Abstract
This survey study uses data from the National Immunization Survey–Child to examine trends in early formula supplementation by race/ethnicity among US children born during the period from 2009 to 2015.
Breastfeeding is the best source of nutrition for most infants. It is associated with a reduction in the risk for some health conditions for both infants and mothers.1,2 The American Academy of Pediatrics recommends that infants be fed only human milk for about the first 6 months of life, with continued breastfeeding along with complementary foods for at least 1 year.3 Previous studies have indicated that early formula supplementation is associated with the exclusivity and duration of breastfeeding,4 but, to our knowledge, trend analysis on formula supplementation among US children is lacking. This survey study examines the trends in early formula supplementation by race/ethnicity using data from the National Immunization Survey–Child (NIS-Child) of US children born from 2009 to 2015.
Methods
The NIS-Child is a national ongoing, random-digit–dialed telephone survey conducted by the Centers for Disease Control and Prevention.5 Our study is based on data from mixed telephone sampling (landline and cellular) from January 1, 2011, through December 31, 2017, among a complex, stratified, multistage probability sample of US households with children aged 19 to 35 months at the time of the survey. Because each annual NIS sample includes children born in 3 calendar years, we analyzed the data by birth year from 2009 through 2015. The protocol under which data were collected by the NIS during the period from 2011 to 2017 was reviewed and approved by the National Center for Health Statistics’ Research Ethics Review Board.
Statistical analysis was performed from September 1, 2019, to February 16, 2020. The prevalence of formula supplementation (with or without other supplementary liquids or solids) before 2 days or 2 weeks of age was calculated among infants who were breastfeeding at each specified age. We examined the changes of population distribution using logistic regressions with each covariate dichotomized and regressed on birth year (Table 1). To indicate differences during the study period, we combined the first 2 (2009-2010) and last 2 (2014-2015) birth cohorts and calculated adjusted prevalence differences and adjusted prevalence ratios using estimated marginal values from log-binomial regressions controlling for infant sex; birth order; participation in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC); marital status; maternal educational level; and household income. To account for cluster sampling design, Taylor series linearization was used for variance estimation using SUDAAN, version 11 (RTI International). Survey design variables and sample weights accounting for differential probabilities of selection, nonresponse, and noncoverage were used to obtain estimates representative of young children in the US. All P values were from 2-sided tests and results were deemed statistically significant at P < .05.
Table 1. Characteristics of Study Sample of Infants Who Were Breastfed for at Least 2 Days, National Immunization Survey–Child, 2009-2015 Birth Cohorts (N = 137 022)a,b.
| Characteristic | Year of birth, No. (%)c | P value | ||||||
|---|---|---|---|---|---|---|---|---|
| 2009 (n = 18 535) | 2010 (n = 19 164) | 2011 (n = 18 822) | 2012 (n = 20 566) | 2013 (n = 21 841) | 2014 (n = 23 075) | 2015 (n = 15 019) | ||
| Race/ethnicity | ||||||||
| Hispanic | 3428 (29.4) | 3634 (28.4) | 3636 (28.4) | 3991 (26.6) | 4323 (27.6) | 4520 (27.9) | 3139 (27.1) | .03 |
| Non-Hispanic | ||||||||
| White | 11 431 (48.2) | 11 619 (48.6) | 11 264 (48.3) | 12 110 (49.2) | 12 675 (49.1) | 13 493 (48.1) | 8652 (48.8) | .72 |
| Black | 1575 (11.0) | 1553 (10.4) | 1477 (10.7) | 1671 (11.0) | 1814 (10.7) | 1799 (10.7) | 1139 (10.7) | .98 |
| Asian | 779 (5.0) | 805 (4.6) | 900 (5.2) | 984 (5.1) | 1054 (5.1) | 1158 (5.5) | 708 (5.1) | .26 |
