6 Natural History of Disease.pptx for the

It is very important to know the history of disease before investigation

1. Natural history of disease Presentation KIU Group 6

2. Objectives of Epidemiology • To determine the extent of disease in populations • To study the natural history and prognosis of disease • Identify the etiology or cause of disease and the risk factors • To evaluate both existing and new preventive and therapeutic measures and modes of health care delivery • To provide the foundation for developing public policy

3. Why study natural history of disease? •   Patients want to know their prognosis • Set priorities for clinical services and public health programs • Establish baseline natural history, to determine effectiveness of new treatments/programs • To compare effectiveness of different treatments (comparative effectiveness)

4. Preclinical Phase Clinical Phase outcome The Natural History of Disease in a p atient

5. Preclinical Phase Clinical Phase Biologic Onset of Disease Pathologic evidence of disease (if care is sought) Signs and symptoms of disease Medical care sought Diagnosis Treatment outcome Prognosis is given at diagnosis and continuously thereafter The Natural History of Disease in a p atient

6. Preclinical Phase Clinical Phase Biologic Onset of Disease Dysplasia/ invasive cancer (if care is sought) Bleeding, difficulty urinating Medical care sought Pap smear/ biopsy Chemotherapy/ radiation outcome Prognosis is given at diagnosis and continuously thereafter The Natural History of Cervical Cancer

7. Prevention •   Primary prevention – An action taken to prevent the development of disease in a person who is well and does not have the disease in question –   Vaccination •   Secondary prevention – An action taken to identify people who already have developed disease, but have not yet developed clinical signs or symptoms –   Cancer screening •   Tertiary prevention – An action taken to prevent complications in those who already have developed signs and symptoms and have been diagnosed. –   Diabetes education

8. Preclinical Phase Clinical Phase Biologic Onset of Disease Dysplasia/ invasive cancer (if care is sought) Bleeding, difficulty urinating Medical care sought Pap smear/ biopsy Chemotherapy/ radiation outcome Prognosis is given at diagnosis and continuously thereafter The Natural History of Cervical Cancer Primary Secondary Tertiary

9. Survival Preclinical Phase Clinical Phase Biologic Onset of Disease Pathologic evidence of disease (if care is sought) Signs and symptoms of disease Medical care sought Diagnosis Treatment outcome Prognosis is given at diagnosis and continuously thereafter

10. When does quantification of survival begin? • At onset of disease? • At diagnosis of disease?

11. When does quantification of survival begin? •   At diagnosis? – Dependent on stage of disease – There is not always a definitive diagnosis for a disease – Diagnosis may be after death Stage 5-year survival for Cervical Cancer IA Above 95% IB1 Around 90% IB2 Around 80%-85% IIA/B Around 75%-78% IIIA/B Around 47%-50% IV Around 20%-30%

12. Some Ways of Expressing Prognosis/Survival • Case Fatality Rate •   5-year Survival •   Observed Survival • Median Survival Time • Relative Survival

13. Total number of individuals dying during a specified period of time after disease onset Number of individuals with the disease of interest Case - Fatality Rate = X 100 Case Fatality Rate

14. Case Fatality Rate • Expresses the likelihood of death given that disease is present –   Proportion • Time is not explicitly stated, but implied by the natural history of the disease • Competing risks of death render this less useful in chronic conditions

15. Five-Year Survival • Proportion of patients alive 5 years after treatment begins or after diagnosis Stage 5-year survival for Cervical Cancer IA Above 95% IB1 Around 90% IB2 Around 80%-85% IIA/B Around 75%-78% IIIA/B Around 47%-50% IV Around 20%-30%

16. 5-Year Survival Year of Treatment No. Subjects Number Alive on Treatment Anniversary 2001 2002 2003 2004 2005 2000 84 44 21 13 10 8 2001 62 31 14 10 6 2002 93 50 20 13 2003 60 29 16 2004 76 43

17. 5-Year Survival Year of Treatment No. Subjects Number Alive on Treatment Anniversary 2001 2002 2003 2004 2005 2000 84 44 21 13 10 8 2001 62 31 14 10 6 2002 93 50 20 13 2003 60 29 16 2004 76 43 8 / 84 = 9.5% survive 5 years

18. Five-Year Survival • Used as an index of treatment success –    A way to compare treatments •    Can also be expressed as 1-, 3-, 10-year survival etc. •    Limitations –    Assumes fixed observation period for all subjects –    Cannot assess prognosis when survival experience is less than 5 years

19. 5-Year Survival Year of Treatment No. Subjects Number Alive on Treatment Anniversary 2001 2002 2003 2004 2005 2000 84 44 21 13 10 8 2001 62 31 14 10 6 2002 93 50 20 13 2003 60 29 16 2004 76 43 8 / 84 = 9.5% survive 5 years

