University of Tasmania

File(s) under permanent embargo

Sudden infant death syndrome

posted on 2023-05-27, 18:14 authored by Ponsonby, AL
The thesis was based on research conducted from 1988 to 1991 inclusive on Sudden Infant Death Syndrome (SIDS) in Tasmania. It has five sections. The first section was a literature review which discusses the history, pathology, epidemiological features and possible mechanisms of SIDS. The second section described preliminary work conducted. The seasonality and age distribution of SIDS was examined with reference to a biphasic model of SIDS pathogenesis and age distribution. For SIDS deaths in Tasmania during the years 1975 to 1987, the month of birth distribution for SIDS cases did not differ significantly from a uniform, nonseasonal distribution (p>0.1). The month of death distribution of SIDS did display seasonality (p<0.01) with a significant (p<0.001) sinusoidal pattern. Using a multiple comparison test, spring and summer born infants died at a significantly (p<0.05) older age than winter born infants. The relationship between climatic temperature and SIDS incidence for the Australian states was described. Using Australian state data extracted predominantly during the 1980's, a drop of 1¬¨‚àû Celsius in average monthly temperature was associated with a 10.6% (9.6%, 11.7%) increase in the rate of SIDS in all Australian states. Differences in average monthly temperature accounted for 84% of the variation of monthly SIDS rates between Australian states. However, after controlling for the effect of climatic temperatures significant differences in the SIDS rate between the Autralian states remained (p<0.0001). Factors contributing to the difference in SIDS incidence between Victoria and Tasmania were assessed for the period 1985 to 1987. Approximately 82% of the interstate difference in SIDS incidence between Tasmania and Victoria from 1985 to 1987 could be accounted for by interstate differences in climatic temperature, maternal age, birthweight, infant sex, month of birth and feeding intention at hospital discharge. The third section was based on the case control study. For each case, two control infants were chosen ‚ÄövÑvÆ control A was age matched. Control B was age and birthweight matched. Control visits were matched for region of birth, season, time of day and minimum temperature.. Study measurements included a thermal environment assessment, verbal questionnaire and meteorological data. From 1 October 1988 to 1 October 1991, 62 infants died whose cause of death was classified as SIDS. The case response rate for retrospective interview was 93.5% (58/62). The initial response rate for controls was 83.5% (101/121). Thermal assessment at the scene of last sleep was conducted for a subset of the study (case n=28, control n=54). Excess thermal insulation over the infant during last sleep was identified as an independent risk factor for SIDS. For every excess thermal insulation unit (tog), the relative risk of SIDS was 1.26 (1.05, 1.52). For the three year case control sample, univariate risk estimates for factors such as the usual prone position 4.52 (2.12, 9.61) and smoking 3.96 (1.91, 8.24) were obtained. Conditional logistic regression was then used to obtain odds ratios for these factors after adjustment for relevant confounders. A six term model provided the best prediction of SIDS within the case control study. It consisted of 4 positive risk factors, :- prone sleeping position (p<0.001), bedroom heating during last sleep (p= 0.032), family history of asthma (p= 0.041) and maternal smoking (p= 0.01). Maternal age > 25 years (p= 0.012) and more than one child health clinic attendance (p= 0.004) were protective factors. Addition of variables such as infant feeding, birthweigh tor illness did not improve the fit of the model to the data. After adjustment for maternal age and birthweight, the risk associated with usual side position compared to usual supine position was 1.05 (0.27, 5.02) and the risk associated with usual prone position compared to usual supine position was 5.70 (1.67, 25.58). The issue of effect modification in relation to the prone sleeping position was examined. Using a multiplicative model, the risk of usual prone position for SIDS was significantly potentiated by use of a natural fibre matress, swaddling during last sleep, recent infant illness or heating of the infants bedroom. Illness was associated with SIDS among prone-sleeping infants (OR=-5.60 (1.75, 18.47)) but not among side or back lying infants (OR= 0.59(0.15, 2.30})._ The possible mechanism of the prone position was discussed. The characteristics and determinants of the infant thermal environment in healthy control infants were described. The amount of thermal insulation on control infants correlated negatively with room temperature (r= -0.44, p= 0.001). The average level of excess thermal insulation for room temperature was higher (p=0.03) for control infants visited in the winter period centered on July then the 5 month period centered on January. The variation in excess thermal insulation was also greater (p= 0.02) in winter than summer. The fourth section is based on the prospective cohort study. The study methodology was described, including the Tasmanian 'at birth' SIDS scoring system. The source population for the cohort study was all live births occurring at six major obstetric hospitals (94%) of Tasmanian live births. The one-fifth of live births identified by a perinatal scoring system as being at higher risk of SIDS' were eligible to join the study of the 4,180 eligible infants born between 1 January 1988 and 31 December 1990, 3,555 participated in the hospital and home visits, giving a response rate of 85%. The results on the prospective observation of an association between the prone sleeping position and SIDS were discussed. The prospective risk estimate for usual prone position, after adjustment for maternal age and birthweight was calculated (OR= 3.92 (1.37, 11.24)). The population attributable risk for usual prone position and SIDS was 0.38 (0.35, 0.41). The evidence for a causal association between prone sleeping position and SIDS was examined. The rate of SIDS in Tasmania declined significantly ( RR = 0.45 (0.23, 0.84)) in 1991 compared with the years 1981-1990 in Tasmania. Changes in the prevalence of SIDS risk factors in the cohort study were examined. The most striking change was that the proportion of cohort infants usually sleeping prone at one month of age was lower for infants born in 1991 compared with infants born 1988-1990 (RR= 0.41 (0.35, 0.46)). A formal evaluation of the contribution of changes in the prevalence of prone position to the SIDS rate decline is required. The fifth section of the thesis compared the retrospective case control and prospective cohort methodologies. A comparison of prospective and retrospective maternal responses to identical questions was conducted. Good agreement was found for some variables (usual sleep position, infant feeding, maternal smoking) but not others such as family history and patterns of bedroom heating. Finally, future directions in SIDS research were discussed. The emphasis in future SIDS research should shift from the identification of new risk factors to research work aimed at understanding how well identified risk factors operate to precipitate death during infancy.


Publication status

  • Unpublished

Rights statement

Copyright 1992 the author - The University is continuing to endeavour to trace the copyright owner(s) and in the meantime this item has been reproduced here in good faith. We would be pleased to hear from the copyright owner(s). Includes bibliographical references. Thesis (PhD)--University of Tasmania, 1993

Repository Status

  • Restricted

Usage metrics

    Thesis collection


    No categories selected


    Ref. manager