Top Banner
simple statistics
25

Medical Statistics (basis)

Jul 03, 2015

Download

Healthcare

Simple Medical Statistics (basis)
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Medical Statistics (basis)

simple statistics

Page 2: Medical Statistics (basis)

simple statisticssensitiviteit

Page 3: Medical Statistics (basis)

sensitiviteit = hoe vaak is de test positief als de ziekte er is

sensitiviteit = 18/(18+4) =82%

Page 4: Medical Statistics (basis)

simple statisticsspecificiteit

Page 5: Medical Statistics (basis)

specificiteit =hoe vaak is de test negatief als de ziekte er niet is

specificiteit = 23/(23+4) =85%

Page 6: Medical Statistics (basis)

simple statisticsPos. voorspellende waarde

Page 7: Medical Statistics (basis)

PPV =als de test positief is, hoe groot is dan de kans dat de zieket er is

PPV = 18/(18+4) =82%

Page 8: Medical Statistics (basis)

simple statisticsNeg. voorspellende waarde

Page 9: Medical Statistics (basis)

NPV =als de test negatief is, hoe groot is dan de kans dat de ziekte er niet is

NPV = 23/(23+4) =85%

Page 10: Medical Statistics (basis)

simple statisticsLikelihoof ratio

Page 11: Medical Statistics (basis)

LR = als de test positief is, hoeveel groter is de kans dan dat patient ziek is tov niet

LR =sensitiviteit/(100%-specificiteit)

Dus hier: 82%/(100-85%) = 5.5

Page 12: Medical Statistics (basis)

simple statisticsStandaard deviatie

Page 13: Medical Statistics (basis)
Page 14: Medical Statistics (basis)
Page 15: Medical Statistics (basis)

simple statisticsbetrouwbaarheidsinterval

Page 16: Medical Statistics (basis)
Page 17: Medical Statistics (basis)

simple statisticsP-waarden

Page 18: Medical Statistics (basis)

P-waarden

P value: the probability of any observed difference having

happened by chance

p = 0.05 = significantp = 0.01 = sterk significantp = 0.001 = zeer sterk significant

Page 19: Medical Statistics (basis)

P-waarden

Valkuilen:- te kleine studie- te grote studie- ook klinisch significant?

Page 20: Medical Statistics (basis)

Risk / Odds Ratio1000 voetballers en 1000 niet-voetballers: # eerste 12, tweede 4

Risk?Risk ratio?Odds ratio?

Page 21: Medical Statistics (basis)

Risk / Odds RatioRisk voetballers 12/125 = 0.096, anderen 4/134 = 0.030Risk ratio: 0.096 gedeeld door 0.030 = 3,2Odds # voetballers: 12/113 =0.11, bij anderen 4/130 =0.031Odds ratio: 0.11/0.031 = 3.55

Page 22: Medical Statistics (basis)
Page 23: Medical Statistics (basis)

Correlatie

Page 24: Medical Statistics (basis)
Page 25: Medical Statistics (basis)

http://sumed.sun.ac.za/Portals/0/Repository/Medical-Statistics-Made-Easy.3f8ceb88-f35b-4f29-8d70-17e78ce071d8.pdf