- Journal List
- HHS Author Manuscripts
- PMC5552234
As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsem*nt of, or agreement with, the contents by NLM or the National Institutes of Health.
Learn more: PMC Disclaimer | PMC Copyright Notice
Epidemiology. Author manuscript; available in PMC 2018 Sep 1.
Published in final edited form as:
Iny Jhun,1,2 Douglas A. Mata,1,3 Francesco Nordio,4 Mihye Lee,2 Joel Schwartz,2 and Antonella Zanobetti2
Author information Copyright and License information PMC Disclaimer
The publisher's final edited version of this article is available at Epidemiology
Associated Data
- Supplementary Materials
Abstract
Background
Sudden infant death syndrome (SIDS) is a leading cause of infant mortality in the United States. While thermal stress is implicated in many risk factors for SIDS, the association between ambient temperature and SIDS remains unclear.
Methods
We obtained daily individual-level infant mortality data and outdoor temperature data from 1972 to 2006 for 210 United States cities. We applied a time-stratified case-crossover analysis to determine the effect of ambient temperature on the risk of SIDS by season. We stratified the analysis by race, infant age, and climate.
Results
There were a total of 60,364 SIDS cases during our study period. A 5.6°C (10°F) higher daily temperature on the same day was associated with an increased SIDS risk of 8.6% (95% confidence interval (CI): 3.6%, 13.8%) in the summer, compared to a 3.1% decrease (95% CI: -5.0%, -1.3%) in the winter. Summer risks were greater among Black infants (18.5%, 95% CI: 9.3%, 28.5%) than White infants (3.6%, 95% CI: -2.3%, 9.9%), and among infants 3-11 months old (16.9%, 95% CI: 8.9%, 25.5%) than infants 0-2 months old (2.7%, 95% CI: -3.5%, 9.2%). The temperature-SIDS association was stronger in climate clusters in the Midwest and surrounding northern regions.
Conclusions
Temperature increases were associated with an elevated risk of SIDS in the summer, particularly among infants who were Black, 3 months and older, and living in the Midwest and surrounding northern regions.
Keywords: Sudden infant death syndrome, temperature, case-crossover
Introduction
Sudden infant death syndrome (SIDS) is the third leading cause of overall infant mortality and the leading cause of postneonatal infant mortality (1 month to 1 year of age) in the United States.1–3 SIDS typically occurs during sleep and is defined as the sudden death of an infant under one year of age that remains unexplained even after a thorough case investigation, including an autopsy.4 While the incidence of SIDS decreased drastically following initiatives that promoted supine sleeping positions in the early 1990s, rates have remained stagnant for the past 15 years.3
The cause of SIDS is unknown. However, challenges to the thermoregulatory capacity of infants (i.e., thermal stress) have been implicated in many of the well-known risk factors, such as prone sleeping position, overwrapping infants, and elevated room temperature.3,5–8 For infants in the critical developmental period for homeostatic control, thermal stress can increase demand on the thermoregulatory mechanism during sleep and can impair arousal mechanisms, respiratory drive, cerebral oxygenation, and cardiac responses.9 Studies have found that bedroom heating increases SIDS risk,10 whereas well-ventilated bedrooms and use of a fan is associated with decreased risk of SIDS.11,12 These findings suggest that indoor heat is an important risk factor for SIDS. In addition, indoor and outdoor temperatures have been shown to correlate strongly, particularly in the warm season.13–15 Therefore, outdoor heat is a plausible risk factor for SIDS, both directly from the thermal stress of outdoor heat exposure and indirectly through its effect on indoor temperature.
Despite numerous studies investigating risk factors related to thermal stress, there is limited literature on the role of outdoor heat as a risk factor for SIDS. The majority of studies on SIDS and outdoor temperature date to the 1990s that focus on outdoor cold and report a negative association between SIDS and temperature.16 It was generally concluded that hypothermia was not a direct cause, but rather that low temperature was a marker for parental behavioral changes (e.g., the use of excessive infant clothing and bedding) and seasonal increases in respiratory infection.17,18 Since then, there has been a paucity of studies investigating the association between outdoor temperature and SIDS. Among the few recent studies, one study reported that heat exposure is a risk factor for SIDS19 while others reported a null association.20–22
As climate change may lead to greater frequency and intensity of heat waves, it is critical to evaluate temperature as a potential risk factor for SIDS. In this study, we conducted a large national study using individual-level infant mortality data to assess the effect of ambient temperature on SIDS risk over a 35-year period in 210 United States cities.
