Maternal and Child Health Journal

, Volume 16, Issue 3, pp 609–614

Pacifier Use and Sids: Evidence for a Consistently Reduced Risk

Authors

    • Goldberg Center for Community Pediatric HealthChildren’s National Medical Center
    • Department of PediatricsGeorge Washington University School of Medicine and Health Sciences
  • Kawai O. Tanabe
    • Department of Family MedicineUniversity of Virginia School of Medicine
  • Diane Choi Yang
    • Department of Epidemiology and BiostatisticsGeorge Washington University School of Public Health and Health Sciences
  • Heather A. Young
    • Department of Epidemiology and BiostatisticsGeorge Washington University School of Public Health and Health Sciences
  • Fern R. Hauck
    • Department of Family MedicineUniversity of Virginia School of Medicine
    • Department of Public Health SciencesUniversity of Virginia School of Medicine
Article

DOI: 10.1007/s10995-011-0793-x

Cite this article as:
Moon, R.Y., Tanabe, K.O., Yang, D.C. et al. Matern Child Health J (2012) 16: 609. doi:10.1007/s10995-011-0793-x

Abstract

Pacifier use at sleep time decreases sudden infant death syndrome (SIDS) risk. It is yet unclear whether pacifier use can modify the impact of other sleep-related factors upon SIDS risk. The objective of this study was to examine the association between pacifier use during sleep and SIDS in relation to other risk factors and to determine if pacifier use modifies the impact of these risk factors. Data source was a population based case–control study of 260 SIDS deaths and 260 matched living controls. Pacifier use during last sleep decreased SIDS risk (aOR 0.30, 95% CI 0.17–0.52). Furthermore, pacifier use decreased SIDS risk more when mothers were ≥20 years of age, married, nonsmokers, had adequate prenatal care, and if the infant was ever breastfed. Pacifier use also decreased the risk of SIDS more when the infant was sleeping in the prone/side position, bedsharing, and when soft bedding was present. The association between adverse environmental factors and SIDS risk was modified favorably by pacifier use, but the interactions between pacifier use and these factors were not significant. Pacifier use may provide an additional strategy to reduce the risk of SIDS for infants at high risk or in adverse sleep environments.

Keywords

SIDSPacifierSleep positionBed sharingRisk factor

Introduction

Sudden infant death syndrome (SIDS) is defined as “the sudden death of an infant less than 1 year of age, which remains unexplained after a thorough case investigation, including performance of a complete autopsy, examination of the death scene, and review of the clinical history” [1]. As a result of risk reduction campaigns encouraging families to place infants on their backs to sleep, the rates of both prone positioning and SIDS have declined more than 50% [2]; however, SIDS remains the leading cause of death for infants aged 1 month to 1 year [2]. Epidemiologic studies have identified other behavioral factors that affect SIDS risk, including smoke exposure [36], soft bedding [716], and bedsharing [1724].

Pacifiers, through a yet undefined mechanism, when used at sleep time, may reduce the risk of SIDS by as much as 90% [2528]. Indeed, the American Academy of Pediatrics (AAP) Task Force on SIDS recommends that parents consider pacifier use for their infants at naptime and bedtime [29]. While several studies have examined the association between pacifier use and SIDS, it is still unclear if pacifiers remain protective in the context of known SIDS risk factors, and what, if any, interactions exist between pacifier use and these factors in association with SIDS risk. A recent population-based case–control study in California found pacifiers to reduce the risk of SIDS for every category of maternal and infant factor examined, for example, for breastfeeding babies as well as for formula fed babies [27]. It also found that with pacifier use, the increased risk of SIDS with prone sleeping, bedsharing, or soft bedding was reduced, although these results did not reach statistical significance. To our knowledge, no other studies have been published that examine these more complex interactions.

Methods

We analyzed data from the Chicago Infant Mortality Study (CIMS). CIMS was designed to examine risk factors for SIDS and other sudden unexpected deaths in infancy, with the principal goal of gathering information to aid in eliminating disparities in postneonatal infant mortality between African–Americans and Whites. The study resulted from concern that the city’s SIDS and postneonatal infant mortality rates were among the highest in the country, and that African–American infants there were over 3 times more likely to die from SIDS and 4 times more likely to die from all causes of postneonatal mortality than White or Hispanic infants [3032]. CIMS enrolled all Chicago resident infants, birth to 1 year, who died suddenly and unexpectedly between November 1993 and April 1996. The autopsy was performed by the Office of the Medical Examiner of Cook County, Chicago. An extensive death scene investigation with 400 questions was completed soon after the medical examiner received notification about the infant’s death. The questions asked details describing events leading up to the event, risk factors, and socioeconomic factors. Approximately 2 weeks after the death, a separate interview was conducted with the primary caregiver in the home. This consisted of 235 questions that focused on issues such as routine sleep habits, social stressors, and access to health care. The controls were living infants who were matched to each case infant by maternal race/ethnicity, infant age, and birth weight [13]. For this study, we analyzed data for 260 SIDS cases and 260 SIDS controls.

