Using Data Mining to Predict Injuries Mining - Injury Prediction - Steve... · Astrology and...
Transcript of Using Data Mining to Predict Injuries Mining - Injury Prediction - Steve... · Astrology and...
Using Data Mining to Predict Injuries:Using Data Mining to Predict Injuries:
A new model for injury prediction and preventionA new model for injury prediction and prevention
Steve Thompson, ARM, COSSSteve Thompson, ARM, [email protected]@aspenrmg.com
Aspen Risk Management GroupAspen Risk Management Group
ERGOhealthyERGOhealthy
ASSE Member since 1989ASSE Member since 1989
2054
2
PredictionPrediction
The way things will happen in the future
Large corporations invest heavily on possibleevents, risks and business opportunities
Prediction in business "... is at least two things –important and hard“
3
Supernatural predictionSupernatural prediction
Paranormal or supernatural
Astrology and “occult” sciences
Not substantiated by controlledexperiments
One in 10 people in the world areconcerned that the world couldend in 2012 (Reuters poll)
4
Paranormal or supernatural
Astrology and “occult” sciences
Not substantiated by controlledexperiments
One in 10 people in the world areconcerned that the world couldend in 2012 (Reuters poll)
7
PrecogPrecog
NostradamusNostradamus
• 16th =century French “seer”
• Studied astrology and “occult”sciences
• Best known for The Prophecies
• Louis Pasteur
Nostradamus predictionsNostradamus predictions
• Atomic bomb
Nostradamus predictionsNostradamus predictions
• Hitler and World War II
Nostradamus predictionsNostradamus predictions
• Sept. 11, 2011
Nostradamus predictionsNostradamus predictions
• The end of the world
Nostradamus predictionsNostradamus predictions
14
Edward Cayce (theEdward Cayce (the ““Sleeping ProphetSleeping Prophet””))
Two World Wars (start and end dates)
The economic crisis in 1929
Fall of communism and the collapse ofthe Soviet Union
Revival of Soviet Union to begin in 2010
Growth of political power in China
Global warming
Scientific predictionScientific prediction
In science a prediction is rigorous and quantitative
Computer models are frequently used to describethe past and future behavior of something
Accurate workplace injury prediction and forecastingis very difficult
In insurance an actuary uses a life table to “predict”life expectancy
““Life tableLife table””
Scientific predictionScientific prediction
Neptune was predicted by Sir IsaacNewton
Radio waves were predicted by JamesClerk Maxwell
Anti-matter was predicted by Paul Dirac
Old Farmer’s Almanac
Injury prediction studiesInjury prediction studies
Early prognostic factors for duration ontemporary total benefits in the first year amongworkers with compensated occupational softtissue injuries http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1740514/
Chronic musculoskeletal pain predictedhospitalization due to serious medical conditionsin a 10 year follow up study http://www.biomedcentral.com/1471-2474/11/127
Injury prediction studiesInjury prediction studies
Prediction of chronic disability in work-relatedmusculoskeletal disorders: a prospective,population-based study http://www.biomedcentral.com/1471-2474/5/14
Early predictors of chronic work disability: aprospective, population-based study of workerswith back injuries (ISSLS prize winner: http://www.ncbi.nlm.nih.gov/pubmed/19050587)
Injury prediction studiesInjury prediction studies
Listening to injured workers: how recoveryexpectations predict outcomes--a prospectivestudy http://www.ncbi.nlm.nih.gov/pmc/articles/PMC99453/?tool=pubmed
Worker recovery expectations and fear-avoidancepredict work disability in a population-basedworkers' compensation back pain samplehttp://www.ncbi.nlm.nih.gov/pubmed/16540874
Injury prediction studiesInjury prediction studies
Eight predictive factors associated with responsepatterns during physiotherapy for soft tissueshoulder disorders were identifiedhttp://www.ncbi.nlm.nih.gov/pubmed/16632137
High pain ratings predict treatment failure inchronic occupational musculoskeletal disordershttp://www.ncbi.nlm.nih.gov/pubmed/16452743
Injury prediction studiesInjury prediction studies
Biopsychosocial multivariate predictive model ofoccupational low back disabilityhttp://www.ncbi.nlm.nih.gov/pubmed/12461399
Injury prediction studiesInjury prediction studies
Life values as predictors of pain, disability andsick leave among Swedish registered nurses: alongitudinal study http://www.ncbi.nlm.nih.gov/pubmed/21958331
Injury prediction studiesInjury prediction studies
A model for occupational injury risk assessmentof musculoskeletal discomfort and theirfrequencies in computer usershttp://www.sciencedirect.com/science/article/pii/S0925753510000822
Injury prediction studiesInjury prediction studies
Symptoms of stress predict musculoskeletaldisorders http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1052853/
Prediction todayPrediction today
Baby Gender
Future Height
Tornado, Hurricane, Weather
Earnings of a company (earnings per share)
“Your Future”
Predict Your Social Security Payout
How to predict Mitt Romney’s running mate
Computer predicts sports injuries (experimental)
Common safety and healthCommon safety and health ““predictorspredictors””
Gas detection devices
Health risk survey
Safety culture, and other safety and ergo surveys
Severity index (predictive analytics in workers’ comp)
Underwriters use predictive modeling to price risk
Peer behavior, and environmental observations
Pre and new hire orientation
Casualty prediction
Strains, sprains, and discomfortStrains, sprains, and discomfort
40% of WC costs = musculoskeletal disorders
Cisco and Accenture pilot studies (10,000 EEs)
Persistent discomfort - pulled off task 5 minutes outevery 15 minutes
Lost time/day/employee: 2.6 hours
Lost time/week/employee: 13 hours
62% decrease in reported discomfort
39% decrease in claims for ergonomic injuries
Minutes add upMinutes add up
13.3 hours/week x 4 weeks = 53.3 hours of lostproduction/employee/month
Cost of lost production for 1 employee:
$15/hr x 53.3 hours = $799.50/month
$25/hr x 53.3 hours = $1,332.50/month
$50/hr x 53.3 hours = $2,665/month
*(20 min. per hour) x (8 hours) x (5 days) x (4 weeks)
Pain and discomfort predicts productivity loss
Injury risk assessmentInjury risk assessment –– why not addwhy not add
Safety inspections
Safety audits
Incident investigations
Near misses
Review of preventative maintenance records
… and on and on...
Why notWhy not……
Fear of what’s “lurking beneath the surface”
Fear of cost to “fix”
Fear…
What to do with the information
What “bucket” does it fit in
Now what
No action, or poor planning – post data mining
Would youWould you……
Introduce process to determine culture andrisk levels
Explore with employees discomfort and risklevels to improve productivity and predictinjuries
Data-mine existing records – health, safety,workers’ compensation
Be prepared to “do something” about yourfindings
If you were wonderingIf you were wondering
Scientific date for the destruction of the planet
Unknown
Precisely because of human fallibility,extraordinary claims require
extraordinary evidence ~ Carl Sagan
Using Data Mining to Predict Injuries:Using Data Mining to Predict Injuries:
A new model for injury prediction and preventionA new model for injury prediction and prevention
Steve Thompson, ARM, COSSSteve Thompson, ARM, [email protected]@aspenrmg.com
Aspen Risk Management GroupAspen Risk Management Group
ERGOhealthyERGOhealthy
ASSE Member since 1989ASSE Member since 1989
2054
Summary and QASummary and QA