Methods
Infant Mortality Data
We obtained daily infant mortality data collected between 1972 and 2006 for 210 United States cities from the National Center for Health Statistics (NCHS). Infant deaths were defined as deaths of infants less than 1 year of age. SIDS cases were identified using International Classification of Diseases 8th Revision (ICD-8), 9th Revision (ICD-9), and 10th Revision (ICD-10) codes 795.0, 798.0, and R95.0, respectively.
Environmental Data
We obtained daily 24-hr average outdoor air and dew-point temperatures measured at airport weather stations from the National Oceanic and Atmospheric Administration (NOAA). Relative humidity was calculated with the following formula:
where Tair and Tdewpt denote 24-hr average air temperature and dew-point temperature, respectively. The closest weather station was matched to each city. For missing measurements, values from the nearest station within 60 kilometers were used as in previous studies.23,24
As one study found an association between ozone and SIDS,25 we obtained ozone (daily max 8-h mean) data from the United States Environmental Protection Agency's Air Quality System Technology Transfer Network. Daily max 8-h mean ozone data were available for 1985-2006 in 202 of the 210 cities that we examined.
Statistical Analyses
We used a time-stratified case–crossover analysis to assess the association between same-day air temperature and SIDS. The case–crossover design is ideal for assessing associations between short-term exposures and rare outcomes such as SIDS,26 and has been used in previous studies assessing heat exposure and SIDS.19,22 The case–crossover design compares each subject's exposure in a time period just prior to a case-defining event with that subject's exposure at other times. Since each subject is his or her own control, all measured or unmeasured time-invariant subject characteristics (e.g., sex, race, socioeconomic status, year/season of birth) are inherently adjusted for, such that there is no confounding by these characteristics. We chose control days as every three days preceding and/or following the date of death within the same month.
We investigated the association between daily average air temperature and SIDS using a conditional logistic regression model (PROC PHREG in SAS, version 9.4, SAS Institute, Cary, NC). An estimate for each season was obtained through an interaction term between temperature and season (defined as spring, Mar-May; summer, Jun-Aug; fall, Sept-Nov; winter, Dec-Feb). We used a continuous variable for air temperature and adjusted for relative humidity as a continuous linear variable and day of the week with indicator variables. We scaled the results to represent a percent excess SIDS risk per 5.6°C (10°F) difference in temperature. We examined the effect of same-day temperature exposure (lag 0) as well as exposures on previous days (i.e., lag 1, lag 2). In addition, we stratified the analysis age, race, sex, and climate (described below).
Climate Clusters
Given the variability in climate across the United States, we stratified our analysis by different climate characteristics. To do so, we categorized the 210 cities into eight climate clusters (Figure 1). The clusters were created through an agglomerative hierarchical approach to minimize within-cluster heterogeneity and maximize between-cluster heterogeneity based on the mean and standard deviation of temperature and relative humidity for all four seasons in each city (more details in Nordio et al., 2015).24 While geographic proximity was not a factor in the clustering algorithm, cities were grouped in a manner in which a predominant geographic pattern can be identified (e.g., West Coast, Northeast, Gulf Coast, etc.).