Because the cases and controls were matched, we first performed conditional logistic regression analyses to measure the associations between several sleep related and other factors and SIDS. Univariate conditional logistic regression models were used to determine the unadjusted odds ratios associated with each of the individual risk factors. A multiple conditional logistic regression model was then constructed to determine the odds of SIDS for each of the individual risk factors while adjusting for known confounders, including maternal age, marital status, maternal education, and Kessner Index, which measures adequacy of prenatal care [33]. We then conducted unconditional logistic regression models (since we had to break the matches) to measure the association of pacifier use at last sleep with SIDS for each category of the risk factors described in Table 1, both unadjusted and adjusted for the four confounding variables. Finally, we constructed conditional logistic regression models, with interaction terms and adjusting for the confounding variables, to determine if pacifier use modified the association between specific factors known to affect SIDS risk (sleep position, maternal smoking during pregnancy, bedsharing, and breastfeeding) and SIDS. Odds ratios were calculated for each variable among pacifier users and non-users, and a P value of <0.05 for the interaction term was considered significant, i.e., the odds ratios were significantly different.
Table 1

Unadjusted and adjusted odds ratios of variables and risk of sudden infant death syndrome

Variable

Unadjusted odds ratio(95% CI)

Adjusted odds ratio (95% CI)*

Maternal smoking during pregnancy

4.85 (3.0–7.85)

4.30 (2.44–7.57)

Ever breast-fed

0.21 (0.13–0.34)

0.39 (0.23–0.68)

Breast-feeding now

0.20 (0.11–0.38)

0.31 (0.15–0.67)

Placed prone

2.38 (1.66–3.41)

2.36 (1.54–3.60)

Found prone

3.79 (2.50–5.74)

4.53 (2.74–7.49)

Soft bedding

5.05 (3.09–8.26)

5.52 (3.12–9.77)

Pillow use

2.52 (1.52–4.18)

3.13 (1.70–5.75)

Pacifier use

0.33 (0.21–0.54)

0.30 (0.17–0.52)

Bedsharing

2.79 (1.83–4.27)

2.02 (1.26–3.24)

* Adjusted for maternal age, marital status, maternal education, and adequacy of prenatal care (Kessner Index)

This study was approved by the institutional review board of the University of Virginia and granted exemption from the institutional review board of Children’s National Medical Center.

Results

After adjusting for known confounders, pacifier use reduced the risk of SIDS by approximately 70% (OR 0.30, 95% CI 0.17–0.52) (Table 1). Pacifier use was also associated with a reduced risk of SIDS in every category of the maternal and infant factors examined (Table 2). With regard to maternal and infant characteristics, pacifier use appeared to decrease SIDS risk even more when mothers were ≥20 years of age, married, nonsmokers, had adequate prenatal care, and if the infant was ever breastfed. Pacifier use also appeared to decrease the risk of SIDS when the infant was in particular sleep environments, such as sleeping in the prone or side position, bedsharing, and when soft bedding was present.
Table 2

Modifying effect of pacifier use on SIDS risk by individual factors

 

SIDS CASES

CONTROLS

Unadjusted OR (95% CI)

Adjusted OR (95% CI)*

#

Used pacifier (%)

#

Used pacifier (%)

Maternal characteristics

 Maternal age**

<20

68

12 (17.7)

62

18 (29.0)

0.52 (0.23–1.20)

0.49 (0.21–1.13)

20–24

107

17 (15.9)

75

27 (36.0)

0.37 (0.17–0.68)

0.24 (0.11–0.54)

≥25

76

8 (10.5)

106

29 (27.4)

0.31 (0.13–0.73)

0.25 (0.10–0.66)

 Maternal education**

High school or less

221

36 (16.3)

146

54 (37.0)

0.32 (0.20–0.54)

0.33 (0.20–0.55)

Some college

39

3 (7.7)

114

29 (25.4)

0.24 (0.07–0.85)

0.29 (0.08–1.06)

