The annual meetings of the Society for Pediatric and Perinatal Epidemiologic Research (SPER) and the Society for Epidemiologic Research (SER) will take place in Baltimore, Maryland from June 18 to 22, 2018. We will present a scientific poster at each of these meetings. At SPER, Mortality related to cerebral palsy in the United States: Analysis of multiple causes of death and comparison with deaths in the general population from 2005 to 2014 will explore deaths in the US that include cerebral palsy (CP - ICD 10 code G80) as immediate, underlying, or contributing cause, or as an “other condition” of significance mentioned on a death certificate. At the SER meeting, Competing risks survival bias among US astronauts will explore whether deaths due to external causes (primarily accidents, including spacecraft tragedies) among astronauts might be obstructing our view of natural-cause deaths in this occupational cohort.

Our SPER poster will demonstrate that deaths that include any mention on a death certificate of CP occur at much younger ages than deaths overall (mean age at death for CP was 42 years, for those without CP 73 years), and have quite a different distribution of immediate and underlying causes of death. In many cases, the National Center for Health Statistics Multiple Cause of Death Data (available here) list CP as the underlying cause of death (i.e., the condition that set the chain of events in motion that ultimately led to death), and this is true in some cases where, technically, another cause should be considered the underlying cause. For this reason, by far the most common underlying cause of death broad grouping for deaths involving CP is “nervous system disorders”, ICD 10 G00-G99, which group includes CP (ICD 10 G80). To shed further light on the chain of events leading to deaths that include any mention of CP, and in particular on what set that chain of events in motion, we recoded underlying cause of death: to the last item listed in Part I of death certificates when this item was not CP, and to the penultimate item (if there was one) when the last item was CP. As a result, the leading cause of death broad category for those deaths mentioning CP anywhere on a death certificate was “respiratory diseases”, ICD 10 J00-J99, rather than nervous system disorders. After re-coding, the top five leading categories of underlying cause of death, from most prevalent to least, were:

1. Respiratory (J00-J99);
2. Other symptoms and signs (R00-R99), which included “convulsions not elsewhere classified”;
3. Nervous system disorders (G00-G99), which included CP, seizures, and hydrocephalus, to name a few);
4. Circulatory diseases (I00-I99); and
5. External causes (V00-V99), which included inhalation or ingestion of food or other objects (W79, W80) as the most common sub-category.

We hope to make a pdf version of the poster available in an update to this post after the SPER meeings have concluded. Now available here.

At the SER meeting, Competing risks survival bias among US astronauts will explore whether deaths due to external causes (primarily accidents, including spacecraft tragedies) among astronauts might be obstructing our view of natural-cause deaths in this occupational cohort. The methods involved in exploring this question form the crux of the poster and are outlined in the abstract, available here. The extremely low chronic disease mortality that has been reported in previous studies of US astronauts does not appear to be an artifact related to the unusually high rate of deaths (at least in some decades and at young ages) due to external causes in this cohort.

We hope to make a pdf version of this poster available in an update to this post after the SER meeings have concluded. Now available here.