Infant Mechanical Suffocation Deaths in the United States, 1980–1997

Wedging

Wedging was primarily a pattern among the age group 3 to <7 months. Three- to 6-month-old infants have begun to develop several motor skills. They usually can rock from side to side and roll over. They may be able to crouch up on hands and knees and move by pushing against a flat surface. Thus, they have the capability to move to corners in beds and cribs. Yet, they do not have the muscle development or motor control to be able to lift themselves out of a wedged position. More than half of the wedgings involved a bed, including 32% wedged between the bed and wall. Wedging involving a bed is not a new pattern of suffocation,^{10 ,11} nor is it a pattern that occurs only in the United States. Wedging suffocation deaths attributed to slipping between a mattress and bed sides or wall have been reported in South Australia¹² and in England and Wales.⁹ In the latter study, beds and cots were implicated in nearly one third of accidental asphyxia deaths among children. In 1977, Smialek suggested that economic reasons or temporary-use circumstances might be factors in the use of a bed for an infant.¹¹ Those still may be valid reasons today. The positioning of a bed close to a wall may be a physical necessity or it may be the result of parents' belief that pushing a bed against a wall prevents a child from falling out. In 1976, Sturner and associates suggested that parents be educated as to the danger of poorly positioned beds;¹⁰ however, the fact that suffocation deaths involving a bed and wall continue to occur indicates that the message has not been communicated successfully.

Approximately 22% of wedgings occurred in a crib. This suggests improper fit of a mattress or lack of structural integrity of the crib frame. The federal crib standard⁸ requires that a mattress fit within a crib frame so that no more than a 1-inch space can be created between the mattress and any side. However, cribs may be used generationally, so that some old cribs in use may not meet the standard, whereas others may have met the standard when first manufactured, but may fail to meet it now because of wear and tear effects on structural integrity or improper size of replacement mattresses. The information from the 38 crib investigations (Table 3) supports these possibilities. In most cases, the age of the crib was unknown. Several were described as hand-me-downs and were likely to be quite old, prestandard cribs. Compromised structural integrity (missing or loose hardware or slats etc) suggested the effect of wear and tear. Improper mattress fit suggested the effect of age, the original mattress having been replaced. For that subset of 38, only five cribs were known to be subject to the federal regulation. Of those five, the compliance status of one was unknown. The other four complied theoretically with the regulation when new; however, the compliance status at the time of the incident was likely to have been compromised. Two cribs were reported damaged in moving, one had been assembled improperly, and one had been used with a previous child.

Oronasal Obstruction

Oronasal obstruction was the second leading cause of suffocation death. Infants in the age groups 0 to <3 and 3 to <7 months of age made up 80% of the cases. Up to 2-month-olds exhibit largely reflexive actions, whereas 3- to 6-month-olds, relatively more developed, display more oral and manual exploratory behavior and more motor ability. Because bedding, pillows, and plastic bags were involved equally in the oronasal obstruction deaths of the 0- to <3-month-olds, rebreathing may play a role in their suffocation. Chiodini and Thatch conducted an observational study of 11 infants 0.2 to 6 months of age and reported that all infants slept face down at some point, either spontaneously or after positioning, and that all infants regardless of age could turn the head out of the face down position.¹³ They concluded that although capable of turning the head, an infant may retain the face down position, in which rebreathing usually occurs. Alternative possibilities are that infants' head-turning is interfered with by the presence of bedding or pillows or that infants may turn the head into a physical barrier made of bedding or pillows. There is insufficient detail in the data used for the present study either to support or to refute these alternative theories. However, from a developmental point of view, they are plausible because this age group would be unable to retreat from or move a barrier.

More than half of the oronasal obstructions among the age group 3 to <7 months of age involved plastic bags. This suggests a more deliberate, exploratory interaction with the environment than the younger infants exhibit. Three- to 6-month-olds can swipe at and manipulate objects; however, this ability is coupled with fledgling motor skills that are insufficient to enable the infant to extricate him/herself from danger. Thus, infants of this age can bring plastic to the face, but may be unable to remove it. They also may fall from elevated surfaces onto filled plastic bags, as reported in two follow-up investigations, and be unable to move away.

