For what it’s worth, I tweeted this to my moderate-sized audience of mostly food allergy parents. Feel free to retweet or adapt.
Pediatric emergency docs have been running the stats. During the school year, we predict a mass shooting (defined as 3 or more non-shooter victims) every 30 days (95% CI limit 5 months). I feel ill. Do you? Evidence also shows tougher gun laws will save future lives. Time to act.
If you are interested in what’s up with anaphylaxis and some other PEM stuff and random musings, you can follow me @donteatmagnets.
Julie Brown, MD, MPH
Associate Professor, University of Washington
Attending Physician, Pediatric Emergency Medicine,
Seattle Children's Hospital, MB.7.520,
Seattle, WA 98105-0371
Office: (206) 987-4016
Fax: (206) 729-3070
From: Paul M [mailto:[log in to unmask]]
Sent: Monday, February 26, 2018 10:22 AM
To: Brown, Julie
Cc: [log in to unmask]
Subject: Re: statistical help
More or less the same results as Matt - 95% CI upper limit every ~5months. More important though is what we can expect to occur, which is every 30 days during the school year.
Mass shooting definition: shooting resulting in at least 3 victims, excluding the perpetrator (per definition of the Investigative Assistance for Violent Crimes Act of 2012, signed into law on Jan 2013)
SPC T chart of days between mass school shootings (n=121), last 100 years (Upper Control Limit = 3SD); not adjusted for population growth or seasonality; CL = average of 1 every 113 days
[Inline image 1]
The most recent period of stability goes back from event #86 in the figure above, starting at the beginning of 2013 until the current date. This time period makes a new T chart below, which is stable (no significant change) by all special cause rules (1 point >3SD from the center line, 9 points in a row on same side of center line, 6 points in a row all increasing or decreasing, 2 of 3 points >2 SD on same side from CL, 3 of 4 points >1 SD on same side of CL). Now, there is 1 every 41 days, with 3SD UCL of 271 days; not adjusted for seasonality, population growth, or transformed for normality). To the original question of 95% CI question, the UCL for 2 SD is 170 days. The good news is that this is a stable process, so the effect of interventions could be associated with changes in the outcomes. The bad news is that this is a stable process.
[Inline image 7]
But kids are in school seasonally - with decreases seen in spring break and summer. So when school's in session, it's probably more like one mass shooting every 30 days.
[Inline image 3]
If you expand the criteria from school mass shooting to any school shooting (at least 1 injury or death), January is the most frequent month for any school shooting.
[Inline image 4]
Implications: Mass shooter drills in September? Increase mental health resources/screenings in schools/staff/communities after winter break? Other ideas with likely bipartisan support?
Long road ahead...
Paul Mullan, MD, MPH
Director of Research and Quality Improvement, Division of Emergency Medicine
Children's Hospital of the King's Daughters
Associate Professor, Eastern Virginia Medical School
Cell: (713) 855-4827<tel:(713)%20855-4827>. Office: (757) 668-9220<tel:(757)%20668-9220>
[log in to unmask]<mailto:[log in to unmask]>
CONFIDENTIALITY NOTICE: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information protected by law. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message.
For more information, send mail to [log in to unmask] with the message: info PED-EM-L
The URL for the PED-EM-L Web Page is: