Table 1.
Ref. | Author, year; Study methodolog y strength score2 |
Country | Sample size | Child age | Other sample characteristics | Dietary assessment | Whether child intake was self- reported | Main findings | Main strengths and limitations, and additional important findings and notes |
---|---|---|---|---|---|---|---|---|---|
[3] | Laskarzewski et al., 1980 Score=12 |
U.S. | 294 children 294 parents |
6–19 y | 4/5 (n=234) of the sample were white families, and 1/5 (n=60) were black families | One 24-hr dietary recall. Data were obtained on the same day for 1/3 of the families (n=99), on different days within a week for 2/3 families (n=195). | Yes |
Spearman correlation: All parent-child pairs (n=294) CHO: 0.28* saturated fat: 0.15* PUFA: 0.19* calories: 0.24* cholesterol: 0.004 One parent-one child (n=198) CHO: 0.30* saturated fat: 0.15* PUFA: 0.22* calories: 0.26* cholesterol: −0.004 Note: Nutrient intake was expressed as per kg body weight. Pearson correlations (not presented) were similar to the Spearman correlations. |
Strengths: Sampled both white and black US children. Limitations: Only used one 24- hr recall. Additional findings: ANOVA adjusted for covariates shows that 23–97% of the variation in children’s intakes was accounted for by those of parents’ (23% for cholesterol among all parent-child pairs, and 97% for CHO of black fathers over age 40 and their children). The proportion for black parent-child pairs were higher than that for white for CHO, saturated fat, and calories. |
[5] | Perusse et al., 1988 Score=17 |
Canada | 885 children 712 parents From 375 families |
mean age of 14.5 for males and 14.8 for females | French descent Quebec city area | 3-day dietary record for children and parents | Yes |
Intraclass correlation: Parent-child: Energy (kcal): 0.25* Energy (kcal/kg/d): 0.26* CHO (g): r=0.26* Fat (g): r=0.27* Protein (g): r=0.26* CHO (% energy): r=0.29* |
Strengths: Used the BETA model to separate phenotypic variance of intake measurements into biological and cultural. |
[4] | Patterson, et al., 1988 Score=14 |
U.S. | 132 Anglo (A) children and 154 parents 169 Mexican- American (MA) children and 143 parents |
9–15 y | Minority sample | one 24-hr recall, 3-day food record, FFQ | Yes |
Correlation for food frequency: Father-younger child: A: 0.56* MA: 0.17 Father-older child: A: 0.29 MA: 0.34* Mother-younger child: A: 0.36* MA: 0.29* Mother-older child: A: 0.42* MA: 0.42* |
Strengths: a) BMI was adjusted; b) multiple dietary assessment approaches. Limitations: Correlations based on 24-hour recall and 3-day food record were not reported. |
[6] | Oliveria et al., 1992 Score=14 |
U.S. | 91 children; 87 mothers and 83 fathers | 3–5 y | white middle class families | 4 sets of 3-day food record at 3- month intervals | No (complete d by mother with help from child or adult personnel in day care) |
Pearson correlation: parent-child (n=83): Energy: 0.20 Protein 0.37* CHO: 0.31* Total fat: 0.32* Saturated fat: 0.47* MUFA: 0.29* PUFA: 0.27* Cholesterol: 0.41* Calcium: 0.30* Father-child (n=83): Energy: 0.16 Protein 0.34* CHO: 0.18 Total fat: 0.15 Saturated fat: 0.34* MUFA: 0.13 PUFA: 0.10 Cholesterol: 0.34* Calcium: 0.21 Father-son (n=50): Energy: 0.19 Protein 0.25 CHO: 0.25 Total fat: 0.15 Saturated fat: 0.40* MUFA: 0.15 PUFA: 0.06 Cholesterol: 0.45* Calcium: 0.15 Father-daughter (n=33): Energy: −0.03 Protein 0.25 CHO: 0.03 Total fat: 0.12 Saturated fat: 0.22 MUFA: 0.09 PUFA: 0.18 Cholesterol: 0.10 Calcium: 0.33 Mother-child (n=87): Energy: 0.17 Protein 0.29* CHO: 0.37* Total fat: 0.46* Saturated fat: 0.48* MUFA: 0.43* PUFA: 0.33* Cholesterol: 0.37* Calcium: 0.29* Mother-son (n=54): Energy: 0.03 Protein 0.14 CHO: 0.28* Total fat: 0.40* Saturated fat: 0.43* MUFA: 0.43* PUFA: 0.31* Cholesterol: 0.19 Calcium: 0.22* Mother-daughter (n=33): Energy: 0.22 Protein 0.42* CHO: 0.47* Total fat: 0.49* Saturated fat: 0.51* MUFA: 0.42* PUFA: 0.35* Cholesterol: 0.56* Calcium: 0.30 |