 Marital status**

Married

42

4 (9.5)

89

22 (24.7)

0.32 (0.10–1.0)

0.17 (0.04–0.66)

Not married

218

35 (16.1)

171

61 (35.7)

0.35 (0.21–0.56)

0.40 (0.21–0.60)

 Kessner index**

Adequate

100

9 (9.0)

163

57 (35.0)

0.18 (0.09–0.39)

0.14 (0.06–0.31)

Inadequate/Indeterminate

160

30 (16.1)

97

26 (26.8)

0.63 (0.35–1.15)

0.66 (0.34–1.26)

 Smoking during pregnancy

No

134

17 (12.7)

211

65 (30.8)

0.33 (0.18–0.59)

0.27 (0.15–0.51)

Yes

126

22 (17.5)

49

18 (36.7)

0.36 (0.17–0.76)

0.41 (0.18–0.97)

Infant characteristics

 Race

Not African– American

65

7 (10.8)

65

17 (26.2)

0.34 (0.13–0.89)

0.27 (0.08–0.93)

African–American

195

32 (16.4)

195

66 (33.9)

0.38 (0.24–0.62)

0.35 (0.21–0.60)

 Age

≤1 month

22

2 (9.0)

22

5 (22.7)

0.34 (0.06–1.98)

0.36 (0.05–2.55)

2–3 months

180

28 (15.6)

183

62 (33.9)

0.36 (0.22–0.60)

0.30 (0.17–0.53)

≥4 months

57

8 (14.0)

55

16 (29.1)

0.40 (0.15–1.03)

0.34 (0.12–0.95)

 Ever breastfed

No

205

36 (17.6)

130

50 (38.5)

0.34 (0.21–0.56)

0.35 (0.21–0.60)

Yes

55

3 (5.5)

130

33 (25.4)

0.17 (0.05–0.58)

0.14 (0.04–0.51)

Sleep conditions

 Last sleep position

Back

57

10 (17.5)

87

25 (28.7)

0.53 (0.23–1.21)

0.50 (0.18–1.34)

Side/prone

202

28 (13.9)

172

58 (33.7)

0.32 (0.19–0.53)

0.27 (0.16–0.48)

 Any Bedsharing

No

128

22 (17.2)

180

62 (34.4)

0.40 (0.23–0.69)

0.36 (0.19–0.65)

Yes

132

17 (12.9)

79

21 (26.6)

0.41 (0.20–0.83)

0.30 (0.14–0.68)

 Sleep location

Room share/no bedsharing

42

7 (16.7)

63

21 (33.3)

0.40 (0.15–1.05)

0.33 (0.12–0.95)

Bedshare with parent(s)

112

16 (14.3)

69

17 (24.6)

0.51 (0.24–1.09)

0.39 (0.17–0.94)

Other room

75

13 (17.3)

107

34 (31.8)

0.45 (0.22–0.93)

0.42 (0.19–0.93)

 Soft bedding

No

133

22 (16.5)

210

66 (31.3)

0.43 (0.25–0.74)

0.37 (0.20–0.68)

Yes

127

17 (13.4)

50

17 (34.0)

0.30 (0.14–0.65)

0.25 (0.10–0.63)

 Pillow use

No

192

27 (14.1)

224

71 (31.7)

0.35 (0.22–0.58)

0.28 (0.16–0.48)

Yes

68

12 (17.7)

36

12 (33.3)

0.43 (0.17–1.09)

0.36 (0.13–1.05)

* Adjusted for maternal age, marital status, maternal education, and adequacy of prenatal care (Kessner index)

** Each of these variables is adjusted for the other 3

The final analyses measured the associations between prone sleeping, maternal smoking during pregnancy, bedsharing, and breastfeeding and SIDS, when pacifiers were not used and when they were used (Table 3). Without pacifier use, prone sleep position, maternal smoking, and bedsharing were associated with increased risk for SIDS, while breastfeeding was associated with decreased risk. With pacifier use, the odds ratios for prone sleeping, maternal smoking, and bedsharing all reduced to under 1 (ranging from 0.25 to 0.33). The odds ratio for breastfeeding remained about the same (0.33). However, the differences in the respective odds ratios (i.e., as measured by the interaction between pacifier use and the variables of interest) were not statistically significant.
Table 3

Risk factors for SIDS without and with pacifier use

 

Did not use pacifier

Used pacifier

P value for difference in OR**

 

Case

Control

aOR (95% CI)*

Case

Control

aOR (95% CI)*

Last sleep position

Back/side

91

111

2.15 (1.33–3.46)

