Objective. Because breastfeeding is thought to be protective against sudden infant death syndrome (SIDS), environmental or child care factors that promote breastfeeding might reduce infant vulnerability to SIDS. The effect of mother-infant bedsharing on nocturnal breastfeeding behavior was studied in 20 routinely bedsharing and 15 routinely solitary sleeping Latino mother-infant pairs when the infants were 3 to 4 months old.
Methodology. All pairs were healthy and exclusively breastfeeding at night. The videotape portion of all-night laboratory polysomnographic studies was used for the analyses. For each pair, an adaptation night was followed by one night each of bedsharing and solitary sleeping.
Results. The most important finding is that when tested in their usual sleeping conditions, routinely bedsharing infants breastfed approximately three times longer during the night than infants who routinely slept separately: this reflected a two-fold increase in the number of breastfeeding episodes and 39% longer episodes. Breastfeeding was also facilitated on the bedsharing night relative to the solitary night within the routinely bedsharing group: the number and total duration of breastfeeding episodes were significantly larger on the bedsharing night.
Conclusions. We suggest that, by increasing breastfeeding, bedsharing might be protective against SIDS, at least in some contexts. Furthermore, maternal reproductive physiology could be impacted because nursing frequency affects ovulation. This is the first study to directly measure nocturnal breastfeeding behavior in any cultural group.
From an evolutionary perspective, breastfeeding and parent-infant co-sleeping constituted an integrated system throughout human history in which both the mother's and infant's sleep physiology were entwined in adaptive ways. These child care practices were probably designed by natural selection to maximize the chances of infant survival and parental reproductive success,1-3and even today they remain inseparable and inevitable for the vast majority of the world's societies.4,5 Only in the last 100 to 200 years, and mostly in western industrialized societies, has nocturnal breastfeeding been disassociated from the co-sleeping environment within which it evolved.
The potential impact of this relatively sudden change in child care on infant development and survival and on maternal reproductive physiology has not yet been addressed. Understandably, the need for privacy after retiring for bed makes collection of reliable data on nocturnal breastfeeding particularly difficult, regardless of the type of sleeping arrangement. In fact, most breastfeeding data collected in both diverse nonwestern and western ethnographic settings emerge from either diurnal observations or from structured interviews or daily feeding diaries that rely on maternal recall.6-8Unfortunately, maternal recall has not proven to be very reliable.9 For example, Vitzthum's8 field studies of breastfeeding among Peruvian women found that mothers significantly underestimated the frequency of breastfeeding while consistently overestimating its duration, compared with the researcher's direct observations.
Because there exists near universal agreement that increased breastfeeding reduces infant morbidity and mortality worldwide,10-12 including in some populations the infant's chances of dying of sudden infant death syndrome (SIDS),13-15 it is unfortunate that clinical, experimental, and ethnographic studies have not addressed the role that sleeping arrangements play in either promoting or inhibiting this practice. The absence of research in this area may be explained partially by the cultural context within which pediatric research unfolds. Western cultural values clearly favor early weaning and solitary infant sleeping arrangements believed to help promote infant autonomy as early in life as possible.16,17 For example, Pinilla and Birch18 devised a behavioral strategy by which starting early in infancy mothers can reduce or eliminate nocturnal feeds, thereby minimizing the need for nighttime parental interventions. Moreover, research in the area of infant sleeping arrangements has focused on child care strategies that accord with parental work schedules and preferences, without scientific determination of whether decreased nocturnal parent-infant contact represents any disadvantages for infants.19 Solitary infant sleeping and early weaning presumably improve the mother's sleep, but possible biological trade-offs for the infant, whose social and biological interests are different from the mother's, remain unexplored.
With the unprecedented worldwide declines in SIDS rates after recent public campaigns against the prone sleeping position, infant care practices, including breastfeeding and bedsharing, are being scrutinized in ways not previously anticipated. However, the complexity of infant care practices and the diversity of approaches used to study them have led to inconsistent and/or contradictory perspectives and findings. For example, not every epidemiologic study shows that breastfeeding provides increased protection from SIDS,20but a serious problem in comparing findings is that the type, frequency, and duration of breastfeeding are not always differentiated or categorized in the same way.21,22 Also, results from an epidemiologic study from New Zealand suggest that in some populations bedsharing in conjunction with maternal smoking increases SIDS risk.23 In contrast, we have argued from an evolutionary perspective that, under otherwise safe conditions, bedsharing and other forms of co-sleeping may offer infants an adaptive advantage and be protective against SIDS.24,25 The apparent diversity of co-sleeping environments, within as well as between cultures, makes comparisons between existing studies challenging at best.