| American Indian or Alaska Native | 242 (0.9) | 279 (1.1) | 258 (0.9) | 264 (0.9) | 289 (0.8) | 266 (0.9) | 197 (1.0) | .72 |
| Male infant | 9491 (50.9) | 9829 (51.7) | 9734 (51.5) | 10 494 (50.6) | 11 287 (51.0) | 11 870 (51.1) | 7770 (51.5) | .93 |
| First born child | 6764 (38.9) | 7428 (39.9) | 7416 (39.4) | 8317 (42.1) | 9181 (41.5) | 9598 (40.6) | 5987 (39.8) | .04 |
| Receiving WICd | ||||||||
| Yes | 7028 (52.7) | 7611 (52.2) | 7480 (51.4) | 8099 (50.3) | 8392 (49.3) | 8383 (47.4) | 5543 (47.8) | <.001 |
| No, but eligible | 1216 (7.2) | 1403 (7.1) | 1340 (7.4) | 1566 (7.8) | 1622 (7.0) | 1934 (8.9) | 1297 (8.5) | <.001 |
| Ineligible | 10 221 (40.1) | 10 069 (40.7) | 9911 (41.2) | 10 813 (42.0) | 11 679 (43.6) | 12 594 (43.7) | 8078 (43.7) | <.001 |
| Married mothers | 14 502 (68.7) | 14 748 (68.4) | 14 432 (67.4) | 15 610 (67.5) | 16 485 (66.8) | 17 505 (66.2) | 11 229 (67.0) | .01 |
| Maternal educational level | ||||||||
| <High school | 1670 (17.1) | 1663 (16.0) | 1617 (15.9) | 1735 (14.2) | 1730 (13.4) | 1871 (14.0) | 1296 (13.3) | <.001 |
| High school graduate | 2777 (24.4) | 2951 (23.6) | 2852 (22.4) | 3228 (23.1) | 3360 (22.6) | 3421 (21.7) | 2257 (21.2) | <.001 |
| Some college or technical school | 4736 (22.6) | 4945 (22.3) | 4947 (23.0) | 5288 (23.5) | 5615 (24.4) | 5765 (22.8) | 3730 (23.5) | .09 |
| College graduate | 9352 (35.9) | 9605 (38.1) | 9406 (38.7) | 10 315 (39.2) | 11 136 (39.7) | 12 018 (41.6) | 7736 (42.0) | <.001 |
| Household federal poverty level | ||||||||
| <100% of poverty level | 3835 (33.1) | 4154 (33.0) | 4003 (32.1) | 4474 (31.6) | 4643 (30.6) | 4492 (28.2) | 3066 (27.7) | <.001 |
| 100%-199% of poverty level | 3352 (20.5) | 3777 (20.4) | 3716 (20.7) | 3945 (19.8) | 4054 (19.3) | 4351 (20.4) | 2791 (20.3) | .51 |
| 200%-399% of poverty level | 5149 (21.9) | 5285 (22.5) | 5104 (22.0) | 5604 (23.6) | 5851 (22.8) | 6201 (23.1) | 3930 (23.1) | .04 |
| 400%-599% of poverty level | 3487 (14.4) | 3300 (12.9) | 3226 (13.4) | 3553 (13.6) | 3722 (14.0) | 3980 (13.8) | 2540 (14.5) | .32 |
| ≥600% of poverty level | 2712 (10.1) | 2648 (11.3) | 2773 (11.9) | 2990 (11.3) | 3571 (13.3) | 4051 (14.5) | 2692 (14.3) | <.001 |
Abbreviation: WIC, Special Supplemental Nutrition Program for Women, Infants and Children.
Each birth cohort includes data from 3 survey years except for 2009 and 2015 births when only 2 survey years of data were available.
Infants who are non-Hispanic Hawaiian or Pacific Islander or are of 2 or more races/ethnicities are included in the total estimates and by other demographics analysis, but not presented separately in this table owing to relatively small sample size among these 2 groups.
Percentages are weighted.
The sample sizes in each subgroup for this variable do not add up to the total owing to missing values.
Results
Table 1 indicates some significant demographic differences during the study period. Overall, the crude prevalence of formula supplementation before 2 days of age among breastfed infants decreased from 23.3% (95% CI, 22.1%-24.4%) in 2009 to 17.2% (95% CI, 16.0%-18.4%) in 2015, while the crude prevalence of formula supplementation before 2 weeks of age decreased from 28.5% (95% CI, 27.2%-29.7%) in 2009 to 21.7% (95% CI, 20.4%-23.0%) in 2015 (Table 2). For differences between the 2009-2010 and 2014-2015 birth cohorts, adjusted analyses (controlling for infant sex, birth order, WIC participation, marital status, maternal educational level, and household income) indicated statistically significant decreases in formula supplementation both before 2 days of age (adjusted prevalence difference, −5.8% [95% CI, −6.9% to −4.7%]; adjusted prevalence ratio, 0.74 [95% CI, 0.70-0.79]) and before 2 weeks of age (adjusted prevalence difference, −6.7% [95% CI, −7.9% to −5.5%]; adjusted prevalence ratio, 0.76 [95% CI, 0.72-0.80]). Except for non-Hispanic American Indian and Alaska Native individuals, all racial/ethnic groups had a significant reduction in early formula supplementation, with the largest decreases among non-Hispanic Black infants.
