Overdiagnostik — en uundgåelig konsekvens af medicinsk ...jydskmedicinskselskab.org/wp-content/uploads/2015/... · –en uundgåelig konsekvens af medicinsk screening John Brodersen

Overdiagnostik – en uundgåelig konsekvens af medicinsk screening

John Brodersen Professor, speciallæge i almen medicin, ph.d.

Center for Forskning & Uddannelse i Almen Medicin, IFSV, KUForskningsenheden for Almen Praksis, Region Sjælland

[email protected]

mailto:[email protected]

Overdiagnostik v. screening

Post-moderne medicin

Definition

Overdetektion ved screening

Hvorfor?

Eksempler

Stadieskifte

Incidens og mortalitet

Konsekvenser af overdiagnostik

Fremtiden?

2



Definition


Hvorfor?

Eksempler

Stadieskifte



Fremtiden?

3

Traditionel medicin

symptomer

klinisk undersøgelse

4

Curative Medicine

Help!

5

Moderne medicin

symptomer

klinisk undersøgelse

mikroskopi/mikrobiologi/medikoteknologi/epidemiologi

afvigelse/abnormitet/risikofaktor/patologi

diagnose6

Preventive Medicine

I have something to offer you

7




korrekt diagnose + overdiagnostik

illness

8




korrekt diagnose + overdiagnostik

illness

9



Definition


Hvorfor?

Eksempler

Stadieskifte



Fremtiden?

10

Overdiagnostik - definition

“Overdiagnosis is the diagnosis of ‘illnesses’ that would never have caused patients harm but potentially exposes them to treatments where the risks outweigh the benefits.”

Doust & Glasziou. Is the problem that everything is a diagnosis?Australian Family Physician 42:856-859, 2013.

11

Overdiagnostik - beskrivelse

“Overdiagnosis occur when individuals are diagnosed with conditions that will never cause symptoms or death”

“…the ultimate criterion for overdiagnosis: at the end of life, if the person never developed a problem from her condition, she has been overdiagnosed”

Welch, Schwartz, Woloshin. Overdiagnosed. Making People Sick in the Pursuit of Health, Boston: Beacon Press, 2011.

12

Overdiagnostik - min definition

Diagnosticering af afvigelser, abnormiteter, risikofaktorer og/eller patologiske forandringer, som uopdaget eller udiagnosticeret aldrig i sig selv vil;

give personen symptomer (eks. afvigelser),

vil medføre sygelighed eller

vil være årsag til personens død.

13

Overdiagnostik - 3 områder

Nedsætte grænseværdier og/eller udvide sygdomsklassifikation

Opfinde nye ”sygdomme” (disease mongering)


14

Overdiagnostik - 3 områder

Nedsætte grænseværdier og/eller udvide sygdomsklassifikation

Opfinde nye ”sygdomme” (disease mongering)


15



Definition


Hvorfor?

Eksempler

Stadieskifte



Fremtiden?

16

The Screening Cascade

Negative

screening result

Positive

screening result

People who

are screened

True positive False positive

Treatment

works better

early than later

Rapidly

progressive

disease; person

would die even if

treated

Mild, easily

treatable disease;

person would do

well even if

treated later

Person would

never have

developed

symptoms, even

if untreatedBenefit

No Benefit

Incidental

finding

Separate

cascade

Workup

Treatment

Indeterminate

findingSurveillance

Modified slide: Professor Russ Harris 17


Negative

screening result

Positive

screening result

People who

are screened


Treatment

works better

early than later

Rapidly

progressive

disease; person

would die even if

treated

Mild, easily

treatable disease;

person would do

well even if

treated later

Person would

never have

developed

symptoms, even

if untreatedDelayed Benefit

No Benefit

Incidental

finding

Separate

cascade

Workup

Treatment

Indeterminate

findingSurveillance

Types of “true positives”



Negative

screening result

Positive

screening result

People who

are screened


Treatment

works better

early than later

Rapidly

progressive

disease; person

would die even if

treated

Mild, easily

treatable disease;

person would do

well even if

treated later

Person would

never have

developed

symptoms, even

if untreatedDelayed Benefit

No Benefit

Incidental

finding

Separate

cascade

Workup

Treatment

Indeterminate

findingSurveillance

Types of “true positives”

Immediate harms/costs



Negative

screening result

Positive

screening result

People who

are screened


Treatment

works better

early than later

Rapidly

progressive

disease; person

would die even if

treated

Mild, easily

treatable disease;

person would do

well even if

treated later

Person would

never have

developed

symptoms, even

if untreated

No Benefit

Incidental

finding

Separate

cascade

Workup

Treatment

Indeterminate

findingSurveillance

Overdiagnosis

Delayed Benefit


Screening - på godt og ondt

GODT

Nedsat morbiditet

og mortalitet

Mere skånsom

behandling

Tryghed for

ikke at være syg

Primær forebyg-

gende effekt

ONDT

Patienter i flere år

Overdiagnostik

Falsk tryghed

Falsk alarm

Overbehandling

Øget morbiditet

og mortalitet

Induceret sygdom

21

Screening - på godt og ondt

GODT

Nedsat morbiditet

og mortalitet

Mere skånsom

behandling

Tryghed for

ikke at være syg

Primær forebyg-

gende effekt

ONDT

Patienter i flere år

Overdiagnostik

Falsk tryghed

Falsk alarm

Overbehandling

Øget morbiditet

og mortalitet

Induceret sygdom

22



Definition


Hvorfor?