As a beginning step toward disentangling some of these complex issues, the present study was designed to quantify nocturnal breastfeeding behavior in routinely bedsharing (RB) and routinely solitary sleeping (RS) Latino mother-infant pairs, in both bedsharing and solitary sleeping laboratory environments. This is the first study to directly measure nocturnal breastfeeding in any cultural group.
Results are presented for the first 35 mother-infant pairs in an ongoing behavioral and polysomnographic study. Twenty were RB and 15 were RS since birth. RB was defined as bedsharing with the mother for at least 4 hours per night, on at least 5 days per week; RS was defined as bedsharing no more than 1 night per week for any part of the night. Two-week daily sleep logs were completed at home just before the sleep recordings to confirm maternal reports of the infants' usual sleep environment. For the 33 pairs who completed all 14 nights of the log, the mean (±SD) number of bedsharing nights was 13.7 ± .5 for the RB group versus .6 ± .9 for the RS group. Fathers were excluded from the study protocol, regardless of whether or not they participated in infant bedsharing, to isolate the influence of maternal-infant bedsharing on breastfeeding.
Subjects were recruited from the Birthing Center at the University of California Irvine Medical Center. All mother-infant pairs visiting the Birthing Center for postpartum follow-up who met our inclusion criteria were asked to participate. Inclusion criteria for mothers were that they: be Latino; be <38 years old; be exclusively or predominantly breastfeeding (no more than two 4-ounce bottles of formula per day and none after 3 pm); have had prenatal care; have no present or past history of drug or alcohol abuse; have no history of smoking, alcohol, or illicit drug use during pregnancy; have had uncomplicated pregnancies, labors, and deliveries; be in good health and free of sleep disorders; be taking no medications known to affect sleep pattern; and have chosen sleeping practice for reasons other than infant temperament (eg, response to a fussy infant). A physician trained in sleep medicine performed the sleep histories. RB mothers were 27.0 ± 5.9 (SD) years of age, and RS mothers were aged 24.3 ± 8.5 years.
Inclusion criteria for infants were that they: be 11 to 15 weeks old at the time of the sleep studies; be in good health, with normal growth and development; weighed >2500 g and were >37 weeks' gestational age at delivery; had a 5-minute Apgar score ≥8; have no history of SIDS in first degree relatives; and no history of prolonged apnea or an apparent life-threatening event. RB infants were 11 boys and 9 girls, aged 13.0 ± 1.3 weeks when sleep testing was performed; RS infants were 4 boys and 11 girls and were 12.9 ± 1.3 weeks old.
Procedure and Data Analysis
Mother-infant pairs underwent 3 consecutive nights of laboratory study: an initial adaptation night matching the routine home sleeping arrangement followed by a bedsharing night (BN) and a solitary sleeping night (SN) in randomly assigned order. Sleep studies were performed in the University Medical Center Sleep Disorders Center, a facility accredited by the American Sleep Disorders Association. For solitary sleeping, infants slept in a standard crib in the room adjacent to the mother's (in hearing range) with the doors between them ajar. On the bedsharing night(s), mother-infant pairs shared the same twin-size bed used by the mother for solitary sleeping. Infants were maintained on their usual feeding and sleeping schedules, with mothers performing all care taker interventions ad lib. Mothers were blind to all experimental hypotheses and instructed only to prepare and respond to their infants as they would at home. Mothers also retired at their usual times, and monitoring was terminated after mother and infant had awakened the next morning at their usual times.
Monitoring each night in infants and mothers included continuous infrared audiovisual recordings as well as the standard, noninvasive polysomnographic measures required by the larger ongoing study (including electroencephalograms, electrooculograms, and surface chin electromyogram for determination of sleep-wake stages26). A large digital clock placed in the camera field allowed calculation of breastfeeding data to the nearest minute.
Breastfeeding was defined as nipple attachment, which was usually verifiable through observation of the video recordings. Therefore, identification of breastfeeding did not rely on infant sucking behavior, with the exception of the few occasions when the infant's head was obscured by blankets; in those instances the sucking artifact, which is readily seen on the chin electromyogram, was used to identify breastfeeding.
