Table 2.
Cohort studies included in the systematic review ordered according to quality score
| Authors | Country | Local Setting (Initial and Final Date) | Sample | Children’s Age at Dental Examination | Dental Examination Criteria/Index (Calibration) | Types of Malocclusion Analyzed | Instrument and Time Interval of Feeding Habit Evaluation | Statistics (Adjusted for Confounders) | Outcomes(OR; 95% CI) or (p-value) | QualityScore |
|---|---|---|---|---|---|---|---|---|---|---|
| Davis & Bell, 1991 [16] | Canada | National database (Beginning with newborns in 1983 and the children were examined in 1988) | Initial = 670 Final = 108 | 5 years | Does not report the use of an index/criteria (NR †) | Molar relationships, crossbite, overjet, overbite, crowding, drifting | Questionnaires answered by mothers, monthly, from 1983 to 1988 | Chi-square and Kruskal-Wallis (no) | No significant association between malocclusion and feeding method (p > 0.05) except for overjet which was associated with exclusive bottle feeding (RR ‡‡ = 6.62; p = 0.006) | 4 (10) |
| Karjalainen et al., 1999 [17] | Finland | Recruited from a prospective baby trial (NR †) | Initial = 179 Final = 148 | 3 years | Does not report the use of an index/criteria (NR †) | Posterior crossbite, anterior open bite, overjet | Parent interviews recorded 10 times at 1–3 month intervals until the child reached 3 years of age | Chi-square tests and covariance analysis (no) | Children breastfed for 4.7 months had greater frequency of posterior crossbite than children breastfed for 7.6 months (p < 0.01). Breastfeeding was not associated with overjet or anterior open bite (p > 0.05) | 4 (10) |
| Warren & Bishara, 2002 [12] * | USA | Recruited from hospitals (began with newborns in 1992–1995 and the children were examined at 4.5–5 years of age) | Initial = 700 Final = 372 | 4.5–5 years | Study models evaluated using Angle classification (NR †) | Primary canine relationship, anterior and posterior crossbite, anterior open bite, overjet, overbite | Questionnaires answered by mothers at 3, 6, 9, 12, 16, 20 and 24 months of age and yearly thereafter | ANOVA(no) | No significant association between malocclusion and duration of breastfeeding (p ≥ 0.05) | 5 (10) |
| Viggiano et al., 2004 [18] | Italy | Recruited from a school (began with newborns in 1993–1995 and the children were examined in 1998) | Initial = 1130 Final = 1099 | 3–5 years | Does not report the use of an index/criteria (NR †) | Molar relationships, posterior crossbite, anterior open bite | Structured questionnaire. The data was collected only once | Logistic regression(NNSH ‡) | Bottle feeding associated with crossbite (OR: 2.54; 95% CI: 1.66–4.03), but not with open bite (OR: 0.93; 95% CI: 0.65, 1.33) or malocclusion (OR: 1.28; 95% CI: 0.99, 1.66) | 5 (10) |
| Bishara et al., 2006 [19] * | USA | Recruited from hospitals (began with newborns in 1992–1995 and the children were examined at 4.5–5 years of age) | Initial = 547 Final = 372 | 4.5–5 years | Study models evaluated using Angle classification (NR†) | Molar relationships, posterior crossbite, overjet, overbite, anterior open bite | Questionnaires answered by mothers at 3, 6, 9, 12, 16, 20 and 24 months and yearly thereafter | McNemar test (no) | No significant difference between children breastfed for 6–12 months without NNSH and children who were not breastfed but had NNSH <12 months (p > 0.05) | 5 (10) |
| Vázquez-Nava et al., 2006 [7] | Mexico | NR † | Initial = NR † Final = 1160 | 4–5 years | Does not report the use of an index/criteria (NR †) | Anterior open bite, posterior cross bite | Validated questionnaire. The data was collected only once | Qui-square and logistic regression (NNSH ‡, allergic rhinitis) | Bottle feeding associated with malocclusion (OR: 1.37; 95% CI: 1.06, 1.78) and crossbite (OR: 1.95; 95% CI: 1.07, 3.54). Bottle feeding was not associated with open bite (OR: 1.27; 95% CI: 0.98, 1.64) | 5 (10) |
| Peres et al., 2007 [13] * | Brazil | Recruited from hospitals (began with newborns in 1993 and the children were examined in 1999) | Initial = 400 Final = 359 | 6 years | Foster and Hamilton criteria (Kappa ≥ 0.85) | Open bite | Interviews with mothers at 1, 3, 6 and 12 months (1993) and in the child’s fifth year of life | Multivariate analysis (NNSH ‡: pacifier/finger sucking, socioeconomic indicators, maternal characteristics) | Open bite was not associated with bottle feeding at 5 years of age in the adjusted analysis (p > 0.05). Open bite was associated with breastfeeding <9 months (OR: 2.7; 95% CI: 1.4,6.8, adjusted for dental caries, NNSH‡ , maternal schooling and maternal behavioral characteristics) | 8 (10) |
| Peres et al., 2007 [14] * | Brazil | Recruited from hospitals (began with newborns in 1993 and the children were examined in 1999) | Initial = 400 Final = 359 | 6 years | Foster and Hamilton criteria(Kappa ≥ 0.85) | Anterior open bite, posterior crossbite | Interviews with mothers at 1, 3, 6 and 12 months (1993) and in the child’s fifth year of life | Multivariate analysis, Poisson regression (Time of breastfeeding and NNSH ‡‡: pacifier/finger sucking, gender, maternal schooling) | Posterior crossbite was associated with duration of breastfeeding (p = 0.036). Posterior crossbite was associated with duration of breastfeeding even after adjustment for the time of NNSH ‡ (OR = 7.6; 95% CI: 1.5, 39.5). Anterior open bite was associated with breastfeeding <9 months (p = 0.004). After adjustment for the use of pacifier, breastfeeding duration lost significance (OR = 1.2; 95% CI: 0.8, 1.7). | 8 (10) |
| Caramez da Silva et al., 2012 [20] | Brazil | Recruited from a hospital (began with newborns in 1993 and the children were examined between 3–5 years-old | Initial = 220 Final = 153 | 3–5 years | Foster and Hamilton criteria (NR †) | Distocclusion (Class II) | Interview with mothers at 7, 30, 60, 120, 180 days of life and between 3–5 years | Chi-square and Poisson regression (adjusted for duration of pacifier use and bottle-feeding) | Breastfeeding for 12 months or longer protects against canine Class II relation (PR ††: 0.44; 95% CI: 0.23, 0.82) | 8 (10) |
| Moimaz et al., 2014 [21] | Brazil | Recruited from a program of prenatal care (began with newborns in November 2008 and the children were examined in May 2010) | Initial = 120 Final = 80 | 30 months | Own criteria(Kappa = 0.92) | Posterior crossbite, anterior crossbite, open bite | Interviews (semi-structured questionnaires) with mothers at 12, 18 and 30 months | Chi-square test and Fisher’s exact test (no) | Posterior crossbite was associated with bottle feeding at 12 and 30 months (p = 0.02 and p = 0.04, respectively). Overjet > 3mm was associated with breastfeeding at 12 and 18 months (p < 0.0001) and at 30 months (p = 0.01). Open bite was associated with breastfeeding at 12, 18 and 30 months (p < 0.001, p = 0.001 and p = 0.01, respectively) | 6 (10) |
* Publications belonging to same epidemiological study reporting different data; † NR = not reported; ‡ NNSH = non-nutritive sucking habits; †† PR = prevalence ratio; ‡‡ RR = relative risk.