TABLE 1

Studies Included in the Meta-analysis

Study (Year)CountryYears of StudyTotal Cases, NTotal Controls, NBreastfeeding Cases, n (%)Breastfeeding Controls, n (%)Crude OR (95% CI)CovariatesaTime to InterviewbFailed Criteria
Any breastfeeding
    Bartholomew and MacArthur24 (1988)ScotlandUnknown797915 (19)25 (32)0.51 (0.23–1.13)2 wk after deathNone
    Naeye et al25 (1976)United States1952–196612537516 (13)60 (16)0.77 (0.41–1.44)No interviewNone
    Biering-Sørensen et al15 (1978)Denmark1956–197112352098 (75)480 (92)0.33 (0.18–0.59)No interviewNone
    Steele and Langworth26 (1966)Canada1960–19618015721 (26)75 (48)0.39 (0.21–0.73)NAcNone
    Protestos et al27 (1973)United Kingdom1960–19729413522 (23)71 (53)0.28 (0.15–0.52)No interviewNone
    Murphy et al28 (1982)United Kingdom1965–1973, 1975–19779947 22316 (16)14 223 (30)0.54 (0.29–1.01)NANone
    Grice and McGlashan29 (1981)Tasmania1970–197612115353 (44)101 (66)0.40 (0.24–0.67)No interviewNone
    McGlashan30 (1989)Tasmania1980–1986167334115 (69)252 (75)0.72 (0.47–1.11)Up to 5 wk after deathNone
    Ford et al6 (1993)New Zealand1987–19903561529275 (77)1371 (90)0.39 (0.29–0.53)Within 1 mo of deathNone
    Gilbert et al14 (1995)United Kingdom1987–1989, 1990–19919819656 (57)144 (73)0.48 (0.27–0.84)2 interviews; immediately and at 2–3 moNone
    Ponsonby et al18 (1995)Tasmania1988–19915812022 (38)63 (53)0.50 (0.26–0.98)1, 46 wk after deathNone
    Klonoff-Cohen and Edelstein31 (1995)United States1989–1992200200114 (57)151 (76)0.43 (0.27–0.68)3, 4, 6, 13, 17NANone
    Mitchell et al5 (1997)New Zealand1991–199312091898 (82)809 (88)0.60 (0.35–1.03)1–6, 7, 9, 11, 12, 19NANone
    Wennergren et al10 (1997)Denmark, Norway, Sweden1992–1995244863184 (75)729 (84)0.59 (0.41–0.85)4, 6, 9NANone
    Schellscheidt et al11 (1997)Germany1993–19945815629 (50)129 (83)0.21 (0.10–0.44)4, 6, 15, 16, 19Within 2 wkNone
    Fleming et al17 (1996)United Kingdom1993–199519578088 (45)470 (60)0.50 (0.35–0.71)1–7, 9, 10, 14, 16, 18, 192 interviews; within 5 d and 2 wk of deathNone
    Hauck et al8 (2003)United States1993–199626026055 (21)130 (50)0.20 (0.12–0.35)1, 5, 11, 132 wk after deathNone
    Vennemann et al7 (2009)Germany1998–2001333998165 (50)827 (83)0.19 (0.14–0.25)1–7, 15, 16, 181 mo after deathNone
Breastfeeding ≥2 mo
    Biering-Sørensen et al15 (1978)Denmark1956–19719750322 (23)278 (55)0.24 (0.14–0.41)No interviewNone
    Mitchell et al5 (1997)New Zealand1991–19936477846 (72)600 (77)0.76 (0.41–1.39)1–6, 7, 9, 11, 12, 19NANone
    Schellscheidt et al11 (1997)Germany1993–1994581567 (12)50 (32)0.29 (0.11–0.74)4, 6, 15, 16, 19Within 2 wk of deathNone
Studies excluded for not meeting eligibility criteria
    Fedrick31 (1974)United Kingdom1966–197015440963 (37)157 (31)1.11 (0.75–1.65)No interview1–3
    Watson et al16 (1981)United Kingdom1975–1979308236164 (53)161 (68)0.53 (0.37–0.76)2 interviews; immediately and 3 wk later1–3
    Chen and Rogan2 (2004)United States19885917740187 (32)3073 (40)0.70 (0.59–0.84)1, 3, 4, 5, 9, 12, 19NA1–4
    Stray-Pedersen et al19 (2005)Norway1989–2003237216 (73)65 (90)0.29 (0.09–0.97)4, 71 mo after death4, 5
    Jonville-Béra et al9 (2001)France1995–199711134137 (33)160 (47)0.57 (0.35–0.92)3, 4, 6, 9, 14, 17, 192 interviews; day of death and 3 mo after death2
    Alm et al33 (2002)Denmark, Norway, Sweden1992–1995239841109 (46)626 (74)0.29 (0.21–0.39)4–6, 8NANonec
  • Studies were scored on the following criteria: (1) an appropriate definition for SIDS; (2) autopsies performed in >98% of cases; (3) an adequate description of SIDS ascertainment in the study population; (4) matched control subjects; (5) an adequate description of the process of control selection; and (6) inclusion of sufficient data to calculate ORs and 95% CIs or the actual ORs and CIs were provided. NA indicates not available, not provided.

  • a Covariates: 1, maternal age; 2, parity; 3, birth weight; 4, infant exposure to tobacco smoke (before or after delivery); 5, factors related to socioeconomic status; 6, infant sleep position; 7, bed-sharing; 8, infant age; 9, infant gender; 10, gestation; 11, marital status; 12, race/ethnicity; 13, factors related to prenatal care; 14, factors relating to surface on which infant was placed; 15, pillow/cushion use; 16, factors related to overheating; 17, postneonatal infant health problems; 18, pacifier use; 19, other.

  • b Time to interview was defined as time from infant's death or identification of controls to interview with parents.

  • c ORs in the article were provided for intervals of breastfeeding duration; thus, we could not identify an OR to use in calculating the SOR for this meta-analysis.