Table 2. Prevalence of Formula Supplementation Among Breastfed Infants Born During the Period From 2009 to 2015, National Immunization Survey–Childa.
| Formula supplementation | Prevalence, % (95% CI)b | 2009-2010 vs 2014-2015c | |||||||
|---|---|---|---|---|---|---|---|---|---|
| 2009 | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | Adjusted prevalence difference (95% CI) | Adjusted prevalence ratio (95% CI) | |
| Before 2 d (n = 137 022) | |||||||||
| Overalld | 23.3 (22.1 to 24.4) | 22.2 (21.0 to 23.4) | 20.1 (19.0 to 21.2) | 18.8 (17.8 to 19.9) | 16.7 (15.7 to 17.6) | 16.0 (15.1 to 16.9) | 17.2 (16.0 to 18.4) | −5.8 (−6.9 to −4.7) | 0.74 (0.70 to 0.79) |
| Hispanicd | 28.4 (25.8 to 31.1) | 29.9 (27.0 to 32.8) | 27.3 (24.4 to 30.1) | 25.8 (23.2 to 28.3) | 21.6 (19.2 to 24.0) | 19.9 (17.6 to 22.2) | 23.2 (20.1 to 26.4) | −6.5 (−9.2 to −3.8) | 0.77 (0.69 to 0.86) |
| Non-Hispanic | |||||||||
| Whited | 18.1 (16.9 to 19.3) | 17.0 (15.8 to 18.2) | 15.4 (14.2 to 16.5) | 14.0 (12.8 to 15.2) | 12.5 (11.5 to 13.4) | 12.9 (12.0 to 13.9) | 12.7 (11.5 to 13.9) | −4.6 (−5.7 to −3.5) | 0.74 (0.68 to 0.79) |
| Blackd | 29.4 (25.5 to 33.3) | 26.1 (22.7 to 29.5) | 22.1 (18.7 to 25.4) | 23.4 (20.1 to 26.8) | 22.2 (19.3 to 25.1) | 18.3 (15.5 to 21.1) | 20.9 (17.2 to 24.5) | −7.9 (−11.3 to −4.5) | 0.71 (0.61 to 0.83) |
| Asiand | 30.1 (24.2 to 36.1) | 26.0 (19.6 to 32.3) | 22.9 (18.6 to 27.2) | 24.6 (19.9 to 29.3) | 22.0 (17.5 to 26.4) | 21.4 (17.4 to 25.3) | 21.4 (16.2 to 26.7) | −6.6 (−11.9 to −1.3) | 0.76 (0.62 to 0.94) |
| American Indian or Alaska Native | 19.5 (8.7 to 30.3) | 13.5 (7.6 to 19.5) | 14.8 (7.2 to 22.4) | 22.3 (11.5 to 33.2) | 11.5 (6.3 to 16.8) | 13.9 (7.7 to 20.0) | 14.3 (6.9 to 21.7) | −3.2 (−11.0 to 4.6) | 0.81 (0.48 to 1.35) |
| Before 2 wk (n = 134 491) | |||||||||
| Overalld | 28.5 (27.2 to 29.7) | 27.4 (26.1 to 28.6) | 24.5 (23.3 to 25.7) | 23.4 (22.3 to 24.5) | 21.3 (20.3 to 22.3) | 20.3 (19.3 to 21.3) | 21.7 (20.4 to 23.0) | −6.7 (−7.9 to −5.5) | 0.76 (0.72 to 0.80) |
| Hispanicd | 32.2 (29.4 to 35.1) | 33.4 (30.3 to 36.5) | 30.8 (27.8 to 33.8) | 30.2 (27.3 to 33.0) | 24.8 (22.2 to 27.4) | 23.2 (20.8 to 25.7) | 27.0 (23.7 to 30.3) | −7.0 (−9.9 to −4.1) | 0.78 (0.71 to 0.87) |
| Non-Hispanic | |||||||||
| Whited | 23.8 (22.5 to 25.1) | 23.2 (21.8 to 24.5) | 20.3 (19.0 to 21.5) | 19.0 (17.7 to 20.3) | 17.3 (16.2 to 18.4) | 17.4 (16.3 to 18.6) | 17.5 (16.1 to 18.8) | −6.1 (−7.4 to −4.8) | 0.74 (0.70 to 0.79) |
| Blackd | 34.8 (30.7 to 38.9) | 30.0 (26.4 to 33.7) | 25.7 (22.1 to 29.2) | 27.0 (23.5 to 30.4) | 26.7 (23.5 to 29.9) | 22.6 (19.5 to 25.7) | 25.3 (21.3 to 29.3) | −8.3 (−12.0 to −4.6) | 0.74 (0.65 to 0.85) |
| Asiand | 34.6 (28.4 to 40.7) | 32.7 (26.1 to 39.3) | 27.4 (22.6 to 32.3) | 28.5 (23.5 to 33.6) | 29.1 (24.1 to 34.2) | 27.7 (23.0 to 32.4) | 25.5 (19.9 to 31.1) | −6.3 (−12.0 to −0.6) | 0.81 (0.67 to 0.98) |
| American Indian or Alaska Native | 21.5 (10.2 to 32.8) | 17.9 (10.0 to 25.9) | 18.3 (9.8 to 26.8) | 23.7 (12.4 to 34.9) | 21.3 (13.5 to 29.1) | 15.0 (9.0 to 21.0) | 19.9 (11.6 to 28.2) | −4.0 (−12.1 to 4.1) | 0.80 (0.52 to 1.24) |