Eksempler

Stadieskifte



Fremtiden?

23

Mammografiscreening

24

17.08.05 17.11.05 01.03.06 16.10.06 10.10.07

29.11.08 26.02.09 10.08.09 10.08.09 17.08.09

Lungecancerscreening m. CT-skanning

25

Natural History of Lung Cancer

R. M. Lindell, T. E. Hartman, S. J. Swensen, J. R. Jett, D. E. Midthun, and J. N. Mandrekar. 5-year lung cancer screening experience: growth curves of 18 lung cancers compared to histologic type, CT attenuation, stage, survival, and size. Chest. 136 (6):1586-1595, 2009.

26

Heterogenitet i vækstrater ved AAA-screening

Thompson et al. Systematic review and meta-analysis of the growth and rupture rates of small abdominal aortic aneurysms: implications for surveillance intervals and their cost-effectiveness. Health Technol.Assess. 17 (41):1-118, 2013.

27

Person AFalse negative screen

from very fast growing

cancer

Size at which cancer can be detected

Abnormal cell

Size at which cancer causes symptoms

Cancer size

Size at which cancer causes death

Death from other causes

Person HOverdiagnosed: Cancer

spontaneously regressed

before death from other cause

Person FOverdiagnosed: Cancer would

NOT have caused symptoms

before death from other causes

Person GNo cancer diagnosis before

death from other cause

Screening

Test 1

Screening

Test 2

Screening

Test 3

Screening

Test 4

Screening

Test 5

Cancer

diagnosis

Cancer

diagnosis Cancer

diagnosisCancer

diagnosis

Time

Death from other causes

Persons B, C, D & EScreen-diagnosed cancer which would

have progressed to symptoms and death

Persons B, C, D & EScreen-diagnosed cancer which would

have progressed to symptoms and death

Model: what happens at cancer screening?

Brodersen J., Schwartz L.M., Woloshin S. Overdiagnosis: How cancer screening can turn indolent pathology into illness. APMIS 122, 2014.



Definition


Hvorfor?

Eksempler

Stadieskifte



Fremtiden?

29

AAA screening: 38-44% ODx

M. Johansson, A. Hansson, and J. Brodersen. Estimatingoverdiagnosis in screening for abdominal aorticaneurysm: could a change in smoking habits and loweredaortic diameter tip the balance of screening towardsharm? BMJ 350:h825, 2015.

30

AAA screening: 38-44% ODx

http://www.bmj.com/content/350/bmj.h825/infographic

31

http://www.bmj.com/content/350/bmj.h825/infographic

Breast, mammography

Lung, chest x-ray

Lung, CT

Liver, ultrasound

Prostate, PSA

Krag et al. Quantification of overdiagnosis in randomisedtrials of cancer screening: an overview of systematicreviews. POD conference 201532

Breast, mammography

Lung, chest x-ray

Lung, CT

Liver, ultrasound

Prostate, PSA

Krag et al. Quantification of overdiagnosis in randomisedtrials of cancer screening: an overview of systematicreviews. POD conference 201533

Overdiagnostik (ODx) vedlungecancer CT screening

NLST: 18,5% ODx

Kontrolgruppen: Rtg. af thorax

Mayo Lung Project: 51% ODx

Best case: 18,5% ODx

Worst case: 40,3% ODx

34

Danish lung cancer screeningtrial (DLCST): design

Pedersen et al. The Danish Randomized Lung Cancer CT Screening Trial – Overall Design and Results of the Prevalence Round. Journal of Thoracic Oncology 4 (5):608-614, 2009.

35

Hvor meget ODx i DLCST?

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Best case: 18,5% ODx

Worst case: 40,3% ODx

36

Overdiagnosis in DLCST at 5 year follow-up

37

96

53


68

24

38

68

24

96

5344

43


39

96

53

43


68

24

44

40

Extra antal af LC: 43 (96-53)

ODx: 43/68=63% [95% CI;33%-88%]

Overdiagnostik i DLCST efter 5 års follow-up

Wille MMW et al. Results of the randomized Danish Lung Cancer Screening Trial with focus on high-risk profiling. Accepted for publication August 2015 in AJRCCM

41

Styrker & svagheder

Ingen screening i kontrolgruppen

Lille kontaminering i kontrolgruppen

For kort follow-up?

Ulige fordeling af storrygere efter randomiseringen?

Participation bias i DLCST?

42

Kardio-vaskulær risiko

CVD diseases and number of risk factors

0

10

20

30

40

50

60

70

80

90

100

20 25 30 35 40 45 50 55 60 65 70 75

Age

%

1 risk factor

0

CVD, diabetes and

treated hypertension

≥ 3 risk factors

2 risk factors

H. Petursson et al. Can individuals with a significant risk for cardiovascular disease be adequately identified by combination of several risk factors? J.Eval.Clin.Pract. 15 (1):103-109, 2009.