Strengths: a) 4 sets of 3-day food records were used to catch seasonal variations in intakes; b) the correlations (except for total energy intake) were adjusted for total energy intake and parents’ age; c) compared correlations stratified by number of meals eaten at home by parents Additional findings: Correlations were much higher when parents had more meals (fathers>=15 vs less, mothers >=18 vs less) at home, eg, for energy: Father-child: 0.29 vs 0.07 Mother-child: 0.35 vs 0.11 |
[7] | Rossow and Rise, 1994 Score=12 |
Norway | 337 children, 324 mothers, 323 fathers | 16–20 y | both urban and rural areas | short questions on fat intake (milk and butter/margarine consumption) | Yes |
Spearman correlation: Mother-child: 0.47 Father-child: 0.42 Children were more likely to have a low-fat diet if their parents did so: mother’s low: OR: 5.00 (2.97, 8.41) father’s low: OR: 4.44 (2.56, 7.85) |
Strengths: Nationally representative sample. Limitations: Weak dietary assessment. |
[8] | Stafleu et al., 1994 Score=14 |
the Netherlands | 97 young adult women and 200 mothers and grandmother s | 20–30 y | Sampled from low-income inhabitants in Helmond where death rates from ischaemic heart disease were higher than the national average | validated 104- item FFQ (Feunekes et al, 1993) | Yes |
Pearson correlation: Younger-middle generation: Energy (kJ): 0.22* Total fat (% energy): 0.19 Saturated fat (% energy): 0.26* MUFA: 0.22* PUFA: 0.20 Cholesterol (mg): 0.21* Middle-older generation: Energy (kJ): 0.08 Total fat (% energy): −0.02 Saturated fat (% energy): 0.09 MUFA: 0.04 PUFA: 0.10 Cholesterol (mg): 0.29* Younger-older generation: Energy (kJ): 0.14 Total fat (% energy): 0.12 Saturated fat (% energy): 0.05 MUFA: −0..00 PUFA: 0.33* Cholesterol (mg): 0.05 Spearman correlation: Younger-middle generation: Oils and fats: 0.19 VF: 0.12 Middle-older generation: Oils and fats: 0.20* VF: 0.15 Younger-older generation: Oils and fats: 0.09 VF: −0.05 |
Strengths: a) studied 3 generations who were adults and lived apart; b) Validated FFQ was administered by interviewer. Limitations: Small sample size. |
[9] | Vauthier et al., 1996 Score=16 |
France | 774 children and 774 parents from 387 families | 7–21 y | Caucasian middle-class families from a longitudinal survey related to cardiovascular disease, only families with two parents (<65 yrs) and two children (>7 yrs) were included. | 3-day food consumption diary | Mixed: most self reported; for young children, by mothers with help from the children) |
Family correlation: Parent-child (n=1548): Energy (kcal): 0.30 Protein (% energy): 0.33 Fat (% energy): 0.34 CHO (% energy): 0.31 Father-son (n=365): Energy (kcal): 0.35* Protein (% energy): 0.36 Fat (% energy): 0.39 CHO (% energy): 0.37 Father-daughter (n=409): Energy (kcal): 0.33* Protein (%):0.36 Fat (%): 0.28 CHO (%): 0.26 Mother-son (n=365): Energy (kcal): 0.24* Protein (% energy): 0.26 Fat (% energy): 0.31 CHO (% energy): 0.28 Mother-daughter (n=409): Energy (kcal): 0.26* Protein (% energy): 0.31 Fat (% energy): 0.40 CHO (% energy): 0.35 |
