0.31, P < .001) and higher dietary quality (b = 0.24,

389

P < .01). PT-adults’ birth weight was not significantly associated with their dietary

390

restraint.

AC C

EP

TE D

382

391

18

ACCEPTED MANUSCRIPT

392 393

Table 4. Total restraint and sub-scores were significantly lower among preterm born adults (PT-adults) than full term adults (FT-adults)a and lower than normsb,c. Effect Sized d (95% CI)

Total dietary restraint Weight fluctuation

400

FT-adults Vs. Norm

PT-adults Vs. Norm

-26.8±6 -43.4±3 5.1, .03

-0.27 (-0.37, 0.29)

-2.38 (-0.55, -0.18)

20.4±5

0.10 (-0.29, 0.36)

-1.00 (-0.50, -0.13)

18.2±3

F, P values

5.2, .02

Concern for dieting 33.2±6 31.4±3 3.9, .05 0.67 (-0.17, 0.48) -0.69 (-0.35, 0.02) ANCOVA controlling for gender and BMI. b Norm were pooled data of college-age students (Allison et al., 1992; Boerner et al., 2004). c Based on calculated effect size (Hedges, 1981). d Effect size (i.e. standardized mean difference) was the mean of each group individually minus the mean of the norm group, divided by their correspondent pooled standard deviation.

SC

a

M AN U

394 395 396 397 398 399

PTadults

RI PT

FTadults

PT-adults were lower sensation seekers than FT-adults (b = -0.26, P = .01) in analysis with the latent variable model of sensation seeking formed from liking for spicy

402

food, liking for thrill seeking behavior and reverse-coded neophobia score (χ2 = 0.18, df

403

= 2, P = 0.91; CFI = 1.0 and RMSEA = 0.0). Looking at the food neophobia subscale,

404

across all participants, higher neophobia tended to associate with lower dietary quality,

405

but significantly in sub-analysis of PT-adults (HEPI; b = -0.28, P < .01), including lower

406

liking for Fruits/Vegetables (b = -0.26, P < .01), proteins (b = -0.25, P < .01), and wine (b

407

= -0.28, P < .01) and liking fewer healthy foods (variety score; b = -0.23, P = .01).

EP

AC C

408

TE D

401

For other lifestyle behaviors, the liking and frequency of smoking did not differ

409

significantly between PT- and FT-adults. The liking for smoking cigarettes was

410

associated with frequency of smoking, such that frequent smokers>moderate

411

smokers>non-smokers (F = 5.1, p < .01), suggesting liking for smoking as a proxy for

412

frequency of smoking cigarettes. PT- and FT- adults also reported no significant

413

differences in moderate, vigorous or overall physical activity. The sub-analysis of PT-

19

ACCEPTED MANUSCRIPT

414

adults showed no associations between birth weight and physical activity.

415

3.3. Relationships among risk and protective indexes, dietary quality and CVD risk

416

factors As expected, PT-adults had higher cumulative medical risk scores than FT-adults

RI PT

417

(F = 91.6, P < .001). The cumulative protection index and the environmental risk index

419

and their subcomponents did not differ by term status. These indexes were categorized

420

into high and low groups to test main and interactive effects with term status to predict

421

dietary quality and CVD risk factors. For dietary quality, there was a significant

422

interaction between term status and cumulative protective index—high protection did not

423

improve dietary quality for PT-adults but improved dietary quality 4-fold for FT-adults

424

(Figure 2).

M AN U

SC

418

425

429 430 431

EP

428

AC C

427

TE D

426

20

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

432

Figure 2. Cumulative family protection index interacts with term status to predict dietary quality scores (F = 4.42, P < .05) in ANCOVA controlling for gender, dietary restraint and BMI. PT-adults: Adults born premature; FT-adults: Adults born full-term.

438

categories on associations between term status and CVD risk factors. There was

439

significant interaction between dietary quality and medical risk (F = 6.1, P < .05) on

440

blood pressure; high cumulative medical risk group with low dietary quality had six

441

mm/Hg higher systolic pressure (Figure 3). Similar interaction effects trended for TG,

442

LDL and Waist/Hip ratio (P ≤ 0.2).

TE D

433 434 435 436 437

AC C

EP

There were no significant main or interactive effects of risk and protective

21

443 444 445 446

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 3. Dietary quality interacts with medical risk status to predict systolic blood pressure (F = 6.1, P < .05) in ANCOVA controlling for gender, BMI and term status in all adults (N = 167). High medical risk group only consisted adults born prematurely.

447

3.4. Structural Equation Models depicting the influences of dietary quality on the term

449

status-CVD risk relationship

450

TE D

448

PT-adults, particularly males, had higher systolic/diastolic blood pressures, lower HDL, higher TC/HDL, and LDL/HDL than the reference 2011-2012 NHANES data

452

(Centers for Disease Control and Prevention (CDC) & National Center for Health

453

Statistics (NCHS)) matched for sex and age (large effect size ds of > 0.8). In contrast,

454

FT-adults from the current study had similar CVD risk factors. The exception was FT-

455

adults males and females had higher systolic/diastolic blood pressure than NHANES

456

(large effect size ds of > 0.8).

457

AC C

EP

451

Structural equation modeling tested the direct effects of term status on CVD risk

458

factors, controlling for gender (females have favorable risk factors than males), BMI

459

(elevated adiposity was associated with less favorable risk factors) and physical activity.

22

ACCEPTED MANUSCRIPT

Among CVD risk factors, SBP and HDL were significantly predicted by term status; PT-

461

adults had greater SBP (b = 0.17, P < .01) and lower HDL levels (b = 0.14, P < .05) than

462

FT-adults. However, expanding the models showed that dietary quality (HEPI) was a

463

complete mediator of the term status-HDL relationship (indirect effect = -0.04, P < .05);

464

being preterm was associated with lower dietary quality, and lower dietary quality, in

465

turn, was associated with lower HDL, leaving the association between term status and

466

HDL non-significant (Figure 4). For the term status-SBP relationship, dietary quality had

467

no significant complete or partial mediating effect. However, dietary quality (b = -0.16, P

468

< .05) and term status (b = 0.14, P < .05) both significantly predicted SBP with no

469

significant interaction effect.