Breastfeeding behavior was quantified as breastfeeding episodes. Breastfeeding episodes began and ended with nipple attachment and detachment, respectively, but also included very short interruptions during which the mother switched from one breast to the other. These interruptions were a matter of seconds and were below the 1-minute resolution of the digital clock. All breastfeeding episodes were initiated by the mother placing her nipple in the infant's mouth and defined to capture a single, intentional, and continuous act of breastfeeding on the mother's part. A new breastfeeding episode was scored if breastfeeding was interrupted by maternal behaviors that indicated an apparent attempt by the mother to terminate feeding (ie, by closing her bra or nightgown, by attempting to leave the room, or by attempting to return to sleep) but the infant's subsequent refusal to settle prompted the mother to reinitiate feeding.
Three breastfeeding variables were computed each night: number, mean duration, and total duration (sum) of breastfeeding episodes. Analyses were restricted to the time the infants were in bed each night, ie, from lights out to final morning awakening. Within-group comparisons of the BN and SN were performed using the Wilcoxon matched-pairs signed rank test. The Mann-Whitney U test was used for between group (RB versus RS) comparisons. Significance was assigned whenP < .05.
This protocol was approved by the University's Human Subjects Review Committee, and mothers were remunerated for their participation.
The group means (±SD) for the three breastfeeding variables each night are given in the Table. For the RB group, both the number (P = .006) and total duration of breastfeeding episodes (P < .001) were significantly greater on the BN than the SN. The mean duration of breastfeeding episodes was also larger on the BN but the difference just failed to reach statistical significance (P= .050). Because no significant differences were found in either the infants' time in bed (mean ± SD = 544.2 ± 40.0 minutes on BN vs 545.2 ± 39.3 minutes on SN) or total sleep time (461.0 ± 38.9 minutes on BN vs 442.5 ± 52.9 minutes on SN) on the two nights (P < .05), the differences in breastfeeding could not be attributed to different length sampling periods. The net result was that RB infants spent more than twice as long breastfeeding on the BN than on the SN.
All three breastfeeding variables were also larger in the RS group on the BN than on the SN, but none of the differences reached significance. Similarly, comparing the two groups on the BN, the three breastfeeding variables were all larger in the RB group than in the RS group; however, only the difference in total duration of breastfeeding episodes was significant (P = .028).
The largest differences found in breastfeeding were consistently between the two groups in their routine conditions. Comparing the RB group on the BN to the RS group on the SN, the number of breastfeeding episodes was doubled (P < .001), the total duration of breastfeeding episodes was nearly three times as great (P < .001), and the mean duration of breastfeeding episodes was 39% greater (P = .039) in the RB group. A modestly (6%) longer total bed time (P = .039) in RB infants on the BN (544.2 ± 40.0 minutes) than for RS infants on the SN (512.7 ± 49.4 minutes) was insufficient to explain the large differences seen in breastfeeding.
We examined any possible role of infant sex in the above findings by partitioning the three main variables by sex: frequency, mean, and total duration of breastfeeding. For both the BN and SN, no significant differences were found for any variable between male and female infants in either RB or RS groups. This was also true when the two groups were combined.
The number of self-reported nocturnal breastfeeding episodes was also derived from the 2-week home sleep logs. RB pairs averaged 2.4 ± .9 feeds per night compared with 1.6 ± .7 per night in RS pairs, a significant difference (P = .009).
The most important finding of this study is that infants who bedshared routinely at home breastfed three times longer during the night than infants who routinely slept separately, when tested in their routine conditions. This largely reflected a two-fold increase in the number of breastfeeding episodes although the duration of individual episodes was, on average, also longer by 39%. These differences are only partly explained by acute or immediate effects of bedsharing because total breastfeeding time was significantly larger in the RB group than the RS group on the BN. This suggests that routine bedsharing promotes more nocturnal breastfeeding through additional factors not present during occasional bedsharing. Because diurnal breastfeeding was not measured in this study, we do not know if daytime feeding pattern is also affected by routine bedsharing at night. Available data on diurnal breastfeeding, however, provide no evidence that daytime feeding is diminished by breastfeeding at night.7,27 Furthermore, how the results might be modified when a father or other family member is also present during bedsharing with the mother and infant is unknown.