20. Observed Survival •   Life-table –   Uses predetermined intervals •   Kaplan-Meier –   Uses exact time of event to determine intervals • Common methods for estimating survival when the length of observation varies across participants

21. Life Tables

22. Hypothetical Study of Treatment (No loss to follow-up) Year of Treatment No. Subjects Number Alive on Treatment Anniversary 2001 2002 2003 2004 2005 2000 84 44 21 13 10 8 2001 62 31 14 10 6 2002 93 50 20 13 2003 60 29 16 2004 76 43

23. Rearrange Data Showing Survival since Enrollment Year No. Subjects Number Alive at End of Year 1 Yr st 2 Yr nd 3 Yr rd 4 Yr th 5 th Yr 2000 84 44 21 13 10 8 2001 62 31 14 10 6 2002 93 50 20 13 2003 60 29 16 2004 76 43 Year of Treatment No. Subjects Number Alive on Treatment Anniversary ‘01    ‘02    ‘03    ‘04 ‘05 2000 84 44 21 13 10 8 2001 62 31 14 10 6 2002 93 50 20 13 2003 60 29 16 2004 76 43

24. Rearrange Data Showing Survival since Enrollment Year No. Subjects Number Alive at End of Year 1 Yr st 2 Yr nd 3 Yr rd 4 Yr th 5 Yr th 2000 84 44 21 13 10 8 2001 62 31 14 10 6 2002 93 50 20 13 2003 60 29 16 2004 76 43 Year of Treatment No. Subjects Number Alive on Treatment Anniversary ‘01    ‘02    ‘03    ‘04 ‘05 2000 84 44 21 13 10 8 2001 62 31 14 10 6 2002 93 50 20 13 2003 60 29 16 2004 76 43

25. Analysis of Survival Data Year No. Subjects Number Alive at End of Year 1 Yr st 2 nd Yr 3 rd Yr 4 Yr th 5 Yr th 2000 84 44 21 13 10 8 2001 62 31 14 10 6 2002 93 50 20 13 2003 60 29 16 2004 76 43 Total

26. Analysis of Survival Data Year No. Subjects Number Alive at End of Year 1 Yr st 2 nd Yr 3 rd Yr 4 Yr th 5 Yr th 2000 84 44 21 13 10 8 2001 62 31 14 10 6 2002 93 50 20 13 2003 60 29 16 2004 76 43 Total 375 197 71 36 16 8

27. Analysis of Survival Data Year No. Subjects Number Alive at End of Year 1 Yr st 2 nd Yr 3 rd Yr 4 Yr th 5 Yr th 2000 84 44 21 13 10 8 2001 62 31 14 10 6 2002 93 50 20 13 2003 60 29 16 2004 76 43 Total 375 197 71 36 16 8 P = Probability of Surviving 1 year = 197/375 = 0.525 1 st

28. Analysis of Survival Data Year No. Subjects Number Alive at End of Year 1 Yr st 2 nd Yr 3 rd Yr 4 Yr th 5 Yr th 2000 84 44 21 13 10 8 2001 62 31 14 10 6 2002 93 50 20 13 2003 60 29 16 2004 76 43 Total 375 197 71 36 16 8 P = Probability of surviving 2 2 nd year = 71/(197-43) = 0.461

29. Analysis of Survival Data Year No. Subjects Number Alive at End of Year 1 Yr st 2 nd Yr 3 rd Yr 4 Yr th 5 Yr th 2000 84 44 21 13 10 8 2001 62 31 14 10 6 2002 93 50 20 13 2003 60 29 16 2004 76 43 Total 375 197 71 36 16 8 P = Probability of surviving 3 3 rd year = 36/(71-16) = 0.655

30. Analysis of Survival Data Year No. Subjects Number Alive at End of Year 1 st Yr 2 nd Yr 3 rd Yr 4 th Yr 5 th Yr 2000 84 44 21 13 10 8 2001 62 31 14 10 6 2002 93 50 20 13 2003 60 29 16 2004 76 43 Total 375 197 71 36 16 8 P = Probability of surviving 4 4 th year = 16/(36-13) = 0.696

31. Analysis of Survival Data Year No. Subjects Number Alive at End of Year 1 st Yr 2 nd Yr 3 rd Yr 4 th Yr 5 th Yr 2000 84 44 21 13 10 8 2001 62 31 14 10 6 2002 93 50 20 13 2003 60 29 16 2004 76 43 Total 375 197 71 36 16 8 P = Probability of surviving 5 5 th year = 8/(16-6) = 0.80

32. Annual and Cumulative Probabilities of s urvival Year Annual Probability Survival of Cumulative Probability of Survival 1 0.525 0.525 2 0.461 0.525*0.461 = 0.241 3 0.655 0.525*0.461*0.655 = 0.159 4 0.696 0.525*0.461*0.655*0.696 = 0.11 5 0.80 0.525*0.461*0.655*0.696*0.80 = 0.088