Restraint -0.17, *

0.14

-0.0 6

EP

0.15, *

0.12

Physical activity

, *)

HDL

0.34, ***

χ2 = 11.7, df = 11; P = 0.38; CFI=0.99

470 471 472 473 474 475 476 477 478

-0.06

-0.28, ***

0.20, **

AC C

HEPI

(-0. 14

BMI

-0. 07

-0.19, **

TE D

-0.03

Term

SES

M AN U

SC

RI PT

460

Gender

RMSEA= 0.02 Figure 4. Structural equation model showed dietary quality (HEPI) mediating the effect of term status on HDL-cholesterol level, controlling for the effects of dietary restraint (restraint), adiposity (BMI), physical activity and demographics (indirect effect coefficient=-0.04, p<.05). Term: full-term = 1, preterm = 2; SES = socio-economic status; HDL = High Density Lipoprotein; Gender: males = 1, females = 2; arrows are represented by standardized Beta; The coefficient in parenthesis represents the effect of term status on HDL –cholesterol before dietary quality was added to the model; * P ≤ .05, ** P ≤ .01, *** P ≤ .001.

23

ACCEPTED MANUSCRIPT

479 480

In the SEM models with LDL/HDL and TC/HDL, being born prematurely associated with lower dietary quality (b = -0.19, P = .01), and lower dietary quality, in

482

turn, associated with higher LDL/HDL (b = -0.17, P = .01) and TC/HDL (b = -0.16, P <

483

.05) ratios. No significant effects were found for fasting glucose or triglycerides. In all of

484

the SEM models, greater dietary restraint was a significant predictor of higher dietary

485

quality (b = 0.15, P < .05). Although level of physical activity was not associated with

486

the CVD risk factors, the sub-analysis of PT-adults showed that greater levels of physical

487

activity was associated with lower BMI (b = -0.18, P < .05), which in turn, was

488

associated with lower CVD risks. Neither the sensation seeking latent variable nor liking

489

for smoking cigarettes as an individual variable improved the SEM model fit to explain

490

CVD risk factors and thus, was not included in final SEM models.

SC

M AN U

PT-adults were not different from FT-adults for BMI (Figure 4). However, in

TE D

491

RI PT

481

separate SEM model, replacing BMI with waist/hip ratio, born prematurely (b = 0.19, P <

493

.01) was associated with higher waist/hip ratios. This association weakened to a trend (b

494

= 0.13, P = 0.08) when HEPI was added to the model, with HEPI significantly predicting

495

waist/hip ratio (b = -0.14, P < .05). The mediation effect of HEPI on term-waist/hip ratio

496

was not significant.

498 499

AC C

497

EP

492

4. Discussion

This study assessed direct and indirect effects of dietary quality and dietary

500

behaviors on cardiovascular disease risk factors in a prospectively followed sample of

501

young adults born prematurely versus a full-term comparison group recruited at the same

502

time. Adults born prematurely reported less healthy dietary behaviors as well as lower 24

ACCEPTED MANUSCRIPT

sensation seeking behaviors than the full-term peers. For measured CVD risk factors,

504

preterm adults exceeded the levels for blood pressure and serum lipids nationally-

505

representative data from the U.S. population, 2011-2012 NHANES (Centers for Disease

506

Control and Prevention, 2014) and had higher systolic blood pressure and central

507

adiposity and lower HDL-cholesterol than FT-adults from this study. Structural equation

508

models revealed that dietary quality joined term status as a separate and additive

509

contributor to systolic blood pressure, which is consistent with recent findings between

510

poor dietary behaviors, low birth weight and type 2 diabetes (Li et al., 2015). However,

511

new to our study was that dietary quality acted as possible mediator between term status

512

and cholesterol dyslipidemia. Furthermore, higher dietary quality attenuated some of the

513

CVD risk factor associations among preterm adults born with higher medical risk.

514

However, a protective family environment, while benefiting dietary quality for FT-adults

515

[similar to previous research (Savage, Fisher, & Birch, 2007)], was not associated with

516

dietary quality improvements among adults who were born prematurely.

SC

M AN U

TE D

517

RI PT

503

Born prematurely may influence responses to the food environment and the development of food preferences. Similar to previous research, PT-adults had less liking

519

for protein foods (Barbieri et al., 2009; Migraine et al., 2013) and wine (Cooke, 2004;

520

Roberts et al., 2013), and greater liking for sweets (Ayres et al., 2012; Barbieri et al.,

521

2009; Silveira et al., 2012) than FT-adults. The PT-adults with lower birth weight also

522

had greater liking for sweets while less liking for fruits/vegetables, consistent with prior

523

findings (Barbieri et al., 2009; Kaseva et al., 2013; Silveira et al., 2012). Contrary to

524

previous findings (Stein et al., 2006), birth weight was not associated with liking for salty

525

foods, which could be explained by differences in the study samples, as we only included

AC C

EP

518

25

ACCEPTED MANUSCRIPT

preterm adults with birth weight <1850 grams. New findings from our study were that

527

PT-adults (especially those with lowest birth weight) reported an overall dietary quality

528

incongruent with current dietary recommendations for CVD prevention (Eckel et al.,

529

2013). Individuals born prematurely also may not appropriately self-control these less

530

healthy food preferences. We found that PT-adults had less dietary restraint than the FT-

531

adults or norms of college-aged students, which may parallel findings of higher impulsive

532

eating behaviors among preschoolers born with intrauterine growth restriction (Silveira et

533

al., 2012).

