Sudden infant death syndrome of substance-abusing mothers in infants Sally L. Davidson Ward, MD, Daisy Bautista, BA, RPFT, Linda Chart, PhD, Margo Derry, PHN, Arthur Lisbin, MD, Michael J. Durfee, MD, Kassia S. C. Mills, BS, and Thomas G. Keens, MD From Childrens Hospital Los Angeles, the Department of Pediatrics, University of Southern Cal- ifornia School of Medicine, and the Los Angeles County Department of Health Services, Los Angeles, California A population-based study was performed to determine whether substance abuse during the perinatal period may be a risk factor for sudden infant death syndrome (SIDS). The incidence of SIDS was studied in 2143 infants of substance- abusing mothers (ISAM) born in Los Angeles County during 1986 and 1987 who were reported to the Los Angeles County Department of Health Services because of a history of drug exposure or positive urine test results in the mother, infant, or both. By comparing the ISAM birth reports with records of autopsy-proven SIDS in Los Angeles County, we found 19 SIDScases in the population of 2143 ISAM, a SIDS rate of 8.87 cases per 1000 ISAM (95% c o n f i d e n c e interval 5.3 to '13.8). This was significantly higher than the SIDS rate for the non-ISAM general population: 396 SIDS deaths among 325,372 live births, an incidence rate of 1.22 cases per 1000 births, p 0.00001. The a g e of ISAM at death was 99 _+ 63 (mean ��� SD) days c o m p a r e d with 91 ��� 52 days for the non-ISAM population (not significant). The incidence of SIDS was significantly greater in male infants, during the winter months, in black infants, and in non-Hispanic white infants in the non-ISAM pop- ulation. Such differences were not observed in the ISAM group. A greater inci- d e n c e of symptomatic apnea was reported before SIDS for the ISAM than for the non-ISAM population (22% vs 5.4%, p = 0.022). We c o n c l u d e that ISAM have a higher incidence of SIDS than the non-ISAM general population. However, it was not possible to separate maternal substance abuse from other confounding variables that may also have had an impact on SIDS risk in the ISAM group. (J PEDIATR 1990 117:876-81) Supported in part by grants from the American Lung Association of Los Angeles County the National Center for the Prevention of Sudden Infant Death Syndrome (Baltimore, Md.) the Greater Los Angeles and Washington State Chapters of the National Sudden .Infant Death Foundation the Los Angeles County, Orange County, Inland Empire, and Kern County Chapters of the Guild for Infant Survival the Junior Women's Club of Orange and the Ruth and Vernon Taylor Foundation. Presented in part at the Society for Pediatric Research, May 1989, in Washington, D.C., and at the Eight Annual Conference on In- fantile Apnea, January 1990, Palm Springs, Calif. Submitted for publication April 27, 1990 accepted June 25, 1990. Reprint requests: Sally L. Davidson Ward, MD, Division of Neo- natology and Pediatric Pulmonology, Childrens Hospital Los An- geles, 4650 Sunset Blvd., Los Angeles, CA 90027. 9/20/23409 The risk of sudden infant death syndrome has been reported as 5 to I 0 times greater in infants of mothers who abuse opiates during pregnancy than in the general population.l6 Infants exposed to cocaine during gestation have been SIDS Sudden infant death syndrome ISAM Infant(s) of substance-abusing mother(s) See related articles, pp. 904 and 909. reported to have an increased S1DS risk by some authors 7, s but not by others. 9 A limitation of the majority of previous studies of SIDS risk in infants of substance-abusing moth- ers is that the studies have been program or hospital based. 8 7 6

Volume 117 SIDS in infants with substance-abusing mothers 8 7 7 Number 6 program-based studies have inherent biases that may not accurately estimate risk. To obtain a more accurate estimation of SlDS risk in the ISAM population, we performed a retrospective, popula- tion-based study in Los Angeles County, Calif., examining the rate of SIDS in I S A M born during the calendar years 1986 and 1987. In addition, we collected other information about ISAM who had died of SIDS, including age at death, seasonal incidence, ethnic background, and incidence of symptomatic apnea. M E T H O D S The Los Angeles County Department of Health Services collects all voluntary reports made by hospitals in Los An- geles County regarding ISAM. These reports are made when illicit substance abuse has occurred during all or part of pregnancy as documented by maternal history or positive results of urine studies of the mother, infant, or both. We included only reports describing abuse of opiates, cocaine, phencyclidine, or amphetamines, alone or in combination. Alcohol abuse a n d marijuana abuse were not included. These records identify the race, gender, date of birth, and birth name of the infant, the names and ages of the parents, and other demographic information. The infant's drug ex- posure, based on history or urine studies of the mothei" or infant, is also recorded. The department of health services is also responsible for maintaining records of all SIDS infants in Los Angeles County. This syndrome is defined in Los Angeles County as the sudden, unexpected death of a previously healthy infant that remains unexplained after a thorough postmortem examination. 