1. Decision Analysis Dr M G Dawes Centre for Evidence Based Medicine

2. Contents What is Decision Analysis? Decision Trees How they work Critical Appraisal of Decision Analysis A bit on economics An example of some software

3. What is Decision Analysis? Urinary Tract Infection Patient presents with symptoms What would happen if you opted for one path in preference to another On what basis would you assess outcome

4. How does one start? List all the options and display: Commonly accepted format is a tree diagram. UTI Decision Node Treat Don’t Treat

5. Decision data UTI Chance Node Treat Don’t Treat better Not better better Not better 0.9 0.1 0.2 0.8

6. Result 0.9 is better than 0.2 UTI Chance Node Treat Don’t Treat better Not better better Not better 0.9 0.1 0.2 0.8

7. Add reality to probability scores Cost of tests Cost of treatment Days lost from work Cost of re attending clinician Multiply the probabilities by the costs

8. Cost Data UTI Treat Don’t Treat better Not better better Not better 0.9 0.1 0.2 0.8 2 9 2 9 Costs

9. Rollback Costs UTI Treat Don’t Treat better Not better better Not better 0.9 0.1 0.2 0.8 2 9 2 9 (2*0.9) + (9*0.1)=2.7 (0.2*2)+(0.8*9)=7.6

10. Results More people get better (90 % vs 20%) It is cheaper (2.70 vs 7.60)

11. What does the patient think? Utilities How would you feel? QALYs Quality adjusted life year

12. QALY’s From Alastair Gray

13. Time Trade Off Method (to assess utility of a health state) You have arthritis (severe - unable to walk to shops – need a buggy – in pain most of the time) and are aged 48 Choose between living with arthritis until 80 or living in perfect health for a shorter length of time eg 50 60 70 If 70 – 65 67 69 71 73 75 Etc until chosen a year

14. Time Trade Off Utility is 1- (number of years willing to give up/(80-current age) If age selected was 75 Utility = 1-(80-75)/(80-48) = 0.84 The better your health the less the years you give up

15. Trade Off vs Age

16. Utilities UTI Treat Don’t Treat better Not better better Not better 0.9 0.1 0.2 0.8 9 2 9 2 Utilities

17. Rollback UTI Treat Don’t Treat better Not better better Not better 0.9 0.1 0.2 0.8 9 2 9 2 (9*0.9) + (2*0.1)=8.3 (0.2*9)+(0.8*2)=3.4

18. Results More people get better (90 % vs 20%) It is cheaper (2.70 vs 7.60) The utilities are better (8.3 vs 3.4) Probably should treat?? Sensitivity analysis

19. UTI: What are the options? Treat on symptoms alone Treat after doing a test Exercise : what are the options Diagnosis Therapy Outcome

20. How does one develop this? Need to know all the baseline data. For UTI What proportion of patients with typical symptoms have UTI What is the sensitivity and specificity of a dipstix. What is the success rate of antibiotic treatment?

21. On what basis would you assess outcome? Bacterial eradication Symptoms Diary Questionnaire Interview Time off work What do patients think?

24. How are these studies appraised Are the results valid What are the results Can these be applied to my patients

25. Are the results valid? Were all important strategies and outcomes included? Were all of the realistic clinical strategies compared? Were all clinically relevant outcomes considered? Was an explicit and sensible process used to identify, select and combine the evidence into probabilities?

26. Validity Check (2) Were the utilities obtained in an explicit and sensible way from credible sources? Was the potential impact of any uncertainty in the evidence determined?

27. What are the Results? In the baseline analysis, does one strategy result in a clinically important gain for patients? If not, is the result a toss-up? How strong is the evidence used in the analysis? Could the uncertainty in the evidence change the result?

28. Can I apply the results to my patient? Do the probability estimates fit my patients' clinical features? Do the utilities reflect how my patients would value the outcomes of the decision?

29. Bottom Line The power of decision analysis is not in the numbers at the decision node It is the ability to change the utilities and probabilities Watching how this affects the decision node Thus it should be seen as a dynamic tool Software for this is available

30. EXCEL or WWW.TREEAGE.COMWWW.TREEAGE.COM ‘Nice’ system ‘Quite’ friendly Free download to try US$ 295 academic price