SC

Lower sensation seeking among PT-adults in our study may further impede

M AN U

534

RI PT

526

development of healthy eating and related health outcomes. The PT-adults reported less

536

affinity for sensation seeking (spicy foods, risk taking behaviors) and lower willingness

537

to try new foods. Adverse neonatal environments (Behrman & Butler, 2007) are

538

associated with avoidance of novel stimuli (Pliner & Melo, 1997) consistent with PT-

539

adults reporting lower affinity for sensation seeking/risk behaviors. Although avoidance

540

of some risky behaviors (smoking, excessive alcohol consumption) decreases CVD risk

541

(Kajantie & Hovi, 2014), our findings and those of others (Contento, Zybert, & Williams,

542

2005; Goulet et al., 2008; Moreira, de Almeida, & Sampaio, 2005; Sproesser, Strohbach,

543

Schupp, & Renner, 2011) support that PT-adults could achieve healthier diets through

544

higher cognitive control, self-regulation, early and frequent exposures to novel foods

545

(Blissett & Fogel, 2013) and novel strategies to improve the sensory characteristics of

546

healthier foods (Bartoshuk & Klee, 2013; J. E. Hayes & Duffy, 2008; Sharafi, Hayes, &

547

Duffy, 2013; Wilkie & Capaldi Phillips, 2014).

AC C

EP

TE D

535

26

ACCEPTED MANUSCRIPT

548

We observed higher blood pressure, lower HDL-cholesterol and higher waist/hip ratio among PT-adults, which is consistent with early manifestation of blood pressure

550

elevations during childhood (Bonamy, Kallen, & Norman, 2012; de Jong et al., 2012;

551

Poplawska et al., 2012), abnormal lipid profile (Parkinson et al., 2013) and impaired fetal

552

growth relating to central fat distribution (Barbieri et al., 2009; Labayen et al., 2006;

553

Mathai et al., 2013; Thomas et al., 2011). Our findings of additive effects of term status

554

and dietary quality on blood pressure suggest that simultaneous improvement of both

555

prenatal and postnatal factors is needed to manage this CVD risk factor. We did not find

556

a significant relationship between liking for salty foods, term status and blood pressure

557

among PT-adults. Instead overall dietary quality acted as a potential mediator between

558

premature birth and multiple CVD risk factors—dietary quality was lower among PT-

559

adults simultaneously leading to lower HDL-cholesterol and higher TC/HDL-cholesterol,

560

LDL/HDL-cholesterol, and waist to hip ratio. Healthy lifestyle also may protect adults

561

born premature against CVD. Although our findings on the level of physical activity and

562

smoking behaviors at age 23 did not show differences between PT- and FT-adults, the

563

sub-analysis of preterm born adults revealed higher levels of physical activity lowered the

564

risks for CVD by reducing the adiposity levels. Similar to previous research (Kajantie &

565

Hovi, 2014), our results support that adults born prematurely can benefit from early

566

adoption of healthy dietary and lifestyle behaviors to promote health in aging.

SC

M AN U

TE D

EP

AC C

567

RI PT

549

The current study provides insights into early-life influences on the relationships

568

between diet and CVD risk in early adulthood. Through cumulative, longitudinal indexes,

569

we capture constructs of family protection, the environmental and medical risks from

570

birth through adolescence. We found no differences between PT- and FT-adults regarding

27

ACCEPTED MANUSCRIPT

the cumulative family protection score, which included maternal perception of child

572

vulnerability, maternal self-esteem, maternal involvement and maternal control style

573

from birth to adolescence. Surprisingly, the greater family protection was found to only

574

benefit adults born full term by positively influencing their dietary quality. As parent-

575

child interactions early in life shape children’s dietary behaviors (Cerro, Zeunert,

576

Simmer, & Daniels, 2002; Savage et al., 2007), we expected to see a similar benefit of

577

family protection for PT-adults. Even with similar food environments, PT-adults may still

578

be at greater dietary risks of CVD due to effects of the prenatal environment to shaping

579

their dietary preferences.

SC

M AN U

580

RI PT

571

As fetal programming theory (Barker, Eriksson, Forsen, & Osmond, 2002; Barker, Osmond, Forsen, Kajantie, & Eriksson, 2007; Sullivan, Hawes, Winchester, &

582

Miller, 2008) suggests, individuals exposed to adverse prenatal environments develop

583

compensatory physiological responses to survive that become maladaptive when the

584

environment alters (Sullivan et al., 2008). Rapid transition from undernourished

585

intrauterine environment to outside obesogenic environment may result in mismatches

586

between predicted and actual environments leading to altered food preferences and

587

dietary behaviors with increased susceptibility to future CVD disease. Heightened

588

attention to dietary quality might be most important for PT-adults who develop medical

589

risk. In the present study, PT-adults who had higher cumulative medical risk and higher

590

dietary quality had 6 mm/Hg lower systolic blood pressure, which could translate to at

591

least 14% less risk for stroke incidents and 9% reduction in coronary heart disease

592

mortality (Stamler, 1991).

593

AC C

EP

TE D

581

The study’s limitations and strengths need acknowledgement. Although part of a

28

ACCEPTED MANUSCRIPT

longitudinal study, the dietary behavior assessment only was incorporated into Wave VIII

595

(age 23 years) with less information about maternal intake during pregnancy [e.g. flavor

596

exposure in uterine environment (Mennella, Jagnow, & Beauchamp, 2001)] and early

597

nutrition (e.g. breastfeeding), which may link to future cardiovascular health (Singhal,

598

2006). The lack of dietary intake data during childhood limits the ability to draw

599

conclusions on whether unfavorable dietary behaviors observed among PT-adults were

600

influenced by factors beyond the prenatal environment (e.g. repeated exposure to

601

unhealthy foods at childhood (Kaikkonen et al., 2013; Lioret, McNaughton, Spence,

602

Crawford, & Campbell, 2013; Schwartz, Scholtens, Lalanne, Weenen, & Nicklaus,

603

2011). Nonetheless at age 23, the liking survey may capture more usual dietary behaviors

604

with less misreporting than food recalls/records or frequency surveys (Duffy et al., 2009;

605

Duffy et al., 2007; Sharafi, Faghri, et al., 2015; Sharafi, Peracchio, et al., 2015). The use

606

of SEM extends beyond traditional regression analysis, which specifies default models,

607

assume measurement occurs without error, and is somewhat inflexible. SEM allows

608

flexibility to incorporate measured variables and latent constructs, while explicitly

609

specifying measurement error to simultaneously determine direct and indirect

610

associations between term status, dietary quality, and CVD risk factors. The FT-adults in

611

the present study showed higher blood pressures than reference values from NHANES.

