Evidence Table 5. Studies associating behavioral problems with Failure to Thrive patients compared to healthy control subjects in developed countries Part II

Author, YearCorrelates & correlate measuresMethod / InstrumentAssociations found
Pollitt, 1976 76109110 FTT CtrlReported or observed feeding, history of other behaviors, direct measuresFTT were reported to have significantly more feeding behavior problems than controls in difficulty in feeding (χ2 = 5.97, p < .02); eating patterns (χ2 = 8.74, p < .03); supported by caloric intake measures (t = 2.06, p = .03). FTT had a trend for exhibiting autoerotic & self-harm behavior problems; analyzed for sum any of these atypical behaviors, the FTT exhibited significant higher incidence (χ2 = 8.19, p < .01)
Feeding difficulty102
Meal pattern-often skipped/skimpy166
Caloric intake1400(300)1700(500)
Response to food -
correlation to wt %NDND
Eating disorder -
FTT 4 exhibiting eating & drinking nonfood items, polydipsia, gorging, & hiding food
Ctrl none reported
Self harm20
Sum children with abnormal behavior101
Mitchell, 1980 80166667Behavior problems FTT Ctrl Behavior problems questionnaireNS but FTT group had higher incidence of behavioral problems.
Polan, 1991 92098539FTT status on positive & negative affect in feeding in 4 channels of emotion - facial display pattern, vocalization, gesture, body posture & movementKiddie Affect Inventory and Assessment of 11 affects scored independentlyFTT had significantly decreased positive affect during feeding & nonfeeding (t=-2.92, p=.04; t=-3.151, p = .003, respectively) Also FTT had significantly increased negative affect during feeding (t=2.567, p=.01).
Drotar, 1992 92372721 FTT Ctrl California Child Q-set Lock box - Rectangular container w/10 locked compartments Child Behavior ChecklistFTT children have poor problem-solving, behavior problems, & “deficiencies in personality development” versus controls. The ego development score was lower, ego resiliency score was significantly lower (p<.05). FTT behavior organization & problem scores were significant lower (p<.001, p<.05, respectively)
Personality development
• ego control376.67(32.28)413.42(30.73)
• ego resiliency376.06(41.21)397.37(41.45)
Behavioral organization.95(1.92)8.45(12.08)
Behavioral problems58.5(10.7)53.2(12.7)
Hutcheson, 1993 94015754Child temperament FTT1* FTT2† Ctrl1* Ctrl 2† Infant CharacteristicsThe FTT cases rated as more difficult with significant finding on effect of infant age on maternal report of infant level of difficulty (F=4.21, p = 0.04). Though no age or group differences found on perceived life stresses, parenting stresses, informal support, and negative affectivity, the FTT infants had fewer life events than FTT toddlers, whereas the reverse was true for the comparison group. The maternal effect was less positive toward FTT toddlers than FTT infants. Controlling for maternal education, wt for age, & ht for age - significant group by age interaction for maternal factors (F=4.98, p = 0.02) due to maternal affective tone (F=9.33, p = 0.005)
* Infants: 8–13.4 mo
† Toddlers: 13.5–24 mo
Kelleher, 1993 93234174Behavioral disorder FTT Ctrl Bates Temperament scale CBCLSample of LBW or premature infants followed had lower HOME scores at 12 months, NS. There was a trend towards more behavioral problems for the FTT as reported by parents.
Skuse, 1994 94253258Behavior Assessment at 15 moTRIBNo difference between groups for expression of positive affect in relation to task directed behavior or task persistence
Puckering, 1995 95378341Behavior score FTT Ctrl Behavior ScreeningThere was no difference in BSQ scores for the two groups (p=.2). MANOVA performed on calculated mean GCI & BSQ scores vs. unconditional & conditional interactions grouped by high or low (above & below overall mean value for interaction) - results indicate that parenting style does not account for GCI or BSQ differences between FTT & Ctrl.
8.7(4.4)6.9(3.5) Questionnaire (BSQ)
Corbett, 1996 97113595Behavioral disorder FTT Ctrl CBCLThere was a trend towards more behavioral problems for the FTT as reported by teachers (NS, p = .296)
Parental reporting3334
Teacher reporting2314
Wilensky, 1996 97022837Feeding problems FTT Ctrl Maternal questionnaireFTT children had significantly more feeding problems than the controls (p < .05)
• showing hunger53.4%46.2%
• turns head from food59.7%20.7%
• shows pleasure at meals76.7%94.3%
• nervous at meals23.2%5.6%
• eats variety48.8%69.8%
• spits out food65.1%33.9%
Chatoor, 1998 99026462 Gp1 Gp2 Ctrl Ainsworth & Cassidy & Marvin attachment classificationThere was significant relationship between diagnostic group and frequency of insecure attachments (χ2 = 8.0, p < .05). Infant anorexia group exhibited more frequent insecure relationships compared to picky eater or healthy eater groups (χ2 = 6.7, p < .01; χ2 = 3.9, p < .05). There was main effect for group for Attachment security scale (F = 5.8, p < .01) with the infantile anorexia group as least secure. The closer to ideal weight was correlated to increased security (r = 0.31, p < .01).
Attachment classificationStrange Situation Security Scale / 9 point Likert
Attachment security
scale rating4.9(2.1)6.1(1.3)6.0(1.5)
Percent ideal weight
Wright, 2000 20161504Behavior & temperamentParental reporting / Likert scaleFTT had higher maternal reporting of infants as undemanding & shy in temperament (p = .005 & p = .002, respectively). Higher maternal reporting of infancy feeding problems, variable eating pattern.
FTT Ctrl
Kerr, 2000 20277217 Gp1 Gp2 Ctrl1 Ctrl2 There is main effect from risk status (FTT, FTT & maltreatment, maltreatment alone) as being associated with worse scores in cognitive & adaptive performance (F = 3.527, p < .01), school & home behavior (F = 2.986, p < .01; F = 2.453, p < .05, respectively). There was a trend for worse scores in all areas for FTT vs. no risks factors or maltreatment alone.
Adaptive functioning at school15.85 (7.2)13.88 (6.9)16.92 (6.9)17.85 (9.7)Teacher Report Form (↑ numbers = better score)
School behavior31.45 (27.1)53.53 (35.7)29.18 (25.0)28.87 (29.1)Teacher Report Form (↓ numbers = better score)
Home behavior28.11 (22.8)34.85 (21.2)29.00 (20.5)26.67 (18.4)Child Behavior Checklist (↓ numbers = better score)
Steward, 2001 21291243 FTT Ctrl PCERA subscaleFTT group scored significantly lower on the communication subscale (p < 0.02); during interactions with the mother had less visual contact, more gaze aversion & vocalized less than controls, had difficult to read cues. They also scored lower than controls on the mood scale, were more irritable & apathetic (p < 0.05).
Mood effect3.654.26

From: Evidence Tables

Cover of Criteria for Determining Disability in Infants and Children: Failure to Thrive
Criteria for Determining Disability in Infants and Children: Failure to Thrive.
Evidence Reports/Technology Assessments, No. 72.
Perrin EC, Frank DA, Cole CH, et al.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.