1. Trials and tribulations in 25 years of pharmacoepidemiology Samy Suissa McGill University and Jewish General Hospital April 29-30, 2015

2. The daring 90s

5. SASKATCHEWAN The Saskatchewan study: 12,000 asthma patients  Healthcare databases  Computerized  Doctor visits  Medications  Hospitalizations  Death certificates

7. Risk of asthma death with bronchodilators

10. Asthma death and inhaled corticosteroids Suissa et al., N Engl J Med 2000;332–6.

13. Worldwide impact?

14. Japan Portugal Israel New Zealand

15. The roaring 2000s

16. CPRD: UK Clinical Practice Research Datalink  >1000 GPs  10 million patients  1990-today  Diagnoses, lab tests  Drug prescriptions, hospitalizations, etc..

18. The painkiller Vioxx

19. WASHINGTON (Reuters) -- Pharmaceutical giant Merck & Co Inc. had evidence by 2000 that its painkiller Vioxx, which was pulled off the market on Sept. 30 (2004), was not safe, a heart specialist told CBS News program "60 Minutes" on Sunday. Monday, November 15, 2004 Posted: 1516 GMT (2316 HKT)

23. Parkinson’s disease medications

28. Pharmaco- epidemiology methods

30. Time-related biases

41. The soaring 2010s

44. Canadian Pharmacoepidemiology  Healthcare databases  Computerized  Doctor visits  Medications  Hospitalizations  Death certificates

45. Why CNODES ?

46. WASHINGTON (Reuters) -- Pharmaceutical giant Merck & Co Inc. had evidence by 2000 that its painkiller Vioxx, which was pulled off the market on Sept. 30 (2004), was not safe, a heart specialist told CBS News program "60 Minutes" on Sunday. Monday, November 15, 2004 Posted: 1516 GMT (2316 HKT)

48. Ontario Quebec Saskatchewan VIGOR RCT Canadian VIOXX observational studies

49. Ontario Quebec Saskatchewan VIGOR RCT Canadian VIOXX observational studies British Columbia? Alberta? Manitoba? Nova Scotia?

50. British Columbia Alberta 1 - 2 - Nova Scotia *CPRD MarketScan CNODES Quebec Ontario Manitoba Saskatchewan

52. CNODES: Challenges  Small drug effects  Large sample sizes  Computerized databases  Consistent results  Rapid

53. High-dose statins and the risk of acute kidney injury

54. BACKGROUND: Ridker PM et al, JUPITER Study Group. N Engl J Med 2008;359 Event Rosuvastatin 20mg n=8901 Placebo n=8901 Rate RatioP-value n%n% Primary Endpoint 1421.62512.80.57<0.0001 Renal disorder 5356.04805.41.190.08 Does treatment with high potency statins increase the risk of serious acute kidney injury ?

56. METHODS/RESULTS  9 new-user cohorts of statin initiators  Over 2 million patients initiating statin therapy  1/3 starting on high-dose statins  Over 24,000 hospitalized for AKI during the first 2 years of follow-up

57. HIGH-POTENCY STATINS AND AKI

58. CNODES Studies

59. CNODES Methods Team

60. 25 years later ?

62. Visit us at: WWW.CNODES.CA