612

However, similar elevations in blood pressures were reported for Wave IV (24-32 year

613

olds) of National Longitudinal Study of Adolescent Health (Nguyen et al., 2011) in

614

comparison with NHANES 2007-2008. Differences in sample characteristics and setting

615

for blood pressure measurement could explain observed discrepancies with NHANES

616

values. The present study was limited in the ability to provide effects of birth weight on

AC C

EP

TE D

M AN U

SC

RI PT

594

29

ACCEPTED MANUSCRIPT

diet and CVD risk factor relationships in FT-adults (full-term infants with birth weight

618

<2500 g and preterm infants with birth weight >1850 g were not recruited into the study).

619

Finally, the homogeneity improved the internal validity but restricts the generalizability

620

of findings.

621

RI PT

617

In conclusion, this study investigated the simultaneous associations among

prematurity status, dietary behaviors and CVD risk factors through structural equation

623

modeling. Young adults who were born prematurely had less healthy dietary behaviors,

624

which in turn associated with elevated biochemical and anthropometric risk factors for

625

CVD. Although the potential impact of the diet-CVD risk factor association is

626

unknown—most preterm adults in large cohort studies have yet to reach age 55, where

627

86% of first CVD events have occurred (Wald, Simmonds, & Morris, 2011)—the greater

628

survival of former preterm infants into adulthood coupled with obesogenic environments

629

suggests early interventions for individuals born prematurely are warranted to develop

630

healthy eating preferences and habits for promotion of optimal growth and development

631

and for CVD prevention.

632

Highlights

635 636 637 638

M AN U

TE D

EP

634

•

Young adults born prematurely had elevated cardiovascular risk factors and less healthy dietary behaviors.

AC C

633

SC

622

•

Poor dietary quality and being born prematurely were separate and additive contributors of elevated systolic blood pressure.

•

Some of the preterm-cholesterol dyslipidemia relationship was explained (mediated) by poor dietary quality.

30

ACCEPTED MANUSCRIPT

•

640 641 642

Young adults born prematurely (vs. full-term) reported low dietary restraint and low sensation seeking, including food neophobia.

•

A healthier diet attenuated some of the CVD risk factors among adults born prematurely and with medical risk.

RI PT

639

643 Acknowledgement

645

Supported by USDA/Hatch and NIH National Institute of Nursing Research R01

646

NR003695.

SC

644

M AN U

647

Statement of authors’ contributions to manuscript.

649

M. S. carried out the analyses, drafted the initial manuscript and all revisions, and

650

approved the final manuscript as submitted; V. B. D developed the dietary assessment

651

protocol, coordinated and supervised the data analysis, critically reviewed the

652

manuscript, and approved the final manuscript as submitted; R. J. M., and S. B. W.

653

coordinated data collection, working with Dr. Sullivan to design preterm risk indexes,

654

reviewed and revised the manuscript, and approved the final manuscript as submitted; T.

655

B. H. supervised the data analysis, reviewed and revised the manuscript, and approved

656

the final manuscript as submitted. M. C. S. designed the data collection, coordinated and

657

supervised the study, developed the preterm risk indexes, critically reviewed the

658

manuscript, and approved the final manuscript as submitted. All authors have read and

659

approved this manuscript submission

AC C

EP

TE D

648

31

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

References Alley, T. R., & Potter, K. A. (2011). Food Neophobia and Sensation Seeking. In V. R. Preedy, R. R. Watson & C. R. Martin (Eds.), Handbook of Behavior, Food and Nutrition (pp. 707-724). New York: Springer Alley, T. R., Willet, K. A., & Muth, E. R. (2006). Motion sickness history, food neophobia, and sensation seeking. Percept Mot Skills, 102(3), 683-690. Allin, M., Rooney, M., Cuddy, M., Wyatt, J., Walshe, M., Rifkin, L., & Murray, R. (2006). Personality in young adults who are born preterm. Pediatrics, 117(2), 309-316. doi: 10.1542/peds.2005-0539 Allison, D., & Baskin, M. (2009). Handbook of Assessment Methods for Eating Behaviors and Weight-Related Problems (2nd ed.). Los Angeles, LA: Sage. Allison, D. B., Kalinsky, L. B., & Gorman, B. S. (1992). A comparison of the psychometric properties of three measures of dietary restraint. Psychological Assessment, 4(3), 391-398. Antonogeorgos, G., Panagiotakos, D. B., Pitsavos, C., Papageorgiou, C., Chrysohoou, C., Papadimitriou, G. N., & Stefanadis, C. (2012). Understanding the role of depression and anxiety on cardiovascular disease risk, using structural equation modeling; the mediating effect of the Mediterranean diet and physical activity: the ATTICA study. Ann Epidemiol, 22(9), 630-637. doi: 10.1016/j.annepidem.2012.06.103 Ayres, C., Agranonik, M., Portella, A. K., Filion, F., Johnston, C. C., & Silveira, P. P. (2012). Intrauterine growth restriction and the fetal programming of the hedonic response to sweet taste in newborn infants. Int J Pediatr, 2012, 657379. doi: 10.1155/2012/657379 Barbieri, M. A., Portella, A. K., Silveira, P. P., Bettiol, H., Agranonik, M., Silva, A. A., & Goldani, M. Z. (2009). Severe intrauterine growth restriction is associated with higher spontaneous carbohydrate intake in young women. Pediatr Res, 65(2), 215-220. doi: 10.1203/PDR.0b013e31818d6850 Barker, D. J., Eriksson, J. G., Forsen, T., & Osmond, C. (2002). Fetal origins of adult disease: strength of effects and biological basis. Int J Epidemiol, 31(6), 12351239. Barker, D. J., Osmond, C., Forsen, T. J., Kajantie, E., & Eriksson, J. G. (2007). Maternal and social origins of hypertension. Hypertension, 50(3), 565-571. doi: HYPERTENSIONAHA.107.091512 [pii]10.1161/HYPERTENSIONAHA.107.091512 Bartoshuk, L. M., Duffy, V. B., Hayes, J. E., Moskowitz, H. R., & Snyder, D. J. (2006). Psychophysics of sweet and fat perception in obesity: problems, solutions and new perspectives. Philos Trans R Soc Lond B Biol Sci, 361(1471), 1137-1148. doi: 10.1098/rstb.2006.1853 Bartoshuk, L. M., & Klee, H. J. (2013). Better fruits and vegetables through sensory analysis. Curr Biol, 23(9), R374-378. doi: 10.1016/j.cub.2013.03.038 Behrman, R. E., & Butler, A. S. (2007). Preterm Birth: Causes, Consequences, and Prevention. Washington, DC: National Academies Press. Bell, M. J. (1978). Neonatal necrotizing enterocolitis. N Engl J Med, 298(5), 281-282. Blissett, J., & Fogel, A. (2013). Intrinsic and extrinsic influences on children's acceptance of new foods. Physiol Behav, 121, 89-95. doi: 10.1016/j.physbeh.2013.02.013