Consistent with the laboratory findings, the sleep logs revealed that RB pairs were also breastfeeding more often than RS pairs at night at home. This indicates that the laboratory findings are, in general, representative of behavior in the natural environment. However, the number of breast feeds reported at home was approximately half that measured in the laboratory. Although this could reflect that infants actually fed more often in the laboratory, the more likely explanations are that mothers used a different criterion for defining a single feeding episode or that they underestimated the true frequency. Regarding the latter, there is other evidence that mothers underestimate the frequency of breastfeeding, compared with researcher observations.8,9 Whatever the reason(s) for the discrepancy, it serves to emphasize the need for objective measures in quantifying breastfeeding behavior.
Although the mechanisms through which routine bedsharing promotes breastfeeding have yet to be determined, there is evidence that olfactory cues could be important. By 2 weeks of age, breastfed infants preferentially orient (by head turning) toward odors from their mother's breast and axillary regions during sleep as well as waking.28,29 Consistent with some sort of sensory bridge between bedsharing infants and their mothers, we have found that bedsharing infants remain oriented toward their mothers for large portions of the night, up to 100%.30 Inherent to bedsharing, the infant's proximity to the mother's breast would enhance the infant's exposure to olfactory and any other relevant sensory cues from the mother that might facilitate breastfeeding. One can further speculate that exposure to the mother's odors during bedsharing might have the effect of lowering the infant's arousal/hunger threshold. Bedsharing might also promote breastfeeding by permitting mothers to sense their infant's more subtle sounds and movements that increase in both frequency and intensity as breastfeeding episodes are approached (J. J. McKenna, unpublished observation, 1996). Because this kind of maternal responsiveness is not possible when infants sleep in a separate room, only by frank crying can the infant elicit maternal attention and feeding in the solitary condition.
That bedsharing promotes increased breastfeeding could have important implications for infant health. The most visible effect might be on weight gain (not measured in our study). Relatedly, Uvnas-Moberg et al31 reported that increased stimulation of the intraoral cavity of the newborn leads to an increased release of glucose and gastrointestinal food absorption hormones in newborns, and presumably in older infants too. In addition, Barr et al32 showed that more frequent infant feedings, especially those containing sugar nutrients (as opposed to water), reduce crying duration, thereby contributing to energy conservation and calm wakefulness presumably through both preabsorptive (orotactile and orogustatory) and postabsorptive mechanisms. Consistent with the latter, we observed that on the BN infants rarely cried to elicit a maternal response leading to a breastfeeding episode, whereas they almost always did so to elicit a feed on the SN (J. J. McKenna, unpublished observation, 1996).
An increased daily infusion of maternal antibodies at a time when their own immunological systems are least efficient may provide bedsharing/breastfeeding infants with increased protection from infectious diseases, some potentially related to SIDS. According to Arnon,33 breastfeeding is important for protection against infant botulism that may be misdiagnosed as SIDS approximately 5% of the time. Bacterial infections also have been hypothesized by Blackwell et al34 to interact with a range of infantile deficits to cause some SIDS deaths, and these too might be reduced by the increased breastfeeding associated with bedsharing. Arnon33 suggests that the degree of protection from SIDS provided by mother's milk depends on the baby receiving greater amounts of secretory immunoglobulin A. If one assumes that longer total duration of breastfeeding equates with more milk intake, then the additional intake (however small) might help compensate for the natural variability of titers of antibodies that distinguish the content of one mother's milk from another's.33
Two recent epidemiologic studies suggest that breastfeeding lowers the risks of SIDS,14,15 and two others suggest that the extent of protection may be dose-related.13,35 Whereas a protective effect has not been found in every study,20 many international SIDS prevention campaigns, including in the United States, recommend breastfeeding as a way to reduce risk for SIDS. That bedsharing is associated with increased breastfeeding raises the possibility that bedsharing is a factor in the relatively lower rates of SIDS among the population from which our subjects were drawn. Our Latino mothers and infants represent a cultural group living in the Western and Southwestern United States for which their low socioeconomic and recent immigrant status suggest increased SIDS risk. Yet, similar to recent Asian immigrants in Great Britain,36 the most recent epidemiologic data show that California Latino infants of Mexican-born mothers are at lower risk (.8/1000 live