1~ The department's records contain the dates of birth and death, race, gender, and names of the infant and parents. Additional forms are completed by a public health nurse who interviews the family if they agree and can be lo- cated the nurse asks specifically about exposure to illicit drugs during the mother's pregnancy and for any history of symptomatic apnea ("turned blue or stopped breathing") before the infant's death. To obtain an estimate of SIDS risk in ISAM in Los An- geles County, we compared the hospital reports of ISAM during the calendar years 1986 and 1987 with the SIDS re- ports from the calendar years 1986, 1987, and 1988. This allowed us to identify SIDS deaths during the first year of life, by which time 99% of SIDS deaths have occurred. 1~ Rates for SIDS among ISAM were compared with the Los Angeles County SIDS rates based on data for the game cal- endar years. The SIDS rates for female and male infants were calculated separately for ISAM and non-iSAM pop- ulations. The SIDS rates in ISAM and non-ISAM popula- tions were also compared within each ethnic group, includ- ing infants of black, Hispanic, non-Hispanic white, and Asian and Pacific Island origin. The SIDS incidence rates for each subpopulation were estimated with a 95% confidence interval based on the Pearson and Hartley method for the Poisson parameter. 11 We compared SIDS rates in the ISAM and non-ISAM groups within each subpopulation, defined by gender or ethnicity, using relative risk estimates and the 95% confi- dence intervals based on the Taylor series. The tw0-tailed Fisher Exact Test was used to determine statistical signif- icance of the relative risk within each subpopulation. The Mantei-Haenszel weighted relative risk, the Mantel-Haen- szel summary chi-square test, and the 95% Greenland- Robins confidence limits were used to compare the overall incidence rates in the ISAM and non-ISAM populations. The EPI-INFO Statistical software package was used to obtain the relative risk estimates, their 95% confidence in- tervals, Fisher Exact Test values, p values, and the Mantel- Haenszel estimates. ~2 On the basis of information recorded on the hospital ISAM birth reports, the numbers of t S A M who were exposed to each abused substance were tabulated. The SIDS rates for each of these substance exposure groups were compared with the rate for the non-ISAM population by means of the Fisher Exact Test and the 95% Taylor se- ries confidence limits for the relative risks. The mean age at the time of death from SIDS was cal- culated for both ISAM and the infants in the non-ISAM population and compared by means of the Student t test. Seasonal variation in SIDS incidence, based on the numbers of infants who died in each month, was evaluated for the Los Angeles County non-ISAM population by chi-square anal- ysis. Seasonal incidence for the ISAM who died of SIDS was evaluated by the Kolmogorov-Smirnov goodness of fit test. Finally, the numbers of infants who died of SIDS in both the non-ISAM and ISAM groups, with and without a history of symptomatic apnea, were tabulated and com- pared by chi-square analysis. We were unable to locate a hospital birth report for some of the infants who died of SIDS but who were identified as ISAM during the public health nurse interview on the basis of a history of exposure to substances of abuse in utero. These infants were removed from the group of infants who died of SIDS in the non-ISAM population, and they were included in the ISAM-S1DS group for the calculations and comparisons of age at time of death, history of symptomatic apnea, and seasonal incidence of death. However, because it is not known how many unreported ISAM there were, these infants were not included in the calculations and comparisons of SIDS rates for ISAM. Strict confidentiality was maintained throughout the study for the families of SIDS infants, the I S A M and their families, and the reporting persons and institutions. Re- search was limited to the information available in the records of the department of health services. No attempt