AC C

660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705

32

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

Boerner, L. M., Spillane, N. S., Anderson, K. G., & Smith, G. T. (2004). Similarities and differences between women and men on eating disorder risk factors and symptom measures. Eat Behav, 5(3), 209-222. doi: 10.1016/j.eatbeh.2004.01.011 Bonamy, A. K., Kallen, K., & Norman, M. (2012). High blood pressure in 2.5-year-old children born extremely preterm. Pediatrics, 129(5), e1199-1204. doi: 10.1542/peds.2011-3177 Bradley, R. H., Whiteside, L., Mundfrom, D. J., Casey, P. H., Kelleher, K. J., & Pope, S. K. (1994). Early indications of resilience and their relation to experiences in the home environments of low birthweight, premature children living in poverty. Child Dev, 65(2 Spec No), 346-360. Byrnes, N. K., & Hayes, J. E. (2013). Personality factors predict spicy food liking and intake. Food Qual Prefer, 28(1), 213-221. doi: 10.1016/j.foodqual.2012.09.008 Caldwell, B. M., & Bradley, R. H. (2001). Home Inventory Administration Manual (Third ed.). Little Rock, AR: Univerisity of Arkansas at Little Rock. Centers for Disease Control and Prevention (CDC), & National Center for Health Statistics (NCHS). National Health and Nutrition Examination Survey Data 20112012. Retrieved June 16th, 2014, from http://wwwn.cdc.gov/nchs/nhanes/search/nhanes11_12.aspx Cerro, N., Zeunert, S., Simmer, K. N., & Daniels, L. A. (2002). Eating behaviour of children 1.5-3.5 years born preterm: parents' perceptions. J Paediatr Child Health, 38(1), 72-78. Chiva-Blanch, G., Urpi-Sarda, M., Ros, E., Valderas-Martinez, P., Casas, R., Arranz, S., Guillén, M., Lamuela-Raventós, R.M., Llorach, R., Andres-Lacueva, C., & Estruch R. (2013). Effects of red wine polyphenols and alcohol on glucose metabolism and the lipid profile: a randomized clinical trial. Clin Nutr,, 32(2):200-6. doi: 10.1016/j.clnu.2012.08.022. Clemente-Postigo, M., Queipo-Ortuño, M. I., Boto-Ordoñez, M., Coin-Aragüez, L., Roca-Rodriguez, M. M., Delgado-Lista, J., Cardona, F., Andres-Lacueva, C., & Tinahones, F.J. (2013). Effect of acute and chronic red wine consumption on lipopolysaccharide concentrations. Am J Clin Nutr. 97(5):1053-61. doi: 10.3945/ajcn.112.051128. Epub 2013 Apr 10. Contento, I. R., Zybert, P., & Williams, S. S. (2005). Relationship of cognitive restraint of eating and disinhibition to the quality of food choices of Latina women and their young children. Prev Med, 40(3), 326-336. doi: 10.1016/j.ypmed.2004.06.008 Cooke, R. W. (2004). Health, lifestyle, and quality of life for young adults born very preterm. Arch Dis Child, 89(3), 201-206. de Jong, F., Monuteaux, M. C., van Elburg, R. M., Gillman, M. W., & Belfort, M. B. (2012). Systematic review and meta-analysis of preterm birth and later systolic blood pressure. Hypertension, 59(2), 226-234. doi: HYPERTENSIONAHA.111.181784 [pii]10.1161/HYPERTENSIONAHA.111.181784 Duffy, V. B., Hayes, J. E., Sullivan, B. S., & Faghri, P. (2009). Surveying food and beverage liking: a tool for epidemiological studies to connect chemosensation with health outcomes. Ann N Y Acad Sci, 1170, 558-568. doi: NYAS04593 [pii]10.1111/j.1749-6632.2009.04593.x