Sensitivity Analyses
We conducted several sensitivity analyses to examine the effects of the Back to Sleep (BTS) campaign, air conditioning prevalence, and an alternative temperature exposure metric on the temperature-SIDS association. As the incidence of SIDS drastically declined after the introduction of the BTS campaign in the early 1990s, we stratified our analysis by time period (i.e., 1972-1993 and 1994-2006). In addition, air conditioning has been shown to modify the association between temperature and adult mortality.27,28 We calculated the percentage of households in the city with any air conditioning (central or room) based on the Census Bureau's American Housing Survey (more details in Reid et al. 2009).28 We assessed the effect of AC on temperature-related SIDS risk by stratifying the analysis by level of air conditioning prevalence (i.e., lower, middle, and upper tertiles). Lastly, as apparent temperature has been used in previous studies examining temperature-related mortality,22,29 we compared the risk estimates derived from apparent temperature and air temperature. Apparent temperature is defined as an individual's perceived air temperature at a given humidity level. Due to greater completeness of air temperature data, the apparent temperature data was only used for sensitivity analysis. Apparent temperature (AT) was calculated with the following formula:
AT =−2.653 +(0.994×Tair) +(0.0153×(Tdewpt)2)
Results
There were 554,644 infant deaths between 1972 and 2006, of which 60,364 (10.9%) were attributed to SIDS (Table). SIDS incidence was higher among males (60.0%), and was highest in the winter (30.7%) and lowest in the summer (18.9%). The study population consisted of 63.7% White, 33.0% Black, 2.2% Asian/Pacific Islander, and 1.0% Native American infants. More than half of SIDS cases occurred at 0-2 months of age. The number of SIDS cases were greatest in clusters 1, 2, and 5, which in aggregate represented more than 60% of total cases. The eight climate clusters exhibited distinct combinations of temperature and relative humidity profiles (eFigure 1, eTable 1). More detailed descriptions of each climate cluster can be found in Nordio et al. 2015.24
Table
SIDS cases (n=60,364) in 210 United States cities during 1972-2006
| Season | n | (%) |
|---|---|---|
| Spring | 15,460 | (25.6) |
| Summer | 11,413 | (18.9) |
| Fall | 14,952 | (24.8) |
| Winter | 18,539 | (30.7) |
| Male | 36,237 | (60.0) |
| Race | ||
| White | 38,456 | (63.7) |
| Black | 19,923 | (33.0) |
| Asian/Pacific Islander | 1,327 | (2.2) |
| Native American | 608 | (1.0) |
| Other | 50 | (0.1) |
| Climate clustera | ||
| Cluster 1 (Northeast) | 9,672 | (16.0) |
| Cluster 2 (Upper Midwest + AK) | 14,481 | (24.0) |
| Cluster 3 (North Midland) | 4,842 | (8.0) |
| Cluster 4 (South Midland) | 6,630 | (11.0) |
| Cluster 5 (West Coast) | 13,592 | (22.5) |
| Cluster 6 (Gulf Coast) | 4,489 | (7.4) |
| Cluster 7 (FL, bordering Mexico + HI) | 3,197 | (5.3) |
| Cluster 8 (Between CA and NM) | 3,461 | (5.7) |
| Age | ||
| 0-2 months | 34,101 | (56.0) |
| 3-11 months | 26,263 | (44.0) |
aClimate clusters were created based on a statistical approach that grouped cities based on their temperature and relative humidity profiles (see Methods). The regions listed in this table reflect the predominant geographic distribution of each cluster.
SIDS: sudden infant death syndrome. AK: Alaska. FL: Florida. HI: Hawaii. CA: California. NM: New Mexico.
We estimated the association between SIDS and temperature for each season. For a 5.6°C (10°F) increase in temperature, the risk of SIDS increased by 8.6% (95% CI: 3.6%, 13.8%) in the summer (Jun-Aug) and decreased by 3.1% (95% CI: -5.0%, -1.3%) in the winter (Dec-Feb) (Figure 2). There was no association between SIDS and temperature in the fall and spring. We also assessed the effect of temperature exposure from 1 and 2 days prior to the day of death. In the winter, protective associations were also observed for warmer temperatures 1 and 2 days prior that were similar in magnitude to same-day temperature effects. In the summer, there was an 8.0% (95% CI: 3.2%, 13.1%) increased risk of SIDS for a 5.6°C increase in the previous day's temperature, but there was no association with temperature 2 days prior. When the same day and previous day summer temperatures were included in the model simultaneously, the cumulative risk did not explain more risk than using either single temperature model. The results were robust to the adjustment of ozone (eTable 2), as well as the use of apparent temperature (i.e., perceived air temperature at a given humidity level; see Methods section) instead of air temperature (eFigure 2). There was no indication that apparent temperature was a better predictor than air temperature. In addition, there was no evidence of nonlinearity in the relationship between SIDS and temperature in the summer (eFigure 3).
Percent increase in SIDS for a 5.6 °C (10 °F) increase in same-day temperature (lag 0), prior day temperature (lag 1), and 2 days prior temperature (lag 2) by season across 210 cities (1972-2006). 95% confidence intervals are shown.