Strengths: a) Took into account the number of meals shared. b) Large sample size and reported results by parent- child pairs. Additional findings: Cultural inheritance represented 30–40% of dietary intake variance for children. With increasing number of meals eaten together (>45/week vs <=45/week), between-generation components increased by 10% for fat and CHO, but unchanged for protein intake. |
[10] | Adelekan and Adeodu, 1997 Score=11 |
Nigeria | 108 mother- child pairs | 3–5 y | Sampled from rural areas, mothers were largely illiterate and farmers | 3 consecutive days of 24-hour recalls | No, by mothers |
Pearson correlation: energy (kJ): 0.39* protein (g): 0.07 total fat (g): 0.64* iron (mg): 0.55* |
Strengths: Mothers’ age was controlled for. |
[11] | Feunekes et al., 1997 Score=17 |
the Netherlands | 1077 households | 1–30 y | Derived from the national food surveys of 1987 and 1992. | 2-day diet records | Mixed: most self reported, byt for those aged<=13 y, by parents |
Pearson correlation: Father-son (n=914): Energy (MJ): 0.19* Total fat (% energy): 0.40* Saturated fat (% energy): 0.43* MUFA (% energy): 0.38* PUFA (% energy): 0.50* Cholesterol (mg/MJ): 0.41* Father-daughter (n=900): Energy (MJ): 0.24* Total fat (% energy): 0.39* Saturated fat (% energy): 0.37* MUFA (% energy): 0.42* PUFA (% energy): 0.48* Cholesterol (mg/MJ): 0.46* Mother-son (n=1003): Energy (MJ): 0.09* Total fat (% energy): 0.37* Saturated fat (% energy): 0.43* MUFA (% energy): 0.38* PUFA (% energy): 0.50* Cholesterol (mg/MJ): 0.47* Mother-daughter (n=998): Energy (MJ): 0.24* Total fat (% energy): 0.44* Saturated fat (% energy): 0.45* MUFA (% energy): 0.45* PUFA (% energy): 0.50* Cholesterol (mg/MJ): 0.55* |
Strengths: a) large sample size; b) wide child age range; c) child age was controlled for. Additional findings: Associations were higher for foods eaten at home than foods eaten outside of the home. Moderate within- family intake correlations were still found for families with a therapeutic dieting parent (not presented). |
[12] | Feunekes et al., 1998 Score=10 |
The Netherlands | 347 adolescents 309 mothers 270 fathers |
15 y | Urban and rural adolescents were selected from five schools | Self- administered FFQ from the Dutch National Food Consumption Survey (1987- 1988) | Yes |
Pearson correlation: Mother-child: Energy (MJ): 0.19* Energy (MJ/kg): 0.22* fat (% energy): 0.19* Saturated fat (% energy): 0.23* MUFA (% energy): 0.20* PUFA (% energy): 0.38* Cholesterol (mg/MJ):0.00 Father-child: Energy (MJ): 0.13* Energy (MJ/kg): 0.10 Total fat (% energy): 0.18* Saturated fat (% energy): 0.24* MUFA (% energy): 0.26* PUFA (% energy): 0.16* Cholesterol (mg/MJ):0.22* |
|
[34] | Billon S et al, 2002 Score=16 |
France | 398 families (398 fathers, 398 mothers, 383 sons, and 416 daughters) | (mean±sd) girl:14.3± 3.8 y; boy: 14.0±3.5 y | middle-class French families | 3 day food consumption diary | Yes | Correlation for absolute breakfast energy intake (BEI): FS: 0.25; FD: 0.19; MS:0.26; MD:0.27; Correlations for relative breakfast energy intake (RBEI): FS:0.23; FD: 0.19; MS:0.27; MD:0.21 [Note: RBEI =BEI/total energy intake over the 3 days.] |
Strengths: a)Used absolute and relative breakfast energy intake; b) adjusted for physical activity, smoking, BMI, alcohol consumption Limitations: limited sample in mid-class families in one city |
[13] | Cullen et al., 2002 Score=12 |
US | 132 children 132 parents |
mean age=10y; 4th–6th grade | Included multiple ethnic groups: white (42%), African Am (20%), Mexican Am (30%), Asian Am (8%) | Children: up to 7-day food records; parents, 21-item FFQ regarding fat practice over the previous week. | Yes |
Spearman correlation: Low fat practices: 0.28* |