AC C

706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751

33

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

Duffy, V. B., Lanier, S. A., Hutchins, H. L., Pescatello, L. S., Johnson, M. K., & Bartoshuk, L. M. (2007). Food preference questionnaire as a screening tool for assessing dietary risk of cardiovascular disease within health risk appraisals. J Am Diet Assoc, 107(2), 237-245. doi: S0002-8223(06)02493-X [pii]10.1016/j.jada.2006.11.005 Eckel, R. H., Jakicic, J. M., Ard, J. D., Miller, N. H., Hubbard, V. S., Nonas, C. A., . . . Yanovski, S. Z. (2013). 2013 AHA/ACC Guideline on Lifestyle Management to Reduce Cardiovascular Risk: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol. doi: 10.1016/j.jacc.2013.11.003 Erinosho, T. O., Pinard, C. A., Nebeling, L. C., Moser, R. P., Shaikh, A. R., Resnicow, K., . . . Yaroch, A. L. (2015). Development and Implementation of the National Cancer Institute's Food Attitudes and Behaviors Survey to Assess Correlates of Fruit and Vegetable Intake in Adults. PLoS One, 10(2), e0115017. doi: 10.1371/journal.pone.0115017 Finken, M. J., Keijzer-Veen, M. G., Dekker, F. W., Frolich, M., Hille, E. T., Romijn, J. A., . . . Dutch, P.-C. S. G. (2006). Preterm birth and later insulin resistance: effects of birth weight and postnatal growth in a population based longitudinal study from birth into adult life. Diabetologia, 49(3), 478-485. doi: 10.1007/s00125-005-0118-y Friedewald, W. T., Levy, R. I., & Fredrickson, D. S. (1972). Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem, 18(6), 499-502. Goulet, J., Provencher, V., Piche, M. E., Lapointe, A., John Weisnagel, S., Nadeau, A., . . . Lemieux, S. (2008). Relationship between eating behaviours and food and drink consumption in healthy postmenopausal women in a real-life context. Br J Nutr, 100(4), 910-917. doi: 10.1017/S0007114508925459 Hack, M., Schluchter, M., Forrest, C. B., Taylor, H. G., Drotar, D., Holmbeck, G., . . . Andreias, L. (2012). Self-reported adolescent health status of extremely low birth weight children born 1992-1995. Pediatrics, 130(1), 46-53. doi: 10.1542/peds.2011-3402 Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical Mediation Analysis in the New Millennium. Communication Monographs, 76(4), 408-420. Hayes, J. E., & Duffy, V. B. (2008). Oral sensory phenotype identifies level of sugar and fat required for maximal liking. Physiol Behav, 95(1-2), 77-87. doi: 10.1016/j.physbeh.2008.04.023 Hedges, L. V. (1981). Distribution Theory for Glass's Estimator of Effect Size and Related Estimators. Journal of Education Statistics, 6(2), 107-128. Herman, C. P., & Polivy, J. (1980). Restrained eating (In A.J. Stunkard (Ed.) ed.). Philadelphia: Saunders. Hobel, C. J., Hyvarinen, M. A., Okada, D. M., & Oh, W. (1973). Prenatal and intrapartum high-risk screening. I. Prediction of the high-rish neonate. Am J Obstet Gynecol, 117(1), 1-9. Hofman, P. L., Regan, F., Jackson, W. E., Jefferies, C., Knight, D. B., Robinson, E. M., & Cutfield, W. S. (2004). Premature birth and later insulin resistance. N Engl J Med, 351(21), 2179-2186. doi: 10.1056/NEJMoa042275

AC C

752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797

34

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

Hollingshead, A. B. (1975). Four factor index of social status. New Haven, Conn.: [Yale University, Dept. of Sociology]. Hutchins-Wiese, H. L., Duffy, V. B., Watkins, B. A., Li, Y., & Kenny, A. M. (2013). A novel tool for dietary assessment of omega-3 polyunsaturated fatty acids (PUFA) status in healthy postmenopausal women (abstract). FASEB J, 27, 616.611. Jessor, R., Costa, F. M., & Turbin, M. S. (2003). Survey of Personal and Social Development. Retrieved August 29th, 2015, from http://www.colorado.edu/ibs/jessor/questionnaires.html Johnson, F., Pratt, M., & Wardle, J. (2012). Dietary restraint and self-regulation in eating behavior. Int J Obes (Lond), 36(5), 665-674. doi: 10.1038/ijo.2011.156 Kaikkonen, J. E., Mikkila, V., Magnussen, C. G., Juonala, M., Viikari, J. S., & Raitakari, O. T. (2013). Does childhood nutrition influence adult cardiovascular disease risk?--insights from the Young Finns Study. Ann Med, 45(2), 120-128. doi: 10.3109/07853890.2012.671537 Kajantie, E., & Hovi, P. (2014). Is very preterm birth a risk factor for adult cardiometabolic disease? Semin Fetal Neonatal Med, 19(2), 112-117. doi: 10.1016/j.siny.2013.11.006 Kaseva, N., Wehkalampi, K., Hemio, K., Hovi, P., Jarvenpaa, A. L., Andersson, S., . . . Kajantie, E. (2013). Diet and nutrient intake in young adults born preterm at very low birth weight. J Pediatr, 163(1), 43-48. doi: S0022-3476(12)01565-X [pii]10.1016/j.jpeds.2012.12.076 Kourlaba, G., & Panagiotakos, D. B. (2009). Dietary quality indices and human health: a review. Maturitas, 62(1), 1-8. doi: S0378-5122(08)00357-5 [pii]10.1016/j.maturitas.2008.11.021 Labayen, I., Moreno, L. A., Blay, M. G., Blay, V. A., Mesana, M. I., Gonzalez-Gross, M., . . . Bueno, M. (2006). Early programming of body composition and fat distribution in adolescents. J Nutr, 136(1), 147-152. doi: 136/1/147 [pii] Li, Y., Ley, S. H., Tobias, D. K., Chiuve, S. E., VanderWeele, T. J., Rich-Edwards, J. W., . . . Qi, L. (2015). Birth weight and later life adherence to unhealthy lifestyles in predicting type 2 diabetes: prospective cohort study. BMJ, 351, h3672. doi: 10.1136/bmj.h3672 Lioret, S., McNaughton, S. A., Spence, A. C., Crawford, D., & Campbell, K. J. (2013). Tracking of dietary intakes in early childhood: the Melbourne InFANT Program. Eur J Clin Nutr, 67(3), 275-281. doi: 10.1038/ejcn.2012.218 Lubchenco, L. O., Hansman, C., & Boyd, E. (1966). Intrauterine growth in length and head circumference as estimated from live births at gestational ages from 26 to 42 weeks. Pediatrics, 37(3), 403-408. Mathai, S., Derraik, J. G., Cutfield, W. S., Dalziel, S. R., Harding, J. E., Biggs, J., . . . Hofman, P. L. (2013). Increased adiposity in adults born preterm and their children. PLoS One, 8(11), e81840. doi: 10.1371/journal.pone.0081840 McGrath, M. M. (1989a). Maternal Self-Esteem Questionnaire. . Kingston, RI: University of Rhode Island. McGrath, M. M. (1989b). Perception of Child Health. Kingston, RI: University of Rhode Island.