Temperature-related risk of SIDS varied by race and infant age. The summertime risk was greater for Black infants than for White infants (Figure 3A). For a 5.6°C increase in temperature, the excess risk of SIDS in the summer was 18.5% (95% CI: 9.3%, 28.5%) for Black infants compared to 3.6% (95% CI: -2.3%, 9.9%) for White infants. There were not enough SIDS cases to observe a statistically significant association in other race categories (eTable 3).
Percent increase in SIDS for a 5.6 °C (10 °F) increase in same-day temperature by season across 210 cities (1972-2006), stratified by race (A) and age (0-2 months vs. 3-11 months) (B). 95% confidence intervals are shown.
Risk of SIDS was higher among infants 3-11 months of age compared to those 0-2 months of age (Figure 3B, eTable 4). For a 5.6°C increase in temperature, the excess risk of SIDS in the summer was 16.9% (95% CI: 8.9%, 25.5%) for infants 3-11 months old compared to 2.7% (95% CI: -3.5%, 9.2%) for infants 0-2 months old. Summertime temperature-related SIDS risk was greatest among 4 month old infants (20.9%, 95% CI: 4.3%, 40.2%). Furthermore, Black infants 3-11 months old had a 24.3% (95% CI: 9.8%, 40.7%) increased risk of SIDS per 5.6°C increase in temperature during the summer. There were no differences in SIDS risk observed between male and female infants (eFigure 4).
The incidence of SIDS declined drastically following the BTS campaign in the early 1990s. To assess whether changes in infant sleep practices modified the association between temperature and SIDS, we stratified our analysis by pre- and post-BTS periods: 1972-1993 (22 years) and 1994-2006 (13 years) (Figure 4). The summer and winter risk estimates in the post-BTS period were similar in magnitude to those in the pre-BTS period, although with wider confidence intervals because of the substantial reduction in SIDS cases. When we stratified the analysis further by age group, SIDS risk among 3-11 month old infants was the same between the pre-BTS period (16.8%, 95% CI: 7.4%, 27.0%) and post-BTS period (16.6%, 95% CI: 2.0%, 33.4%). This suggests that temperature-related SIDS risk was not modified by any BTS-related changes in infant sleeping conditions.
Percent increase in SIDS for a 5.6 °C (10 °F) increase in same-day temperature by season across 210 cities, stratified by time period (1972-1993 vs. 1994-2006) and age (0-2 months vs. 3-11 months).
Differences in temperature-associated SIDS risk were observed between climate clusters (Figure 5). In the summer, temperature increases were associated with an elevation in SIDS risk in cluster 2 (Upper Midwest/Northeast) by 11.5% (95% CI: 2.8%, 21.0%) and in cluster 3 (North Midland) by 18.4% (95% CI: 2.4%, 36.8%). In addition, a borderline significant positive association was observed in cluster 5 (West Coast) by 11.6% (95% CI: -1.5%, 26.4%; p = 0.08). We observed generally higher risks among clusters with cooler summers, although this trend was not statistically significant (p=0.14) (eFigure 5). Cities in clusters 2 and 3 had the coldest winters and are located predominantly in the Upper and Industrial Midwest (eTable 1). In addition, clusters 2 and 5 had the lowest air conditioning prevalence among all the clusters (eFigure 6). While air conditioning prevalence did not modify the association between temperature and SIDS, the temperature-SIDS association was most robust among cities with moderate AC prevalence (77.4-95.7% prevalence) (eFigure 7). In the winter, an increase in temperature was associated with a decreased risk of SIDS in cluster 5, which included cities on the West Coast. In all other clusters, the associations were also negative but not robust in the winter.
Discussion
In a large, national study of 60,364 SIDS cases, acute (i.e., same or previous day) exposure to increased ambient temperature was associated with an increased risk of SIDS in the summer. This increased risk was greater among Black infants and infants 3 months or older. The risk varied geographically by climate characteristics, with more pronounced heat effects estimated in clusters with cooler summers. Our findings suggest that heat exposure is an important risk factor for SIDS. Subsequently, projected increases in the frequency and intensity of heat waves related to climate change may pose a substantial risk to infants.