Plastic bag dangers have been recognized since the 1950s, when plastic drycleaning bags used as waterproof sheets for infant beds were reported as the principal cause of mechanical suffocation death.¹⁴ The inherent dangers of plastic film are 1) that it adheres, partially by the act of inhaling and partially by static electricity generated by the movement of the plastic film itself; and 2) it is strong, thus, infants cannot tear it.¹⁵Jeffrey¹⁵ reported that if plastic film is not removed from an infant's face within 1 minute, the infant will die of suffocation. The causal association between plastic bags and mechanical suffocation death led to efforts by many groups, including the American Medical Association, the National Institute of Drycleaning, and the Society of the Plastics Industry, Inc, to publicize and warn about the hazard. Over the years, warning labels have been required by some jurisdictions, such as New York and Chicago, but there are no national standards or regulations requiring safety labels on plastic bags. The Society of the Plastics Industry, Inc periodically reminds the industry about the importance of continuing its warning labeling and similar activity;^{16 ,17} however, the content of its informational bulletin has not changed since 1982. The bulletin refers to the tragic plastic bag suffocation deaths that occurred during the 1950s as a “cause celebre” and states, “The danger of misuse of ultra-thin dry cleaner bags has probably become as much a part American child safety folklore as the proverbial warning to children not to touch a hot stove. Even so, it is best to take nothing for granted… so let's make sure those warnings are on all dry cleaner bags—education is still the answer here!” As this study shows, plastic bags continue to be involved with a substantial number of infant suffocations, on average one a month for the past 16 years. Perhaps it is time to revisit the plastic bag hazard and identify an alternative to education/warning. Because other plastic materials, such as plastic wrap, trash bags, toys, and cups, also are involved, albeit in small numbers, it may be useful to revisit the hazard of plastic materials in general. In a 1981 report, CPSC identified infant suffocation cases involving trash bags and other plastic materials thicker than 1 mm and indicated that those findings might suggest that any new campaigns include these thicker materials.¹⁸ We are unaware that any consumer education efforts have included these other plastics.

Overlying

Seventy percent of infants overlain were younger than 3 months old. More than half of the deaths by this pattern occurred in a bed. There may be several factors that place this age group at particular risk. These infants are the smallest and have not yet developed much motor ability. They are engaged primarily in eating and sleeping. New parents may take their infants to bed with them to enhance infant–parent bonding or for feeding convenience.

Overlying historically has been recognized as a cause of infant suffocation death.^{19 ,20} In 1970, Francisco recommended that “Parents must be informed that to sleep in the same bed with an infant is to invite disaster… ”.¹⁹ However, suffocation by overlying became a largely unrecognized entity, superseded by sudden infant death syndrome (SIDS).^{9 ,20} To the contrary, the data in this study suggest that overlying increasingly presents a problem, whether in a bed or on a sofa. The estimated mortality rate for 1986–1990 was more than triple the estimated rate for 1980–1985; the estimated rate for 1991–1995 was nearly twice the rate for 1986–1990. A possible explanation for the increase could be an increase in the rate of infant–parent co-sleeping related to reported benefits, including increased breastfeeding and reduction in the rate of SIDS,^{21 ,22} despite the statement by the American Academy of Pediatrics that there is no basis for encouraging bed-sharing as a strategy to reduce SIDS.²³Knowledge of rates of co-sleeping would have been helpful in understanding the potential causal relationship to overlying.

Other Patterns

Entrapment with suspension, hanging, and entanglement occurred less frequently than did the other patterns. These types of suffocation occurred primarily among infants older than 6 months of age, probably as a reflection of increasing mobility and exploratory behavior that comes with increasing age. The 7- to 11-month-old has much more head control, can push up on hands and knees, may move by “creeping/crawling” along, and can pull to a standing position. Usually, the 12-month-old begins to toddle or walk and can climb, including climbing out of the crib. The entrapments with suspension that occur in a crib or cradle are likely attributable to problems in structural integrity of the product, because the pattern involves falling feet first through a space between the mattress and frame, a space too small to allow the head to pass. Table 3 showed that structural problems were common among the investigated cases. Many hangings involved “nooses” created by necklaces, pacifier cords, and blind cords. Any items worn around the neck create a strangulation hazard because they can catch on protrusions. Since 1977, it has been illegal to sell a pacifier with any type of cord; in addition, all pacifiers are required to have a warning label to inform about the hazards of strangulation.²⁴ The strangulation hazard associated with blinds has been reported by Rauchschwalbe and Mann.²⁵ Pacifier cords and blind cords also were involved in entanglements.

Compression and entrapment without suspension were the least frequent patterns, together accounting for 85 reports. As expected, these involved older infants whose mobility was probably a contributing factor.

Follow-up Investigations

Based on the information from the crib investigations, many contributing factors became obvious (Table 3). In some cases, the investigations provided a basis for clarification. Some of the patterns identified initially as wedging or unknown were found to actually be entrapments with suspension. Therefore, the actual number of wedgings may be somewhat lower and the actual number of entrapments with suspension may be somewhat higher than reported in Table 2. Also, some products described as cribs were found actually to be beds or other products. This emphasizes the importance of follow-up investigation to ensure optimum accuracy of information, better understanding of contributing factors, and provide a basis for effective interventions.

E Codes

Most of the patterns of suffocation were given the code E913, accidental mechanical suffocation. Approximately 80% of wedgings, 60% of overlains, and 40% of oronasal obstructions were coded E913.0, accidental mechanical suffocation in bed or cradle. Such general coding conceals the actual manner of many deaths. It also makes location impossible to identify correctly unless other information is concurrently available, because the code combines data for cradles, products intended for infants and children, with beds, which could be used by adults, infants, or children. Knowledge of both actual manner and location is important in designing successful interventions. Perhaps subsequent ICD revisions should consider one code for beds and one for cribs, cradles, bassinets, and other products generally recognized as intended for infants and toddlers, especially because suffocation deaths are primarily a concern during the first year of life.