AC C

798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841

35

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

McGrath, M. M., & Sullivan, M. (2003). Testing proximal and distal protective processes in preterm high-risk children. Issues Compr Pediatr Nurs, 26(2), 59-76. doi: DOI: 10.1080/01460860390197835 Mennella, J. A., Jagnow, C. P., & Beauchamp, G. K. (2001). Prenatal and postnatal flavor learning by human infants. Pediatrics, 107(6), E88. Mennella, J. A., Pepino, M. Y., & Reed, D. R. (2005). Genetic and environmental determinants of bitter perception and sweet preferences. Pediatrics, 115(2), e216222. doi: 10.1542/peds.2004-1582 Migraine, A., Nicklaus, S., Parnet, P., Lange, C., Monnery-Patris, S., Des Robert, C., . . . Roze, J. C. (2013). Effect of preterm birth and birth weight on eating behavior at 2 y of age. Am J Clin Nutr, 97(6), 1270-1277. doi: ajcn.112.051151 [pii]10.3945/ajcn.112.051151 Moreira, P., de Almeida, M. D., & Sampaio, D. (2005). Cognitive restraint is associated with higher intake of vegetables in a sample of university students. Eat Behav, 6(3), 229-237. doi: 10.1016/j.eatbeh.2005.01.005 Nguyen, Q. C., Tabor, J. W., Entzel, P. P., Lau, Y., Suchindran, C., Hussey, J. M., . . . Whitsel, E. A. (2011). Discordance in national estimates of hypertension among young adults. Epidemiology, 22(4), 532-541. doi: 10.1097/EDE.0b013e31821c79d2 Niswander, K. R., & Gordon, M. (1972). The collaborative perinatal study: The women and their pregnancies: the collaborative prenatal study of the National Intitute of Neurologic Disease and Stroke. Philadelphia: W.B. Saunders. Pallister, T., Sharafi, M., Lachance, G., Pirastu, N., Mohney, R. P., MacGregor, A., . . . Menni, C. (2015). Food Preference Patterns in a UK Twin Cohort. Twin Research and Human Genetics, 1-13. doi: http://dx.doi.org/10.1017/thg.2015.69 Panz, V. R., Raal, F. J., Paiker, J., Immelman, R., & Miles, H. (2005). Performance of the CardioChek™ PA and Cholestech LDX® point-of-care analysers compared to clinical diagnostic laboratory methods for the measurement of lipids. Cardiovasc J South Afr. Papile, L. A., Burstein, J., Burstein, R., & Koffler, H. (1978). Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. J Pediatr, 92(4), 529-534. Parkinson, J. R., Hyde, M. J., Gale, C., Santhakumaran, S., & Modi, N. (2013). Preterm birth and the metabolic syndrome in adult life: a systematic review and metaanalysis. Pediatrics, 131(4), e1240-1263. doi: peds.2012-2177 [pii]10.1542/peds.2012-2177 Peracchio, H. L., Henebery, K. E., Sharafi, M., Hayes, J. E., & Duffy, V. B. (2012). Otitis media exposure associates with dietary preference and adiposity: a communitybased observational study of at-risk preschoolers. Physiol Behav, 106(2), 264271. doi: S0031-9384(12)00062-5 [pii]10.1016/j.physbeh.2012.01.021 Pliner, P., & Melo, N. (1997). Food neophobia in humans: effects of manipulated arousal and individual differences in sensation seeking. Physiol Behav, 61(2), 331-335. Poplawska, K., Dudek, K., Koziarz, M., Cieniawski, D., Drożdż, T., Śmiałek, S., . . . Kwinta, P. (2012). Prematurity-Related Hypertension in Children and Adolescents. International Journal of Pediatrics, 2012. doi:10.1155/2012/537936

AC C

842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887

36

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

Prechtel, H., & Beitema, D. (1967). The neurological examination of the full-term newborn infant. London, England: Heineman. Pridham, K. F., Brown, R., Sondel, S., Clark, R., & Green, C. (2001). Effects of biologic and experiential conditions on the pattern of growth in weight of premature and full-term infants. Res Nurs Health, 24(4), 283-297. Rees, K., Hartley, L., Flowers, N., Clarke, A., Hooper, L., Thorogood, M., & Stranges, S. (2013). 'Mediterranean' dietary pattern for the primary prevention of cardiovascular disease. Cochrane Database Syst Rev, 12;8:CD009825. doi: 10.1002/14651858. CD009825.pub2. Ritchey, P. N., Frank, R. A., Hursti, U. K., & Tuorila, H. (2003). Validation and crossnational comparison of the food neophobia scale (FNS) using confirmatory factor analysis. Appetite, 40(2), 163-173. doi: S0195666302001344 [pii] Roberts, G., Burnett, A. C., Lee, K. J., Cheong, J., Wood, S. J., Anderson, P. J., . . . Victorian Infant Collaborative Study, G. (2013). Quality of life at age 18 years after extremely preterm birth in the post-surfactant era. J Pediatr, 163(4), 10081013 e1001. doi: 10.1016/j.jpeds.2013.05.048 Rogers, L. K., & Velten, M. (2011). Maternal inflammation, growth retardation, and preterm birth: insights into adult cardiovascular disease. Life Sci, 89(13-14), 417421. doi: S0024-3205(11)00363-8 [pii]10.1016/j.lfs.2011.07.017 Samara, M., Johnson, S., Lamberts, K., Marlow, N., & Wolke, D. (2010). Eating problems at age 6 years in a whole population sample of extremely preterm children. Dev Med Child Neurol, 52(2), e16-22. doi: DMCN3512 [pii]10.1111/j.1469-8749.2009.03512.x Savage, J. S., Fisher, J. O., & Birch, L. L. (2007). Parental influence on eating behavior: conception to adolescence. J Law Med Ethics, 35(1), 22-34. doi: 10.1111/j.1748720X.2007.00111.x Scarmo, S., Henebery, K., Peracchio, H., Cartmel, B., Lin, H., Ermakov, I. V., . . . Mayne, S. T. (2012). Skin carotenoid status measured by resonance Raman spectroscopy as a biomarker of fruit and vegetable intake in preschool children. Eur J Clin Nutr, 66(5), 555-560. doi: 10.1038/ejcn.2012.31 Schwartz, C., Scholtens, P. A., Lalanne, A., Weenen, H., & Nicklaus, S. (2011). Development of healthy eating habits early in life. Review of recent evidence and selected guidelines. Appetite, 57(3), 796-807. doi: 10.1016/j.appet.2011.05.316 Sharafi, M., Faghri, P. D., Huedo-Medina, T. B., Minski, K., & Duffy, V. B. (2015). A Simple Liking Survey Captures Behaviors Associated with Weight Loss in a Worksite Program among Women at Risk of Type 2 DiabetesFASEB J, 29, 905.910. Sharafi, M., Hayes, J. E., & Duffy, V. B. (2013). Masking Vegetable Bitterness to Improve Palatability Depends on Vegetable Type and Taste Phenotype. Chemosens Percept, 6(1), 8-19. doi: 10.1007/s12078-012-9137-5 Sharafi, M., Peracchio, H., Scarmo, S., Huedo-Medina, T. B., Mayne, S. T., Cartmel, B., & Duffy, V. B. (2015). Preschool-Adapted Liking Survey (PALS): A Brief and Valid Method to Assess Dietary Quality of Preschoolers. Child Obes. doi: 10.1089/chi.2015.0037