Our results are consistent with a recent case-crossover study in Montreal, Canada of 196 SIDS cases, which found a nearly three-fold increased risk of SIDS on higher temperature days (≥29°C relative to 20°C) during the summer.19 In addition, risks were much greater among infants 3 months or older. This is consistent with the more than six-fold risk we found among infants 3-11 months old compared with infants 0-2 months old. In contrast, null associations have been reported in a case–crossover study (861 SIDS cases) conducted in California30 and in an ecologic analysis of SIDS rates (111 SIDS cases) conducted in Arkansas, Georgia, Kansas, and Missouri.21 A study in Taiwan of 1,671 SIDS cases reported a protective effect of higher temperature days (≥26.4°C relative to 21.9 - <23.3°C in daily maximum temperature).20 Aside from these studies, however, the majority of the literature on outdoor temperature and SIDS date to the 1990s and focus on the effects of colder outdoor temperatures but not heat exposure.16,18,31 Furthermore, given the relatively rare incidence of SIDS, prior studies have been limited by sample size and geographic distribution. To date, this present study represents the largest analysis on SIDS and outdoor temperature.
It is hypothesized that for a vulnerable infant in a critical development period, exposure to external stressors (e.g., prone sleeping) can overwhelm cardiorespiratory autonomic control and sleep arousal mechanisms.7 Many of the well-known risk factors for SIDS (e.g., prone sleeping, infant overwrapping, bedroom heating, bed sharing) are associated with thermal stress.3 Conversely, a well-ventilated bedroom and the use of a fan reduce the risk of SIDS.11,12 Thermal stress can lead to infant death by disrupting the respiratory drive, laryngeal closure reflex, and arousal mechanisms.9,21,32
We found greater heat-related SIDS risk among infants 3-11 months old than infants 0-2 months old. Overwrapping, room heating, and profuse night sweating have been more strongly associated with SIDS among infants older than 2 months of age.32,33 In addition, prone sleeping alters cardiorespiratory autonomic control and results in decreased cerebral oxygenation during sleep, particularly at 2 to 3 months of age.34 Other studies have also suggested that there are temporary protective mechanisms among neonates that disappear as infants develop,35 and that changes in breastfeeding rates may contribute to age-related differences in the susceptibility to SIDS.2 Therefore, multiple factors may contribute to the elevation in temperature-related SIDS risk among infants 3 months and older.
Our results suggest that estimated differential effects of outdoor heat contribute to racial disparities in SIDS rates, given that for the same temperature increase, Black infants had more than a five-fold risk compared with White infants. Race/ethnicity may serve as a proxy for socioeconomic status in the United States, specifically for heat-related health effects.36 However, SIDS rates among infants who are African American, Native American, or Alaskan Native are at least twice the national average, regardless of socioeconomic status.1,7 Therefore, our results may reflect biologic and/or cultural differences that increase susceptibility to SIDS with heat exposure. Studies have investigated the role of serotonergic brainstem abnormalities to explain increased vulnerability of male infants to SIDS,37 but to our knowledge, physiologic differences by race/ethnicity have not been identified. While racial and ethnic differences in sleeping environments (e.g., supine positioning, bed sharing, use of soft bedding) contribute to disparities in the incidence of SIDS,3,38–40 we found that temperature-related SIDS risk ratios remained the same after the BTS campaign drastically reduced SIDS rates in the mid-1990s. This suggests that changes in infant sleeping conditions resulting from the BTS campaign did not modify the association of outdoor temperature exposure with SIDS risk. In other words, while the BTS campaign was associated with reduced incidence of SIDS, infant susceptibility to SIDS resulting from outdoor heat exposure remained unaffected.
We also found important differences among regions with different climate characteristics. Among the eight climate clusters, summertime temperature-related SIDS risk was more pronounced in clusters in the Upper/Industrial Midwest and the surrounding region. We observed a general, albeit not statistically significant, trend of higher risks in clusters with cooler summers. This is consistent with observations that individuals living in cities with low summer temperatures have higher mortality risks during periods of high temperature, after controlling for air conditioning.24 Among adults, there is evidence of acclimatization to local weather conditions and that temperature variability is a key driver of geographic differences in heat exposure effects.41 However, little is known about physiological adaptation to heat among infants. The role of air conditioning in mitigating heat effects is also well established.27 While we found differences in air conditioning prevalence between climate clusters, we did not find evidence that air conditioning prevalence modified the temperature-related SIDS risk. Nonetheless, geographic differences in behavioral adaptation to heat (e.g., avoidance of outdoor exposure, air conditioning use), urban design (e.g., green space, shading on street surfaces), or building characteristics (e.g., building heat penetration) could contribute to differences in susceptibility between the climate clusters. These potential modifying factors and their roles in the geographic and racial/ethnic disparities in SIDS risk would be important topics for future SIDS research.