43

0

50

100

150

200

250

300

350

400

450

500

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

An

tal i

nci

de

nte

can

cer

pro

stat

a

An

tal p

røve

r

Almen praksis er rekvirent Speciallæge/sygehus er rekvirent Incidente cancer prostatae

PSA-test and Pca in DK

T. O. Mukai, F. Bro, K. V. Pedersen, P. Vedsted. Brug af undersøgelse for prostataspecifikt antigen . Ugeskr.Laeger 172 (9):696-700, 2010.

44



Definition


Hvorfor?

Eksempler

Stadieskifte



Fremtiden?

45

A: Optimalt scenarium

Esserman L., Shieh Y., & Thompson I. Rethinking Screening for Breast Cancer and Prostate Cancer. JAMA: 302 (15):1685-1692, 2009.

46

B: Værste scenarium


47

C: Intermediært scenarium


48

Stadieskifte


49

Breast cancer screening


50

PSA-screening


51



Definition


Hvorfor?

Eksempler

Stadieskifte



Fremtiden?

52


Brodersen J., Schwartz L.M., Woloshin S. Overdiagnosis: How cancer screening can turn indolent pathology into illness. APMIS 122, 2014.53

1995

2000

2005

1980

1985

1990

1995

2000

2005

2010

1985

1990

10000

20000

30000

40000

50000

60000

Cases

per year

Years Years Years

LungWithout screening

Breast (women)Increasing screening

ProstateIncreasing screening

Cancer death and invasive cancer diagnosis with and without screening

Lung, breast and prostate. France 1980-2010

Death Diagnosis

1980

1995

2000

2005

1980

1985

1990

2010

2010

54

1995

2000

2005

1980

1985

1990

1995

2000

2005

2010

1985

1990

10000

20000

30000

40000

50000

60000

Cases

per year

Years Years Years






Death Diagnosis

1980

1995

2000

2005

1980

1985

1990

2010

2010

55

1995

2000

2005

1980

1985

1990

1995

2000

2005

2010

1985

1990

10000

20000

30000

40000

50000

60000

Cases

per year

Years Years Years






Death Diagnosis

1980

1995

2000

2005

1980

1985

1990

2010

2010

56

Hvilke kræftsygdom er dette?

57

58

59

60

Screening med smartphone

61



Definition


Hvorfor?

Eksempler

Stadieskifte



Fremtiden?

62

Skadevirkning af overdiagnostik

Overundersøgelse

Overbehandling

Overforbrug

63

financial strain

hassles/inconveniences

medical costs

opportunity costs

physical harms

psychological harms

societal costs

+ work-related costs

Harris R.P. et al. The

Harms of Screening:

A Proposed Taxonomy

and Application to Lung

Cancer Screening,

JAMA 2014

Skadevirkning af overdiagnostik

64



Definition


Hvorfor?

Eksempler

Stadieskifte



Fremtiden?

65

Fremtiden?

Storm P:

”Det er svært at spå – især om fremtiden”

66

Fremtiden?

Mere screening:

Befolkningsundersøgelser og højrisiko

Quantified Self

Big data & Personalised Medicine

67

Nye screeningsprogrammer?

CT-screening for lungekræft blandt storrygere

Abdominalt aorta anuerisme:

Befolkningsundersøgelse eller storrygere?

68

Quantified Self

69


70

http://www.google.dk/url?sa=i&rct=j&q=&esrc=s&frm=1&source=images&cd=&cad=rja&uact=8&ved=0CAcQjRw&url=http://www.amazon.com/Withings-Wireless-Activity-Tracker-Monitoring/dp/B00CW7KK9K&ei=wR1GVLXlGsO_ywOMzoKQCA&psig=AFQjCNHLswGZyN0qR3JRy5Gn3RoMXzg3JA&ust=1413967603561039

http://www.google.dk/url?sa=i&rct=j&q=&esrc=s&frm=1&source=images&cd=&cad=rja&uact=8&ved=0CAcQjRw&url=http://www.amazon.com/Withings-Wireless-Activity-Tracker-Monitoring/dp/B00CW7KK9K&ei=wR1GVLXlGsO_ywOMzoKQCA&psig=AFQjCNHLswGZyN0qR3JRy5Gn3RoMXzg3JA&ust=1413967603561039

P4 or Personalised medicine

Predictive

Preventive

Personalised

Participatory

71

P4 among healthy

Predictive & Preventive = Screening Incognito

72


73

Survivors stories drive screening towards more overdiagnosis

Popularity paradox

More UsefulScreening

Appears To Be

More IntensiveScreening

MoreOverdiagnosis

More ”Survivor”Stories

74

75

Documents

Overdiagnostik — en uundgåelig konsekvens af medicinsk ...jydskmedicinskselskab.org/wp-content/uploads/2015/... · –en uundgåelig konsekvens af medicinsk screening John Brodersen