births) compared with Anglos (1/1000);37 and for infants weighing more than 2500 g, this is also true among Latinos living in the Chicago area where the SIDS rates range from 1.0/1000 for Latino infants to 1.2/1000 for white middle-class infants.38 That breastfeeding and bedsharing are more commonly practiced among recent Latino immigrants than in white American families may explain some of this discrepancy. Eighty-four percent of Mexican-born mothers in the United States initiate breastfeeding in contrast to 58.5% of white American mothers (Demographic and Health Surveys, 1991, unpublished data, Institute for Resource Development, Westinghouse Corporation).39 Moreover, the incidence of all night bedsharing was found to be 21% among Latino families in east Harlem, New York City, significantly higher than the 6% reported for a representative sample of white, middle-class urban infants matched for age and sex.40 That maternal smoking, a known risk factor for SIDS,41 may be relatively uncommon in Latino women may also be contributory. Furthermore, we found that bedsharing minimized use of the prone sleep position in our Latino subjects,30which could also be a factor in the low SIDS rates reported for southern California Latinos because prone sleeping is a known risk factor for SIDS.42
Although bedsharing is most often considered in terms of impact on infants, maternal reproductive physiology, particularly ability to reestablish ovulation, may also be affected. Suppression of ovulation contributes to increased iron status by delaying blood loss associated with menstruation and also extends the recuperative period between pregnancies, permitting the mother's body to replenish fat and nutrient stores required by herself as well as future offspring.21,43 Recent studies suggest that frequent nipple contact or sucking and relatively short intervals between breast feeds (rather than simply breastfeeding versus not breastfeeding) are required to sustain the high levels of circulating prolactin which are known to suppress ovulation, increase birth interval, and thus limit maternal fecundity.27,44 Although the complex interplay of behavior and physiology are yet to be deciphered, sucking duration was found to be important in prolonging lactational amenorrhea when sucking frequency is low, but less critical when sucking occurs 10 to 20 times per day and in which intervals between feeds are relatively short.45 Although we do not know the pattern of our Latino mother's breastfeeding during the day, the more frequent nocturnal breast feeds, the shorter intervals between feeds and the longer total nightly durations occurring with routine bedsharing certainly reveal a nocturnal breastfeeding structure consistent with that known to suppress ovulation among women living in diverse cultural settings.6,27,44 One can speculate that mothers who bedshare may be more likely to become pregnant later, rather than sooner, permitting their infants to benefit longer from less divided maternal attention and resource allocation,46 thereby improving the mother's as well as the infant's prospects of better psychosocial and physical health.47
In summary, these findings on nocturnal breastfeeding behavior are the first based on direct observation, instead of maternal recall, and the first to include the bedsharing environment. The discovery that bedsharing promotes breastfeeding has important implications both for infant and maternal health. Perhaps most important is that bedsharing might reduce the risk of SIDS because breastfeeding is thought to be protective against SIDS for some populations. However, given that bedsharing may increase SIDS risk when practiced in combination with maternal smoking,23 the potential benefits of bedsharing cannot be considered independently of the specific social and physical circumstances within which bedsharing occurs. The potential for overlying or falling out of bed should be considered, in addition to the potential dangers associated with modern sleeping surfaces such as waterbeds, soft, fibrous, or bean mattresses, or loose bedding.48,49 In older infants who continue to breastfeed, the possibility of promoting dental caries should also be considered.
Finally, our study has important implications for appreciating the potentially important effects that cultural choice of infant care practice has on the mother-infant relationship in general and on maternal and infant biology in particular. The differences we report between RB and RS mother-infant pairs underscore the need to reconceptualize normal mother-infant behavior, not in cultural terms alone, but also in biological terms. These findings reinforce the idea that mother and infant breastfeeding behaviors are highly sensitive to the environment within which they find expression and that expression is maximized when in the co-sleeping environment within which breastfeeding evolved.
This research was funded by National Institute of Child Health and Human Development grant R01 27482.
The authors wish to express their appreciation to Sean Drummond and Naz Kajani for their excellent technical assistance.
- Received February 12, 1996.
- Accepted June 7, 1996.
Reprint requests to (J.J.M.) Department of Anthropology, University of Notre Dame, Notre Dame, IN 46556.
