Jon Roesler, MS CSTE Disaster Epidemiology Workshop May 9 th, 2013 Atlanta, GA

3 Examples…

3 Examples

Example 1

Recalling Minnesota Disasters  1991 Halloween Blizzard 37” of Snow in Duluth. Between the blizzard and the ice storm, 22 fatalities and over 100 people were injured.  1997 Red River Flood The most severe flood of the river since Affected MN, ND, Canada. 50,000 evacuated in Grand Forks / E. Grand Forks alone, where record flooding occurred. $3.5 billion in damages. No fatalities.  1998 Comfrey – St. Peter Tornado Outbreak 16 tornados in 4 hours including F2, F3, & F4. $235 million in damages. 2 fatalities.  2005 Red Lake Massacre – Murder/Suicide 16 year-old shot grandfather & girlfriend, then 7 at high school, before killing self; 10 fatalities, 15 injured.  2007 I-35W Mississippi River Bridge Collapse During evening rush hour on August 1, 2007, it suddenly collapsed with 13 fatalities and injuring 145 people.  2009 Red River Flood Record flooding in Fargo / Moorhead. 3 fatalities.

Recalling Minnesota Disasters  1991 Halloween Blizzard 37” of Snow in Duluth. Between the blizzard and the ice storm, 22 fatalities and over 100 people were injured.  1997 Red River Flood The most severe flood of the river since Affected MN, ND, Canada. 50,000 evacuated in Grand Forks / E. Grand Forks alone, where record flooding occurred. $3.5 billion in damages. No fatalities.  1998 Comfrey – St. Peter Tornado Outbreak 16 tornados in 4 hours including F2, F3, & F4. $235 million in damages. 2 fatalities.  2005 Red Lake Massacre – Murder/Suicide 16 year-old shot grandfather & girlfriend, then 7 at high school, before killing self; 10 fatalities, 15 injured.  2007 I-35W Mississippi River Bridge Collapse During evening rush hour on August 1, 2007, it suddenly collapsed with 13 fatalities and injuring 145 people.  2009 Red River Flood Record flooding in Fargo / Moorhead. 3 fatalities.

TEEN LONER KILLS 9, SELF Posted Tuesday, March 22nd 2005, 12:00 AM A CRAZED TEENAGER murdered his grandparents and then went on a shooting rampage through his Minnesota school yesterday, killing a teacher, a security guard and five students before putting a bullet in his own head, authorities said. The bloodbath that killed 10 people also injured as many as 15 other kids at Red Lake High School on the Chippewa Indian Reservation.

Timeline Beltrami County March 21, 2005 Red Lake Massacre - Murder/Suicide Fall 2005 Headwaters Alliance Winter 2007 MDH Contacted Fall 2007 MDH Report

Roesler J: Suicide and Nonfatal Self-inflicted Harm, Beltrami County, St. Paul, MN: Minnesota Department of Health, September 14, 2007

Nonfatal Self-inflicted Harm by Age Group, 2005 Roesler J: Suicide and Nonfatal Self-inflicted Harm, Beltrami County, St. Paul, MN: Minnesota Department of Health, September 14, 2007

Minn Med.Minn Med Aug;92(8):53-5. Nonfatal suicide attempts and other self-inflicted harm. Beltrami County youths, Roesler JRoesler J, Petcoff M, Azam A, Westberg S, Kinde M, Crosby A.Petcoff MAzam AWestberg SKinde MCrosby A

Case Definition: Hospital-treated, Self-Inflicted Harm (SIH) Forms: Self-Inflicted Injury Data Collection Form; Report of Injury (ROI) Data Source(s): Select cases that meet sample criteria from UB discharge data Purpose: Continued epidemiologic surveillance of self-inflicted harm in youth from Beltrami County. Inclusion: Under 25 years of age AND All sequences AND All hospitals AND Year of discharge = AND Beltrami County residents only AND Inpatient or Outpatient AND Non-fatal or Fatal AND With any of the following ICD-9-CM Cause codes: E950.0-E958.9 (self-inflicted injury), V62.84 (suicidal ideation) AND With any of the following ICD-9-CM Diagnostic codes: , , or AND Exclusion: Any with the following ICD-9-CM Diagnostic codes: 909.3, 909.5, , , , , , (late effects and certain adverse effects)

Abstraction All Hospitalized SIH, <25 years old Sample of ED-treated SIH, <25 years old

2 Main Findings Double the expected rate of prior attempts Pump Handle Effect

Expected Prior Attempt Hx 23% Zahl DL, Hawton K: Repitition of deliberate self-harm and subsequent suicide risk: long term follow-up study of 11,583 patients. British Journal of Psychiatry, 185:70-75, % Miranda R, Scott M, Hicks R, et al.: Suicide attempt characteristics, diagnoses, and future attempts: comparing multiple attempters to single attempters and ideators. J Am Acad Child Adolesc Psychiatry, 47(1):32-40, January % Hawton K, Harriss L: Deliberate self-harm in young people: characteristics and subsequent mortality in a 20-year cohort of patients presenting to hospital. J Clin Psychiatry, 68(10): , October 2007.