Behavioral effects are also important for understanding the negative wintertime temperature-SIDS relationship. Consistent with older studies that report increased SIDS incidence with colder temperatures,16,18 we found that warmer temperatures during the winter were protective against SIDS. Studies have shown that the increased SIDS risk associated with colder temperatures in the winter can be explained by excessive clothing and overwrapping of infants.7 Therefore, the winter effects reflect parental behaviors in response to cold temperature rather than the effects of actual exposure to colder temperatures.
Many of the limitations of our study stem from data availability. We utilized monitored outdoor temperatures and not personal temperature exposure data, which can lead to exposure misclassification. We did not have individual mortality data after 2006, as these were no longer available from NCHS. We also did not have additional race/ethnicity categories (e.g., Hispanic) to fully assess racial/ethnic disparities. Nonetheless, we found important differences that warrant targeted public health interventions. While the case–crossover design was ideal for adjusting for time-invariant subject-level confounders that were not available in the data, there were important assumptions. We assumed that the temperature-SIDS relationship was similar across multiple cities. To address this, we stratified our analysis by climate clusters, but there may be important differences within clusters. While a city-level analysis would be ideal, this was not possible due to the limited number of SIDS cases per city. In addition, we were not able to evaluate the role of potential modifiers, such as preterm birth and maternal tobacco use, however, these are time-invariant individual risk factors that are inherently adjusted for in the case-crossover design.
In summary, we identified outdoor temperature as an important risk factor for SIDS, especially among Black infants, infants 3 months and older, and infants living in the Midwest and surrounding northern regions in the United States. Our results suggest that future increases in frequency and intensity of heat waves related to climate change may have important implications for infant health.
Supplementary Material
Supplemental Digital Content
Click here to view.(292K, pdf)
Acknowledgments
We would like to thank Dr. Douglas Levy and Dr. Luc Djousse for their feedback during the conception of the study.
Funding: This work was supported by NIEHS grant R21ES024012 and P30 ES000002, NIH grant P50MD010428-01, and US EPA grant 83615601-0.
Footnotes
Data availability: Because the mortality dataset was used under a Data User Agreement from each state, the dataset and code are not available for replication.
Conflicts of interest: The authors declare no conflict of interest.
References
1. Mathews TJ, MacDorman MF. Infant mortality statistics from the 2010 Period Linked Birth/Infant Death data set. Natl Vital Stat Reports. 2013;62(8):1–33. [PubMed] [Google Scholar]
2. Goldstein RD, Trachtenberg FL, Sens MA, Harty BJ, Kinney HC. Overall Postneonatal Mortality and Rates of SIDS. Pediatrics. 2016;137(1):1–10. doi:10.1542/peds.2015-2298. [PubMed] [CrossRef] [Google Scholar]
3. Moon RY. SIDS and Other Sleep-Related Infant Deaths: Expansion of Recommendations for a Safe Infant Sleeping Environment. Pediatrics. 2011;128(5):e1341–e1367. [PubMed] [Google Scholar]
4. Willinger M, James LS, Catz C. Defining the sudden infant death syndrome (SIDS): deliberations of an expert panel convened by the National Institute of Child Health and Human Development. Pediatr Pathol. 1991;11(5):677–684. [PubMed] [Google Scholar]
5. Ponsonby AL, Dwyer T, Gibbons LE, Cochrane JA, Jones ME, McCall MJ. Thermal environment and sudden infant death syndrome: case-control study. BMJ. 1992;304(6822):277–282. [PMC free article] [PubMed] [Google Scholar]
