BACKGROUND: Some victims of sudden infant death syndrome (SIDS) are found with their heads covered with bedclothes, but the significance of this is uncertain. The aim of this review is to describe the prevalence of head covering, the magnitude of the risk and how far the suggested causal mechanisms agree with current epidemiological evidence. METHODS: Systematic review of population-based age-matched controlled studies. RESULTS: Controlled observations of head covering for the final sleep were found in 10 studies. The pooled prevalence in SIDS victims was 24.6% (95% CI 22.3% to 27.1%) compared to 3.2% (95% CI 2.7% to 3.8%) among controls. The pooled univariate odds ratio (OR) was 9.6 (95% CI 7.9 to 11.7) and the pooled adjusted OR from studies mainly conducted after the fall in SIDS rate was 16.9 (95% CI 12.6 to 22.7). The risk varied in strength but was significant across all studies. In a quarter of cases and controls head covering had occurred at least once previously (pooled adjusted OR = 1.1; 95% CI 0.9 to 1.4). The population attributable risk (27.1%; 95% CI 24.7% to 29.4%) suggests avoiding head covering might reduce SIDS deaths by more than a quarter. CONCLUSIONS: The epidemiological evidence does not fully support postulated causal mechanisms such as hypoxia, hypercapnoea and thermal stress, but neither does it support the idea that head covering is part of some terminal struggle. Head covering is a major modifiable risk factor associated with SIDS deaths and parental advice to avoid this situation should be emphasised.
BACKGROUND: Some victims of sudden infant death syndrome (SIDS) are found with their heads covered with bedclothes, but the significance of this is uncertain. The aim of this review is to describe the prevalence of head covering, the magnitude of the risk and how far the suggested causal mechanisms agree with current epidemiological evidence. METHODS: Systematic review of population-based age-matched controlled studies. RESULTS: Controlled observations of head covering for the final sleep were found in 10 studies. The pooled prevalence in SIDS victims was 24.6% (95% CI 22.3% to 27.1%) compared to 3.2% (95% CI 2.7% to 3.8%) among controls. The pooled univariate odds ratio (OR) was 9.6 (95% CI 7.9 to 11.7) and the pooled adjusted OR from studies mainly conducted after the fall in SIDS rate was 16.9 (95% CI 12.6 to 22.7). The risk varied in strength but was significant across all studies. In a quarter of cases and controls head covering had occurred at least once previously (pooled adjusted OR = 1.1; 95% CI 0.9 to 1.4). The population attributable risk (27.1%; 95% CI 24.7% to 29.4%) suggests avoiding head covering might reduce SIDS deaths by more than a quarter. CONCLUSIONS: The epidemiological evidence does not fully support postulated causal mechanisms such as hypoxia, hypercapnoea and thermal stress, but neither does it support the idea that head covering is part of some terminal struggle. Head covering is a major modifiable risk factor associated with SIDS deaths and parental advice to avoid this situation should be emphasised.