Strengths: a) included multiple ethnic groups; b) Relative strong dietary assessment. |
[29] | Fisher et al., 2002 Score=12 |
US | 191 girls and their parents | 5 y | 1) Non-Hispanic White; 2) Parents were in their mid-30s, well-educated, and majority were employed |
Parents: FFQ; Children: 3 24-hour recalls | No, by mothers in the presence of their daughters |
Correlation coefficient (based on path coefficient from structural model): VF: 0.23* |
Strengths: Used structural equations modeling enable to find the net correlation accounting for other factors. Limitations: FFQ and 24-hr recalls gave different scales of intake, and the data from children and their parents might not be comparable. |
[35] | Longbottom et al, 2002 Score=9 |
Scotland | 36 children (12 boys) and their mothers | 5.5–8.5 y (mean/SD: 6.5±1.0) | 4-day weighed food records for the mother and their children, respectively | Yes |
Spearman’s correlation for density (weight/per 1000 kcal) for mother-child pairs for various foods: crude mean r for 26 food groups: 0.34 fat and oil: 0.24 fish: 0.15 meat & product: 0.21 fruit:0.74** Note: We calculated the crude mean r. |
Strengths: Used data on weighed food intakes. Limitations: small sample size. |
|
[36] | Mitchell et al., 2003 Score=15 |
US | 1364 members of 42 large Mexican American families | >16 y | The San Antonio Family Heart Study (SAFHS), a prospective study of cardiovascular risk in Mexican American families | FFQ modified for Mexican American population | Yes |
Age and sex-adjusted familial correlations: Parent-offspring: Energy: 0.14 Fat (% energy): 0.09 Parent-parent: Energy: 0.27 Fat (% energy): 0.15 Sibling-sibling: energy: 0.10 fat (% energy): 0.04 Note: P values were not reported, but the discussion stated, "For nearly all of these dietary variables, the correlations were statistically significant." |
Strengths: a) concurrent assessment of dietary intake of parent–sibling pairs; b) large sample size; c) reported sibling- sibling and parent- parent correlations. Limitations: a) a highly selective sample; b) small sample size. |
[37] | Runyan et al, 2003 Score=9 |
USA | 72 pairs of daughters and premenopausal mothers | 11–14 y (12.8±0.8 y) | mother aged 33- 51 years old with mean age as 42.4±4.2 | 3-day food records and a calcium intake survey | Yes |
Pearson correlation Calcium: 0.33* |
Limitations: small sample size |
[32] | Stanton et al., 2003 Score=11 |
US | 404 children and their mothers | 12–15 y | 1) 72% white, 28% black; 2) selected from rural areas 3) 97% were female parents; 4) Mothers’ age: 27–71 y; education: 39% <HS, 24% college graduates, 28.5% HS |
Same 35-item FFQ for children and parents | Yes |
Pearson correlation: Fat: Mother-daughter: 0.30** Mother-son: 0.11 Mother-child: 0.22** White dyads: 0.23** Black dyads: 0.18 Fiber: Mother-daughter: 0.13* Mother-son: −0.00 Mother-child: 0.06 White dyads: 0.10 Black dyads: 0.04 |
Strengths: Ethnic diversity |
[31] | Park et al., 2004 Score=15 |
South Korea | 231 children and 260 parents from 134 families | 11–19 y | 2) Excluded: obesity secondary to hypothyroidism or Cushing’s disease severe debilitating diseases or cancer Excluded: women lactating or pregnant, treated with anti- obesity agents, lost more than 10% of normal weight over the past 6 months. | FFQ from Korean National Health and Nutrition Survey Usual intake over the past 6 months | Yes |
Intraclass correlation analysis: Father-son: Energy: −0.03 CHO (% energy): 0.12 Protein (% energy): 0.02 Fat (% energy): 0.20 Saturated fat: 0.04 Cholesterol: −0.06 Father-daughter: Energy: 0.20 CHO (% energy): −0.10 Protein (% energy): 0.07 Fat (% energy): −0.01 Saturated fat: 0.02 Cholesterol: 0.03 Mother-son: Energy: 0.10 CHO (% energy): 0.23* Protein (% energy): 0.23* Saturated fat: 0.28* Cholesterol: 0.27* Mother-daughter: Energy: 0.27* CHO (% energy): 0.17 Protein (% energy): 0.31* Fat (% energy): 0.09 Saturated fat: 0.25* Cholesterol: 0.14 |
Strengths: Use of adjusted models (ICC adjusted for age, age-square and age-cubed). |
[30] | Galloway et al., 2005 Score=12 |
US | 173 girl- mother pairs | 7 y at baseline 9 y at follow-up | Non-Hispanic white girls from central Pennsylvania | Three 24-hr recalls when girls were 7, and again when girls were 9 | No, by mothers |
Correlation coefficient (based on path coefficient from structural model): Mothers’ VF intake when girls were 7 vs. girls’ VF intake at age 9: 0.36** |
Strengths: a) used structural equations modeling; b) longitudinal design. |
[40] | da Veiga and Sichieri, 2006 Score=12 |
Brazil | 391 fathers, 486 mothers, 287 boys, 256 girls | 12–18 y | A cross-sectional population-based survey in Rio de Janeiro | Parents: validated FFQ; Children: adult FFQ plus five new food items obtained from a pretest among adolescents | Yes |
Spearman correlation: Rice (1 soupspoon): Father-daughter: 0.45* Father-son: 0.52* Mother-daughter: 0.49* Mother-son: 0.50* Vegetables (1 portion): Father-daughter: 0.36* Father-son: 0.49* Mother-daughter: 0.46* Mother-son: 0.36* Milk products (1 glass or slice or unit): Father-daughter: 0.51* Father-son: 0.63* Mother-daughter: 0.60* Mother-son: 0.58* Meat (4–6 oz): Father-daughter: 0.53* Father-son: 0.63* Mother-daughter: 0.56* Mother-son: 0.58* Soda (1 glass): Father-daughter: 0.25* Father-son: 0.42* Mother-daughter: 0.39* Mother-son: 0.39* Mean correlation of foods for mother-child pairs: By level of maternal education: ≤4 y: 0.49 5–7 y: 0.40 >8 y: 0.41 By monthly per capita income: US$<100: 0.48 US$100–199: 0.45 US$≥200: 0.36 |
Strengths: a) concurrent assessment of dietary intake of parent–child pairs; b) used a validated comprehensive FFQs; c) tested effect of family socioeconomic level on food intake resemblance. |
[33] | Wang et al., 2009 Score=15 |
US | 121 child- mother pairs | 10–14 y | African American children from urban areas of Chic ago | the Youth and Adolescent Questionnaire(Y AQ) developed by Harvard Univ and the adult version of the FFQ | Yes |
Spearman correlation: Energy (kcal): Mother-daughter: 0.26* Mother-son: −0.24 Mother-child: 0.04 Fat (g): Mother-daughter: 0.30* Mother-son: −0.21 Mother-child: 0.07 Fat (% energy): Mother-daughter: 0.11 Mother-son: 0.19 Mother-child: 0.16 Fiber (g): Mother-daughter: 0.12 Mother-son: −0.08 Mother-child: 0.02 Calcium (mg): Mother-daughter: 0.19 Mother-son: −0.19 Mother-child: 0.02 |
Strengths: a) used comparable comprehensive FFQs for children and their mothers; b) studied the factors might be associated with the resemblance. Limitations: A selective US population group. |
[28] | Beydoun and Wang., 2009 Score=20 |
US | 1061 fathers, 1230 mothers, 1370 sons and 1322 daughters | 2–18 y | A nationally representative multi-stage stratified data from the USDA CSFII 1994–96 | Two 24 hour recalls | Mixed: most self reported, but for children aged 2–9 y, 7% and 11% self- reported dietary intake on days 1 and 2, respectively) | Multivariate-adjusted standardized regression coefficient (can be considered as adjusted correlations): Total dietary quality score: Parent-child: 0.26* Mother-daughter: 0.18 Mother-son: 0.28* Father-daughter: 0.28* Father-son: 0.29* Energy (kcal): Parent-child: 0.22* Mother-daughter: 0.26* Mother-son: 0.23* Father-daughter: 0.14* Father-son: 0.29* Fat (g): Parent-child: 0.24* Mother-daughter: 0.24* Mother-son: 0.28* Father-daughter: 0.18* Father-son: 0.27* Fat (% energy): Parent-child: 0.01 Mother-daughter: 0.02 Mother-son: −0.04 Father-daughter: 0.02 Father-son: 0.01 VF: Parent-child: 0.29 Mother-daughter: 0.37 Mother-son: 0.31 Father-daughter: 0.21 Father-son: 0.29 |
Strengths: a) Used nationally representative data and sampling design complexity was accounted for. b) adjusted for some covariates Limitations: a) Only two 24-h recalls used. b) Clustering at the household level. Additional findings: The study provided many more results about other dietary intake measures including other nutrients and food groups and the correlations by parent-child characteristics such as race, age and family income. |
[38] | Papas et al, 2009 Score=9 |
USA | 109 adolescent mothers and their toddler children | 13 months old | primiparous, low-income, African American mothers and their toddlers | Toddler dietary variety was measured with a 73-item feeding checklist; maternal dietary variety was measured with the Youth Adolescent Food Frequency Questionnaire (YAQ). | No, by mothers |
Spearman correlation: fruit: 0.26** vegetables: 0.40** snacks/disserts: 0.50** meats: 0.29** diary: 0.18* soda: 0.25* |
Strengths: longitudinal study Limitations: 1) highly selective sample; b) dietary assessment: maternal dietary variety was measured as a frequency of food items consumed over the past year, but toddlers’, by food items consumed over the previous week. |
[39] | López- Alvarenga et al., 2007 Score=11 |
Mexico | 552 children (406 subjects filled the questionnaires for both parents and the child and 146 only filled the questionnaires for one parent (father or mother) and the child). | 8–12 y | From the initial sample (n=668), 300 participants were attending a low SES school and 368 participants came from a medium- high SES school. | The families filled in one questionnaire made up of 43 questions about parents’ habits, 34 regarding the children’s habits and 31 questions related with food intake frequency and quantity. | No, by parents |
First canonical root for child- parent food preference correlations: Yogurt, Oaxaca cheese and yellow cheese: 0.75** Skimmed milk, skimmed yogurt and diet soda: 0.67** Omelet, bread and beans: 0.80** Eggs, pork, butter, fried potatoes, tacos, soups, sandwiches and pizza: 0.77** Fish: 0.79** Sausages, beef, poultry and oat: 0.71** French fries, sweet bread, cakes and sweetened cereals: 0.81** VF: 0.74** Sweets and soft drinks: 0.74** Whole milk: 0.50** |
Limitations: a) only limited food groups were included in the questionnaires; b) Assessed food preferences, but not actual food intakes; c) The canonical root method may be an overestimate compared to regular correlation. |
These studies were listed by publication year. CHO, carbohydrate; MUFA, monounsaturated fat; PUFA, polyunsaturated fat; VF, vegetable and fruit; % of energy, % of total energy derived from the nutrient.
The methodological quality of each study is assessed based on the study design, study sample, data collection including how dietary assessment, data analysis and results reported (see Appendix for related details). Seven separate scores (range: 1 to 3 best) was assigned for each of these features, and then a total score (the highest score was 3*7=21) was calculated for each study. The final total score for each study was used in our meta-analysis for those studies included in the meta-analysis.
P<0.05;
P<0.01.