33. Calculating the Life Table P x p x q x l’ x (l –  0.5*w ) x x w x d x l x x 6 2 16 5 th yr 13 7 36 4 th yr 16 19 71 3 rd yr 43 83 197 2 nd yr 178 375 1 st yr Cumulative Survived Died Effective Number at Risk With- drew Died during interval Alive at start of interval Interval --- Proportion ---

34. Analysis of Survival Data Year No. Subjects Number Alive at End of Year 1 Yr st 2 nd Yr 3 rd Yr 4 Yr th 5 Yr th 2000 84 44 21 13 10 8 2001 62 31 14 10 6 2002 93 50 20 13 2003 60 29 16 2004 76 43 Total 375 197 71 36 16 8

35. Calculating the Life Table P x p x q x l’ x (l –  0.5*w ) x x w x d x l x x 6 2 16 5 th yr 13 7 36 4 th yr 16 19 71 3 rd yr 43 83 197 2 nd yr 178 375 1 st yr Cumulative Survived Died Effective Number at Risk With- drew Died during interval Alive at start of interval Interval --- Proportion ---

36. Effective Number at Risk l’   = (l –  0.5*w ) x x x • Assume those that withdrew, withdrew uniformly throughout the interval • Mean time of withdrawal would be at the midpoint of the year (0.5) l ’ x = Number at beginning of interval –  (0.5*withdrawals)

37. 175.5 375.0 Calculating the Life Table P x p x q x l’ x (l –  0.5*w ) x x w x d x l x x 6 2 16 5 th yr 13 7 36 4 th yr 16 19 71 3 rd yr 43 83 197 2 nd yr 178 375 1 st yr Cumulative Survived Died Effective Number at Risk With- drew Died during interval Alive at start of interval Interval --- Proportion ---

38. 13.0 29.5 63.0 Calculating the Life Table 175.5 375.0 P x p x q x l’ x (l –  0.5*w ) x x w x d x l x x 6 2 16 5 th yr 13 7 36 4 th yr 16 19 71 3 rd yr 43 83 197 2 nd yr 178 375 1 st yr Cumulative Survived Died Effective Number at Risk With- drew Died during interval Alive at start of interval Interval --- Proportion ---

39. Calculating the Life Table •   Proportion who died during interval ( q ) x – Died during interval/Effective # at risk of dying • Proportion who survived during interval ( p ) x –   1- proportion who died • Cumulative proportion who survived from enrollment through end of interval ( P ) x –   Multiply proportion who survived each interval –   Cumulative 5-year survival • 0.525 x 0.527 x 0.698 x 0.763 x 0.846 = 0.124

40. 0.277 0.527 0.473 0.525 0.525 0.475 Calculating the Life Table 13.0 29.5 63.0 175.5 375.0 P x p x q x l’ x (l –  0.5*w ) x x w x d x l x x 6 2 16 5 th yr 13 7 36 4 th yr 16 19 71 3 rd yr 43 83 197 2 nd yr 178 375 1 st yr Cumulative Survived Died Effective Number at Risk With- drew Died during interval Alive at start of interval Interval --- Proportion ---

41. 0.124 0.846 0.154 0.147 0.763 0.237 0.193 0.698 0.302 Calculating the Life Table 0.277 0.527 0.473 0.525 0.525 0.475 13.0 29.5 63.0 175.5 375.0 P x p x q x l’ x (l –  0.5*w ) x x w x d x l x x 6 2 16 5 th yr 13 7 36 4 th yr 16 19 71 3 rd yr 43 83 197 2 nd yr 178 375 1 st yr Cumulative Survived Died Effective Number at Risk With- drew Died during interval Alive at start of interval Interval --- Proportion ---

42. Life Table: Cumulative Survival

43. Kaplan –  Meier Method

44. Kaplan-Meier Method Patient 1 Patient 2 Patient 4 Patient 5 Patient 6 Patient 3 Died Died Died Died Lost to follow-up Lost to follow-up 4 10 14 24 Months since enrollment

45. Survival Analysis with Kaplan-Meier Method Time to Death No. Alive at each Time No. Died at each Time Died at that Time Survived at that Time Cumulative Survival ----- Proportion ----- 4 6 1 10 4 1 14 3 1 24 1 1

46. Survival Analysis with Kaplan-Meier Method Time to Death No. Alive at each Time No. Died at Died at that each Time Time Survived at that Time Cumulative Survival ----- Proportion ----- 4 6 1 0.167 10 4 1 14 3 1 24 1 1

47. Survival Analysis with Kaplan-Meier Method Time to Death No. Alive at each Time No. Died at each Time Died at that Time Survived at that Time Cumulative Survival ----- Proportion ----- 4 6 1 0.167 10 4 1 0.250 14 3 1 0.333 24 1 1 1.000