AC C

888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931

37

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

Silveira, P. P., Agranonik, M., Faras, H., Portella, A. K., Meaney, M. J., & Levitan, R. D. (2012). Preliminary evidence for an impulsivity-based thrifty eating phenotype. Pediatr Res, 71(3), 293-298. doi: pr201139 [pii]10.1038/pr.2011.39 Singhal, A. (2006). Early nutrition and long-term cardiovascular health. Nutr Rev, 64(5 Pt 2), S44-49; discussion S72-91. Sipola-Leppanen, M., Vaarasmaki, M., Tikanmaki, M., Hovi, P., Miettola, S., Ruokonen, A., . . . Kajantie, E. (2014). Cardiovascular risk factors in adolescents born preterm. Pediatrics, 134(4), e1072-1081. doi: 10.1542/peds.2013-4186 Sproesser, G., Strohbach, S., Schupp, H., & Renner, B. (2011). Candy or apple? How self-control resources and motives impact dietary healthiness in women. Appetite, 56(3), 784-787. doi: 10.1016/j.appet.2011.01.028 Stamler, R. (1991). Implications of the INTERSALT study. Hypertension, 17(1 Suppl), I16-20. Stein, L. J., Cowart, B. J., & Beauchamp, G. K. (2006). Salty taste acceptance by infants and young children is related to birth weight: longitudinal analysis of infants within the normal birth weight range. Eur J Clin Nutr, 60(2), 272-279. doi: 10.1038/sj.ejcn.1602312 Sullivan, M. C. (2008). Risk and protection in trajectories of preterm Infants: Birth to Adulthood Bethesda, MD: National Institutes of Health, National Institute of Nursing Research. Sullivan, M. C., Hawes, K., Winchester, S. B., & Miller, R. J. (2008). Developmental origins theory from prematurity to adult disease. J Obstet Gynecol Neonatal Nurs, 37(2), 158-164. doi: JOGN216 [pii]10.1111/j.1552-6909.2008.00216.x Terasaki, M., & Imada, S. (1988). Sensation seeking and food preferences. Personality and Individual Differences, 9(1), 87-93. Thomas, E. L., Parkinson, J. R., Hyde, M. J., Yap, I. K., Holmes, E., Dore, C. J., . . . Modi, N. (2011). Aberrant adiposity and ectopic lipid deposition characterize the adult phenotype of the preterm infant. Pediatr Res, 70(5), 507-512. doi: 10.1038/pr.2011.73210.1203/PDR.0b013e31822d7860 Thompson, F. E., & Subar, A. F. (2013). Dietary assessment metholology. . In A. M. Coulston, C. J. Boushey & M. G. Ferruzzi (Eds.), Nutrition in the Prevention and Treatment of Disease (3rd ed.). San Diego, CA: Academic Press. Tomiyama, A. J., Moskovich, A., Haltom, K. B., Ju, T., & Mann, T. (2009). Consumption after a diet violation: disinhibition or compensation? Psychol Sci, 20(10), 1275-1281. doi: 10.1111/j.1467-9280.2009.02436.x Tuorila, H., Huotilainen, A., Lähteenmäki, L., Ollila, S., Tuomi-Nurmi, S., & Urala, N. (2008). Comparison of affective rating scales and their relationship to variables reflecting food consumption. Food Qual Prefer, 19(1), 51-61. Ullman, J. B. (2001). Structural equation modeling. In B. G. Tabachnick & L. S. Fidell (Eds.), Using Multivariate Statistics. Needham Heights, MA: Allyn & Bacon. Wald, N. J., Simmonds, M., & Morris, J. K. (2011). Screening for Future Cardiovascular Disease Using Age Alone Compared with Multiple Risk Factors and Age. PLoS One, 6(5), e18742. doi: 10.1371/journal.pone.0018742 Wardle, J., & Beales, S. (1987). Restraint and food intake: an experimental study of eating patterns in the laboratory and in normal life. Behav Res Ther, 25(3), 179185.

AC C

932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977

38

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

Wilkie, L. M., & Capaldi Phillips, E. D. (2014). Heterogeneous binary interactions of taste primaries: Perceptual outcomes, physiology, and future directions. Neurosci Biobehav Rev, 47C, 70-86. doi: 10.1016/j.neubiorev.2014.07.015 Winchester, S. B., Sullivan, M. C., & Msall, M. E. (2014). Executive Function in Adult Survivors of Prematurity In K. P. Bennett (Ed.), Executive Functioning: Role in Early Learning Processes, Impairments in Neurological Disorders and Impact of Cognitive Behavior Therapy (CBT) (pp. 83-114): Nova Science.

AC C

978 979 980 981 982 983 984 985 986 987

39