6. Ponsonby AL, Dwyer T, Gibbons LE, Cochrane JA, Wang YG. Factors potentiating the risk of sudden infant death syndrome associated with the prone position. N Engl J Med. 1993;329(6):377–382. [PubMed] [Google Scholar]
7. Moon RY, Horne RSC, Hauck FR. Sudden infant death syndrome. Lancet. 2007;370(9598):1578–1587. doi:10.1016/S0140-6736(07)61662-6. [PubMed] [CrossRef] [Google Scholar]
8. Guntheroth WG, Spiers PS. Thermal stress in sudden infant death: Is there an ambiguity with the rebreathing hypothesis? Pediatrics. 2001;107(4):693–698. [PubMed] [Google Scholar]
9. Harper RM, Kinney HC, Fleming PJ, Thach BT. Sleep influences on homeostatic functions: implications for sudden infant death syndrome. Respir Physiol. 2000;119(2-3):123–132. [PubMed] [Google Scholar]
10. Ponsonby AL, Dwyer T, Kasl SV, Cochrane JA. The Tasmanian SIDS Case-Control Study: univariable and multivariable risk factor analysis. Paediatr Perinat Epidemiol. 1995;9(3):256–272. [PubMed] [Google Scholar]
11. McGlashan ND. Sudden infant deaths in Tasmania, 1980-1986: a seven year prospective study. Soc Sci Med. 1989;29(8):1015–1026. [PubMed] [Google Scholar]
12. Coleman-Phox K, Odouli R, Li D. Use of a fan during sleep and the risk of sudden infant death syndrome. Arch Pediatr Adolesc Med. 2008;162(10):963–968. [PubMed] [Google Scholar]
13. Nguyen JL, Schwartz J, Dockery DW. The relationship between indoor and outdoor temperature, apparent temperature, relative humidity, and absolute humidity. Indoor Air. 2014;24(1):103–112. [PMC free article] [PubMed] [Google Scholar]
14. Nguyen JL, Dockery DW. Daily indoor-to-outdoor temperature and humidity relationships: a sample across seasons and diverse climatic regions. Int J Biometeorol. 2016;60(2):221–229. [PMC free article] [PubMed] [Google Scholar]
15. Tamerius J, Perzanowski M, Acosta L, et al. Socioeconomic and Outdoor Meteorological Determinants of Indoor Temperature and Humidity in New York City Dwellings. Weather Clim Soc. 2013;5(2):168–179. [PMC free article] [PubMed] [Google Scholar]
16. Williams SM, Mitchell EA, Stewart AW, Taylor BJ. Temperature and the sudden infant death syndrome. Paediatr Perinat Epidemiol. 1996;10(2):136–149. [PubMed] [Google Scholar]
17. Tuohy PG, Tuohy RJ. The overnight thermal environment of infants. N Z Med J. 1990;103(883):36–38. [PubMed] [Google Scholar]
18. Campbell MJ. Time Series Regression for Counts: An Investigation into the Relationship between Sudden Infant Death Syndrome and Environmental Temperature. J R Stat Soc Ser A (Statistics Soc. 1994;157(2):191. [Google Scholar]
19. Auger N, Fraser WD, Smargiassi A, Kosatsky T. Ambient Heat and Sudden Infant Death: A Case-Crossover Study Spanning 30 Years in Montreal, Canada. Environ Health Perspect. 2015;123(7):712–716. [PMC free article] [PubMed] [Google Scholar]
20. Chang HP, Li CY, Chang YH, Hwang SL, Su YH, Chen CW. Sociodemographic and meteorological correlates of sudden infant death in Taiwan. Pediatr Int. 2013;55(1):11–16. [PubMed] [Google Scholar]
21. Scheers-Masters JR, Schootman M, Thach BT. Heat stress and sudden infant death syndrome incidence: a United States population epidemiologic study. Pediatrics. 2004;113(6):e586–92. [PubMed] [Google Scholar]
22. Basu R, Pearson D, Sie L, Broadwin R. A Case-Crossover Study of Temperature and Infant Mortality in California. Paediatr Perinat Epidemiol. 2015;29(5):407–415. [PubMed] [Google Scholar]
23. Lee M, Nordio F, Zanobetti A, Kinney P, Vautard R, Schwartz J. Acclimatization across space and time in the effects of temperature on mortality: a time-series analysis. Environ Health. 2014;13:89. [PMC free article] [PubMed] [Google Scholar]
