Letter to the Hon. Alex Azar, U.S. Secretary of the Dept. of Health and Human Services - Menendez, Colleagues Continue Pressuring CDC for Accurate Firearm Injury Data

Letter

Dear Secretary Azar:

We write regarding your May 3 response to our March 29 letter concerning the Centers for Disease Control's (CDC) firearm injury data. As we noted in our previous letter, "[s]ince 2010, at least 50 academic articles have cited the CDC's gun injury estimates." Yet, the CDC acknowledges that its national estimate of gun injuries is "unstable and potentially unreliable." We believe the CDC's explanation falls short. Thus, we seek additional answers to the following questions:

1. Your response states the CDC is exploring ways to improve the precision and accuracy of non-fatal firearm injury data by analyzing Healthcare Cost and Utilization Project-National Emergency Department Sample data from 2016 when it becomes available. Your response specifically notes that current data only goes through 2014. Yet, the Agency for Healthcare Research and Quality issued 2016 HCUP-NEDS data in September 2018[1], over six months before your May 3, 2019 letter. Why did your response state that such data are unavailable when the data is indeed available?
2. Your response states "a confidence interval will widen as an estimate increases, even when the relative uncertainty associated with the estimate remains constant." As we understand it, a confidence interval is a range of values generated from sample statistics that can be used to estimate the true value for the entire population. The interval is based on the average of the sample population and the standard deviation (a statistical description of the variability or dispersion of the data). The standard deviation will increase (and, therefore, the confidence interval widen) as the dispersion increases if and only if the sample size being used to describe the population does not also increase. Therefore, it would seem that the problem lies not in the large number of firearm deaths and injuries that comprise the data set, but in the poor quality of data being used to describe those deaths and injuries. In order to estimate a large problem, a large sample of data is necessary. How does the CDC plan to rectify this?
3. As you know, National Violent Death Reporting System (NVDRS) does not provide information on non-fatal firearm injuries. While we are not opposed to expansion, we seek to know how expansion of NVDRS to all 50 states advances the goal of gathering accurate non-fatal firearm injury data? Furthermore, while it is important that a wider geographic area be represented in the data, it is also necessary to expand the number of reporting hospitals. By what percentage will the number of reporting hospitals increase in this expansion?

Thank you in advance for your cooperation. We look forward to receiving your responses by August 1, 2019.