Abbreviation: WIC, Special Supplemental Nutrition Program for Women, Infants and Children.
Infants who are non-Hispanic Hawaiian or Pacific Islander or are of 2 or more races/ethnicities are included in the total estimates, but not presented separately in this table owing to relatively small sample size among these 2 groups.
Crude prevalence of formula supplementation with or without other supplementary liquids or solids before 2 days or 2 weeks of age among infants breastfed at each specified age.
Calculated between 2009-2010 and 2014-2015 by the estimated values from log-binomial regression model controlling for infant sex, birth order, participation in WIC, marital status, maternal educational level, and household income.
P < .05 for linear trend tests with all years’ data of crude prevalence included.
Discussion
Consistent with previous findings on the steady increase in breastfeeding rates,6 this study suggests that early formula supplementation among breastfed infants decreased among all racial/ethnic groups except for non-Hispanic American Indian and Alaska Native individuals. The nonsignificant changes among non-Hispanic American Indian and Alaska Native individuals might be partially explained by the relatively small sample in this group. The reasons for decreases in early formula supplementation during the study period remain undetermined but may include the implementation of evidence-based maternity care practices and of local, state, and national initiatives designed to improve breastfeeding support in the early postpartum period.2
References
- 1.Ip S, Chung M, Raman G, et al. Breastfeeding and maternal and infant health outcomes in developed countries. Evid Rep Technol Asses (Full Rep). 2007;(153):1-186. [PMC free article] [PubMed] [Google Scholar]
- 2.Feltner C, Weber RP, Stuebe A, Grodensky CA, Orr C, Viswanathan M. Breastfeeding Programs and Policies, Breastfeeding Uptake, and Maternal Health Outcomes in Developed Countries. Comparative Effectiveness Review No. 210. Agency for Healthcare Research and Quality; 2018. AHRQ publication 18-EHC014-EF. [PubMed]
- 3.Section on Breastfeeding Breastfeeding and the use of human milk. Pediatrics. 2012;129(3):e827-e841. doi: 10.1542/peds.2011-3552 [DOI] [PubMed] [Google Scholar]
- 4.Eilers MA, Hendrick CE, Pérez-Escamilla R, Powers DA, Potter JE. Breastfeeding initiation, duration, and supplementation among Mexican-origin women in Texas. Pediatrics. 2020;145(4):e20192742. doi: 10.1542/peds.2019-2742 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Centers for Diseases Control and Prevention About the National Immunization Surveys. Accessed July 29, 2020. https://www.cdc.gov/vaccines/imz-managers/nis/about.html
- 6.Li R, Perrine CG, Anstey EH, Chen J, MacGowan CA, Elam-Evans LD. Breastfeeding trends by race/ethnicity among US children born from 2009 to 2015. JAMA Pediatr. 2019;173(12):e193319. doi: 10.1001/jamapediatrics.2019.3319 [DOI] [PMC free article] [PubMed] [Google Scholar]