24. Nordio F, Zanobetti A, Colicino E, Kloog I, Schwartz J. Changing patterns of the temperature-mortality association by time and location in the US, and implications for climate change. Environ Int. 2015;81:80–86. doi:10.1016/j.envint.2015.04.009. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
25. Woodruff TJ, Darrow LA, Parker JD. Air pollution and postneonatal infant mortality in the United States, 1999-2002. Environ Health Perspect. 2008;116(1):110–115. [PMC free article] [PubMed] [Google Scholar]
26. Maclure M. The Case-Crossover Design: A Method for Studying Transient Effects on the Risk of Acute Events. Am J Epidemiol. 1991;133(2):144–153. [PubMed] [Google Scholar]
27. O'Neill MS, Zanobetti A, Schwartz J. Disparities by race in heat-related mortality in four US cities: The role of air conditioning prevalence. J Urban Heal. 2005;82(2):191–197. [PMC free article] [PubMed] [Google Scholar]
28. Reid CE, O’Neill MS, Gronlund CJ, et al. Mapping community determinants of heat vulnerability. Env Heal Persp. 2009;117(11):1730–1736. doi:10.1289/ehp.0900683. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
29. Zanobetti A, Schwartz J. Temperature and mortality in nine US cities. Epidemiology. 2008;19(4):563–570. [PMC free article] [PubMed] [Google Scholar]
30. Basu R. High ambient temperature and mortality: a review of epidemiologic studies from 2001 to 2008. Env Heal. 2009;8:40. [PMC free article] [PubMed] [Google Scholar]
31. Schluter PJ, Ford RP, Brown J, Ryan aP. Weather temperatures and sudden infant death syndrome: a regional study over 22 years in New Zealand. J Epidemiol Community Health. 1998;52:27–33. [PMC free article] [PubMed] [Google Scholar]
32. Fleming PJ, Levine MR, Azaz Y, Wigfield R, Stewart AJ. Interactions between thermoregulation and the control of respiration in infants: possible relationship to sudden infant death. Acta Paediatr Suppl. 1993;82(Suppl 3):57–59. [PubMed] [Google Scholar]
33. Kohlendorfer U, Kiechl S, Sperl W. Sudden infant death syndrome: risk factor profiles for distinct subgroups. Am J Epidemiol. 1998;147(10):960–968. [PubMed] [Google Scholar]
34. Yiallourou SR, Walker AM, Horne RSC. Prone sleeping impairs circulatory control during sleep in healthy term infants: implications for SIDS. Sleep. 2008;31(8):1139–1146. [PMC free article] [PubMed] [Google Scholar]
35. Guntheroth WG, Spiers PS. The Triple Risk Hypotheses in Sudden Infant Death Syndrome. Pediatrics. 2002;110(5):e64–e64. [PubMed] [Google Scholar]
36. Gronlund CJ. Racial and socioeconomic disparities in heat-related health effects and their mechanisms: a review. Curr Epidemiol reports. 2014;1(3):165–173. [PMC free article] [PubMed] [Google Scholar]
37. Paterson DS, Trachtenberg FL, Thompson EG, et al. Multiple serotonergic brainstem abnormalities in sudden infant death syndrome. JAMA. 2006;296(17):2124–2132. [PubMed] [Google Scholar]
38. Hauck FR, Moore CM, Herman SM, et al. The contribution of prone sleeping position to the racial disparity in sudden infant death syndrome: the Chicago Infant Mortality Study. Pediatrics. 2002;110(4):772–780. [PubMed] [Google Scholar]
39. Flick L, White DK, Vemulapalli C, Stulac BB, Kemp JS. Sleep position and the use of soft bedding during bed sharing among African American infants at increased risk for sudden infant death syndrome. J Pediatr. 2001;138(3):338–343. [PubMed] [Google Scholar]
40. Rasinski KA, Kuby A, Bzdusek SA, Silvestri JM, Weese-Mayer DE. Effect of a sudden infant death syndrome risk reduction education program on risk factor compliance and information sources in primarily black urban communities. Pediatrics. 2003;111(4 Pt 1):e347–54. [PubMed] [Google Scholar]
41. Medina-Ramón M, Schwartz J. Temperature, temperature extremes, and mortality: a study of acclimatisation and effect modification in 50 US cities. Occup Environ Med. 2007;64(12):827–833. [PMC free article] [PubMed] [Google Scholar]