19

55

0.35 (0.18–0.67)

0.76 NS

Prone

130

64

19

28

Smoking during pregnancy

No

117

146

3.69 (2.15–6.34)

17

65

0.28 (0.15–0.53)

0.81 NS

Yes

104

31

22

18

Bedsharing***

No

106

118

1.69 (1.01–2.82)

22

62

0.28 (0.15–0.52)

0.55 NS

Yes

115

58

17

21

Ever breastfed

No

169

80

0.32 (0.16–0.64)

36

50

0.35 (0.20–0.60)

0.19 NS

Yes

52

97

3

33

Soft sleep area

No

111

144

2.97 (1.78–4.96)

22

66

0.37 (0.21–0.66)

0.43 NS

Yes

110

33

17

17

Pilllow Use

No

165

153

2.69 (1.49–4.84)

27

71

0.27 (0.16–0.47)

0.26 NS

Yes

56

24

12

12

* Adjusted for maternal age, marital status, maternal education, and adequacy of prenatal care (Kessner index)

** The P value for the interaction between pacifier use and the variable of interest

*** Also adjusted for maternal smoking during pregnancy

Discussion

In the Chicago Infant Mortality Study, a large population-based case–control study, we found pacifier use to be associated with a reduced risk for SIDS. Furthermore, in this population that is demographically at high risk for SIDS, we found that pacifier use may offer additional protection for infants with older, married, non-smoking, breast-feeding mothers who have received adequate prenatal care; all of these characteristics are considered to place an infant at lower risk for SIDS [29]. However, pacifier use may also offer some additional protection against SIDS for infants who sleep in the prone or side position, for bedsharing infants, or when soft bedding is present in the infant’s sleeping environment—all factors that place an infant at higher risk for SIDS.

The finding that pacifier use may be more protective when the infant is in an adverse sleep environment is consistent with the findings of a population based, case control study in a demographically diverse population [27]. As in that study, this current study demonstrates a trend towards increased protection when pacifiers are used in an adverse sleeping environment, but in neither study are the differences in the odds ratios statistically significant. This may be due to inadequate sample size for these sub-analyses.

Although the data used for this study were collected between 1993 and 1996, this analysis continues to be relevant because it is the largest U.S. case–control study in a population that is high risk for SIDS. Furthermore, based on similar key health indicators for Chicago and the U.S. overall, such as infant mortality rates, smoking during pregnancy, and other demographic risk factors for SIDS [34], it is likely that our results are generalizable to other high risk populations in the U.S. The findings from this case–control population have been consistent with results from other SIDS case–control studies conducted during different time periods in the U.S. and around the world.

It is unclear why pacifier use was more protective for those maternal and infant characteristics that place infants at lower risk for SIDS (such as older maternal age and adequate prenatal care), while also being more protective in adverse sleep environments (such as side/prone sleep position and bedsharing). It may be a reflection of as yet undefined factors that predispose to pacifier use that may also affect SIDS risk. It may also be that pacifier use can have a salutary effect with regards to reducing SIDS risk in all infants, no matter what their demographic or environmental risk factor profile is.

The stratified analyses resulted in small sample sizes in some of the analyses, which are reflected by the wide confidence intervals. In addition, even though the confidence intervals for the odds ratios for pacifier use and non-use frequently did not overlap, the interactions between pacifier use and the respective variables were often not significant. This unusual finding is consistent with the non-significant associations found by Li [27] and is likely due to the small number of subjects in these analyses. However, the odds ratios for all analyses with pacifier use consistently trended in the same direction. Finally, because this was a case–control study, one cannot attribute a causal effect between any individual risk factor and SIDS.

In recent years, the incidence of SIDS has seen no further decline [35]. This may be due to multiple factors, but the persistence of specific behavioral risk factors is likely partly responsible. Rates of supine sleep positioning have remained constant in recent years and may be declining [36], bedsharing rates are increasing [37, 38], and there has been no decrease in the use of soft bedding in infants’ sleep environments [39]. This study provides additional epidemiologic evidence for the benefits of pacifier use. Pacifier use may provide an additional strategy to reduce the risk of SIDS for all infants, but its use should be particularly encouraged for those infants who are in adverse sleep environments.

Acknowledgments

We thank Kristen Wells, PhD for statistical assistance. This study was supported by HRSA grant 1R40MC08963-01 and by funding from First Candle.

Copyright information

© Springer Science+Business Media, LLC 2011