- SIDS =
- sudden infant death syndrome •
- RB =
- routinely bedsharing •
- RS =
- routinely solitary sleeping •
- BN =
- bedsharing night •
- SN =
- solitary sleeping night
- ↵Konner M. Evolution of human behavior development. In: Munroe RH, Munroe RL, Whiting JM, eds. Handbook of Cross-Cultural Human Development. New York, NY: Garland STPM Press; 1981:3–52
- ↵Whiting JWM. Environmental constraints on infant care practices. In: Munroe RH, Munroe RL, Whiting JM, eds. Handbook of Cross-Cultural Human Development. New York, NY: Garland STPM Press; 1981:155–179
- McKenna JJ
- ↵Whiting J. Effect of climate on certain cultural practices. In: Goodenough W, ed. Explorations In Cultural Anthropology: Essays In Honor of George Peter Murdoch. New York, NY: McGraw-Hill; 1964:117–139
- ↵LeVine R, Dixon S, LeVine S, et al. Child Care and Culture. Lessons From Africa. Cambridge, England: Cambridge University Press; 1994
- Konner MJ,
- Worthman C
- ↵Elias MF, Nicholson N, Konner M. Two subcultures of maternal care in the United States. In: Taub D, King F, eds. Current Perspectives in Primate Social Dynamics. New York, NY: Von Nostrand Reinhold; 1986:37–50
- Quandt SA
- ↵Cunningham AS. Breastfeeding: adaptive behavior for child health and longevity. In: Stuart-Macadem P, Dettwyler K, eds. Breastfeeding: Biocultural Perspectives. New York, NY: Aldine de Gruyer; 1995:243–264
- ↵Jeliffe DB, Jeliffe EFP. Human Milk in the Modern World. Oxford, England: Oxford University Press; 1978
- Fredrickson DD,
- Sorenson JF,
- Biddle AK,
- et al.
- Mitchell EA,
- Taylor BJ,
- Ford RPK,
- et al.
- ↵Hoffman H, Damus K, Hillman L, et al. Risk factors for SIDS: results of the National Institute of Child Health and Human Development SIDS cooperative epidemiological study. In: Schultz P, Southhall D, Valdes-Dapena M, eds. Sudden Infant Death Syndrome: Cardiac and Respiratory Mechanisms and Interventions. New York, NY: New York Academy of Sciences; 1988;533:13–37
- ↵Ferber R. Solve Your Child's Sleep Problem. New York, NY: Simon and Schuster; 1986
- ↵Spock B, Rothenberg M. Dr. Spock's Baby and Child Care. New York, NY: Pocket Books; 1985
- Pinilla T,
- Birch LL
- ↵Woolridge M. Baby-controlled breast-feeding: biocultural implications. In: Macadem PS, Dettwyler K, eds. Breastfeeding: Biocultural Perspectives. New York, NY: Aldine de Gruyer; 1994:217–230
- Gilbert RE,
- Wigfield RE,
- Fleming PJ,
- et al.
- ↵Dettwyler K. Infant feeding practices and growth. Annu Rev Anthropol. 1992;:21:171–204
- ↵McKenna JJ. The potential benefits of infant-parent cosleeping in relation to SIDS prevention: overview and critique of epidemiological bed sharing studies. In: Rognum TO, ed. Sudden Infant Death Syndrome. New Trends in the Nineties. Oslo, Norway: Scandinavian University Press; 1995:256–265
- ↵Guilleminault C, Souquet M. Sleep states and related pathology. In: Korobkin R, Guilleminault C, eds. Advances in Perinatal Neurology, I. New York, NY: SP Medical and Scientific Books; 1979
- Vitzthum VJ
- Richard C,
- Mosko SS,
- McKenna J,
- Drummond S
- ↵Arnon SS. Breast-feeding and toxigentic intestinal infections: missing links in SIDS. In: Tildon JT, Roeder LM, Steinschneider A, eds. Sudden Infant Death Syndrome. New York, NY: Academic Press; 1983:539–549
- ↵Jura J, Olejar V, Dluholucky S. Epidemiological risk factors of SIDS in Slovakia, 1993, 1994. Program and Abstracts Third SIDS International Conference, Stavanger, Norway July 31-August 4. Oslo, Norway: Holstad Grafisk; 1994;98. Abstract
- Gantly M,
- Davies DP,
- Murcott A
- Ryan AD,
- Rush D,
- Krieger F,
- Lewandowski G
- Schachter FF,
- Fuchs ML,
- Bijur PE,
- Stone RK
- ↵Mitchell EA. Smoking: the next modifiable risk factor. In: Rognum TO, ed. Sudden Infant Death Syndrome: New Trends In the Nineties. Oslo, Norway: Scandinavian University Press; 1994:114–118
- Howie PW,
- McNeilly AS
- McNeilly AS,
- Glasier A,
- Howie PW
- Copyright © 1997 American Academy of Pediatrics