48. Survival Analysis with Kaplan-Meier Method Time to Death No. Alive at each Time No. Died at each Time Died at that Time Survived at that Time Cumulative Survival ----- Proportion ----- 4 6 1 0.167 0.833 10 4 1 0.250 14 3 1 0.333 24 1 1 1.000

49. Survival Analysis with Kaplan-Meier Method Time to Death No. Alive at each Time No. Died at Died at that each Time Time Survived at that Time Cumulative Survival ----- Proportion ----- 4 6 1 0.167 0.833 10 4 1 0.250 0.750 14 3 1 0.333 0.667 24 1 1 1.000 0.000

50. Survival Analysis with Kaplan-Meier Method Time to Death No. Alive at each Time No. Died at each Time Died at that Time Survived at that Time Cumulative Survival ----- Proportion ----- 4 6 1 0.167 0.833 0.833 10 4 1 0.250 0.750 14 3 1 0.333 0.667 24 1 1 1.000 0.000

51. Survival Analysis with Kaplan-Meier Method Time to Death No. Alive at each Time No. Died at Died at that each Time Time Survived at that Time Cumulative Survival ----- Proportion ----- 4 6 1 0.167 0.833 0.833 10 4 1 0.250 0.750 0.625 14 3 1 0.333 0.667 0.417 24 1 1 1.000 0.000 0.000

52. Kaplan-Meier Method 0.833 0.625 0.417 0.000

53. Kaplan-Meier Method

54. Median Survival Time •   Length of time that one-half of the study population survives •   Why median over mean survival? –  Less affected by extreme values –  Cannot calculate mean until all persons in the study die (or reach study endpoint)

55. 0.124 0.846 0.154 0.147 0.763 0.237 0.193 0.698 0.302 Median Survival (Life Table) 0.277 0.527 0.473 0.525 0.525 0.475 13.0 29.5 63.0 175.5 375.0 P x p x q x l’ x (l –  0.5*w ) x x w x d x l x x 6 2 16 5 th yr 13 7 36 4 th yr 16 19 71 3 rd yr 43 83 197 2 nd yr 178 375 1 st yr Cumulative Survived Died Effective Number at Risk With- drew Died during interval Alive at start of interval Interval --- Proportion ---

56. Median Survival: Kaplan-Meier

57. Life Tables vs. Kaplan-Meier •   Life Table –   Interval –   Time (year) used to define interval –   Graph smooth •   Kaplan-Meier –   Exact time –   Data (deaths) are used to define intervals –   Graph is step function Life table and Kaplan Meier will provide very similar results if sample size is large and intervals are small.

58. Relative Survival Compare survival for two (or more) groups of patients = Relative Survival Observed survival in people with disease Expected survival if disease were absent

59. Relative Survival (Eaker et al, 2006) Expected survival based upon the age- and period-specific of the general population of females in the Uppsala/Orebro health care region.

60. Assumptions for Life Tables and KM • No secular or temporal change in the effectiveness of treatment or in survivorship over calendar time • Survival experience among people lost to follow-up is the same as those who remain under observation

61. Temporal Trend Assumption • Secular or temporal change in survival: most relevant in long-term studies –   Treatments may improve over time –   Diagnostics tests may improve –   Access to health care may change –   Economy may improve or worsen –   Health care status may improve or worsen • Evaluate by assessing survival experience early and late in the study period

62. Temporal Trend Assumption

63. Lost to Follow-Up Assumption • Survival experience in subjects lost to follow-up is similar to those who remain under study • Reasons for losses vary and may relate to their disease –   Death from disease –   Death from competing risk – Recover from disease and drop out of study –   Move from the area or seek care elsewhere • Large losses to follow-up threaten the validity of any study (90% rule) • If you have losses, compare baseline prognostic characteristics between subjects who remain in the cohort to those lost

64. Lost to Follow-Up Assumption

65. Lead Time Bias •   Overestimation of survival time, due to the backward shift in the starting point for measuring survival when diseases are detected earlier •   Screening

66. Lead Time Bias Biologic Onset of Disease Death Diagnosis Observed survival 1995 2002 2005 Detected by screening: Diagnosis & treatment 2000 Observed survival Lead time

67. Generalizability of Survival •   Generalizability, also called external validity, is the degree to which results apply to the underlying target population • Nature of study sample must reflect the target population –   May differ due to •   Inclusion and exclusion criteria •   Self-referral/volunteers/doctor referrals •   Refusals

68. Review • Natural history of disease • Prevention • Quantifying survival –    Case-fatality rate –    5-year survival –    Life tables –    Kaplan-Meier method –    Median Survival –    Relative Survival • Assumptions and issues