Prior History Self-Inflicted Harm Minn Med.Minn Med Aug;92(8):53-5. Nonfatal suicide attempts and other self-inflicted harm. Beltrami County youths, Roesler JRoesler J, Petcoff M, Azam A, Westberg S, Kinde M, Crosby APetcoff MAzam AWestberg SKinde MCrosby A

Beltrami County Timeline 2004 Nonfatal SIH Peak: 96 3/21/2005 Red Lake murder/suicide Fall 2005 Headwaters Task Force 2005 Suicides Peak: Nonfatal SIH Decline (2 nd year in row): 61 Winter 2007 MDH contacted Fall 2007 MDH Report 2007 Suicides Decline (2 nd year in row): 8

Removing the Handle from the Broad Street Pump: the epidemic may have already been in rapid decline! “…the attacks had so far diminished before the use of the water was stopped, that it is impossible to decide whether the well still contained the cholera poison…” -John Snow

So… (from a surveillance perspective…)  Cars, Alcohol, Suicide Each kill many more…  I got my fatality numbers from Media, Wikipedia…  Real-time surveillance is hard…  Our role is in post-hoc analysis… Leading to prevention efforts

Example 2

MDH 2012 Syndromic Surveillance System For Heat-Related Illnesses and Deaths Syndromic Surveillance: Use of data from non-traditional sources (e.g., chief complaints from emergency department visits, absenteeism data, over-the-counter drug sales) in order to detect public health events earlier than possible with other methods (e.g., laboratory confirmed diagnosis, physician diagnosis).

MDH 2012 Syndromic Surveillance System For Heat-Related Illnesses and Deaths  Understand the health implications of extreme heat  Discuss current strategies to prevent heat-related morbidity and mortality  Understand available data related to morbidity and mortality related to heat  Begin to assess need for real-time data

MDH 2012 Syndromic Surveillance System For Heat-Related Illnesses and Deaths “In Minnesota, the National Weather Service (NWS) provides weather forecasts and determines the issuance of heat advisories, watches or warnings.” “All Minnesota jurisdictions involved in planning and implementing heat response plans should develop relationships with their local NWS station to ensure:  daily monitoring of weather conditions, and  early detection and transfer of information regarding the characteristics of the upcoming event”

So… (from a surveillance perspective…) The NWS is the lead agency during a heat event The role of MDH injury surveillance is probably at looking at heat-related illness/death post-hoc (after the fact)

Example 3

Predicting Trauma Admissions: The Effect of Weather, Weekday, and Other Variables Kevin A. Friede Marc C. Osborne, MD Darin J. Erickson, PhD Jon S. Roesler, MS Arsalan Azam J. Kevin Croston, MD Michael D. McGonigal, MD Arthur L. Ney, MD November 2009

One of the challenges all hospitals, especially designated trauma centers, face is how to make sure they have adequate staffing on various days of the week and at various times of the year. A number of studies have explored whether factors such as weather, temporal variation, holidays, and events that draw mass gatherings may be useful for predicting patient volume. This article looked at the effects of weather, mass gatherings, and calendar variables on daily trauma admissions at the three Level I trauma hospitals in the Minneapolis- St. Paul metropolitan area. Minn Med.Minn Med Nov;92(11):47-9. Predicting trauma admissions: the effect of weather, weekday, and other variables. Friede KAFriede KA, Osborne MC, Erickson DJ, Roesler JS, Azam A, Croston JK, McGonigal MD, Ney AL.Osborne MCErickson DJRoesler JSAzam ACroston JKMcGonigal MDNey AL

Minn Med.Minn Med Nov;92(11):47-9. Predicting trauma admissions: the effect of weather, weekday, and other variables. Friede KAFriede KA, Osborne MC, Erickson DJ, Roesler JS, Azam A, Croston JK, McGonigal MD, Ney AL.Osborne MCErickson DJRoesler JSAzam ACroston JKMcGonigal MDNey AL

ARIMA: Autoregressive integrated moving average In statistics and econometrics, and in particular in time series analysis,statisticseconometricstime series analysis an autoregressive integrated moving average (ARIMA) modelmodel is a generalization of an autoregressive moving average (ARMA) model.autoregressive moving average These models are fitted to time series data either to better understand the data or to predict future points in the series (forecasting).time seriesforecasting

Using ARIMA statistical modeling, we found that: weekends, summer, lack of rain snowfall were predictive of daily trauma admissions holidays mass gatherings such as sporting events were not predictive. The forecasting model was successful in reflecting the pattern of trauma admissions. Minn Med.Minn Med Nov;92(11):47-9. Predicting trauma admissions: the effect of weather, weekday, and other variables. Friede KAFriede KA, Osborne MC, Erickson DJ, Roesler JS, Azam A, Croston JK, McGonigal MD, Ney AL.Osborne MCErickson DJRoesler JSAzam ACroston JKMcGonigal MDNey AL

Minn Med.Minn Med Nov;92(11):47-9. Predicting trauma admissions: the effect of weather, weekday, and other variables. Friede KAFriede KA, Osborne MC, Erickson DJ, Roesler JS, Azam A, Croston JK, McGonigal MD, Ney ALOsborne MCErickson DJRoesler JSAzam ACroston JKMcGonigal MDNey AL

The forecasting model was successful in reflecting the pattern of trauma admissions. However, it's usefulness was limited in that the predicted range of daily trauma admissions was much narrower than the observed number of admissions. Nonetheless, the observed pattern of increased admission in the summer months and year-round on Saturdays should be helpful in resource planning. Minn Med.Minn Med Nov;92(11):47-9. Predicting trauma admissions: the effect of weather, weekday, and other variables. Friede KAFriede KA, Osborne MC, Erickson DJ, Roesler JS, Azam A, Croston JK, McGonigal MD, Ney AL.Osborne MCErickson DJRoesler JSAzam ACroston JKMcGonigal MDNey AL So… (from a surveillance perspective…)

So… (from a surveillance perspective…) Before / During / After

So… (from a surveillance perspective…) Before / During / After

Jon Roesler, MS Minnesota Department of Health 85 East Seventh Place, Suite 220 PO Box St. Paul, MN Voice: