Robin W Paton PhD FRCS FRCS Orthopaedic BOTA 2015 What is overdiagnosis First do no harm Disease A disorder of structure or function that is not simply a direct result of physical injury ID: 219230
Download Presentation The PPT/PDF document "Over-diagnosis, over-treatment & ove..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Over-diagnosis, over-treatment & over-screening
Robin W Paton PhD FRCS FRCS (Orthopaedic)BOTA 2015Slide2
What is over-diagnosis?
First do no harm!Disease: A disorder of structure or function that is not simply a direct result of physical injury! Oxford dictionary
Over-diagnosis: Asymptomatic individuals are diagnosed with a disease that will not cause them to experience symptoms or an early deathSlide3
What is over-diagnosis? First do no harm
! Patients deserve accurate diagnosis & appropriate treatment
Over-diagnosis of the wellUnder-treatment of the sick‘Conjoint twins of modern medicine’ Heath 2014Slide4
Drivers of over diagnosis
‘Ability to detect smaller abnormalities axiomatically tend to increase the prevalence of any given disease’ Black 1998Faith in early detection unmodified by risks
Legal incentives that punish under diagnosis but not over diagnosisTechnological changes detecting ever smaller ‘abnormalities’Commercial
and professional interests BMJ 2 June 2012 Slide5
Role of fear in over-diagnosis
BMJ 8 November 2014Patients need clinicians that reassert the border between the well & the sickThe risk of missing a diagnosis frightens Doctors: ‘naming & shaming’/ public pilloryingProtocols, guidelines & ‘screening’ leads to over-diagnosis & over-treatment
Heath 2014Slide6
Self perpetuating loop of diagnostics:Disease mongering & Over-medicalisation
New test Increased accuracyDiagnosis of mild disease/ pseudo disease (spectrum shift)
Treatment results improved outcomes/ ‘success’ Hofmann BMJ 2015 Moynihan et al 2012Slide7
Algorithm based medicine
Dreyfus model of learningNovice: learns basic rules & applies mechanically with no attention to context
(level one): MechanicalExpert: rapid, intuitive reasoning informed on imagination, common sense, selected research & other evidence (level five): Judgement
Expert
Proficient
Competent
Advanced Beginner
Novice
Intuition
Experience
Involved
‘Rules’
Analytic
Detached
Slide8
The role of evidence based medicine (EBM)
Problems: Algorithms have replaced patient centred care Statistical significance = to clinical benefits?
EBM guidelines struggle to contend with multiple morbidities Volume & source of evidence (NICE) Industry & Government interference / bias
Greenhalgh 2014 Slide9
The role of evidence based medicine (EBM) &
Expert Consensus Based medicine (ECBM)Prior to EBM: anecdote, theoretical reasoning from basic science (Dreyfus expert?) Risk of conclusions being accepted without critical review:
Eminence based medicine!EBM: evidence from high quality RCTs & observational studies in combination with clinical expertise‘ECBM’ :
the way forward or fudge? individual evidence & shared decisions i.e. NICE Greenhalgh
2014
Stirling 2014Slide10
Choosing Wisely R , ‘Less is more’ & ‘Too much medicine’
Orthopaedics (USA & Canada): Lack of ambition? Avoid routine post-operative DVT ultrasonography screening in elective TKR & THR
Avoid needle lavage in symptomatic OA of the knee for long term reliefAvoid glucosamine & chondroitin in the treatment of symptomatic OA (knee)Avoid lateral wedge insoles in patients with symptomatic medial compartment OA (knee)Avoid post-operative splinting of the wrist after carpal tunnel release for long term reliefSlide11
‘Less is more’: Canada
Canadian health care system: Dominated by hospitals & specialist doctors Not compensated to recommend alternatives to medical treatment i.e. physiotherapy not covered by medicare
: = operation cheaper?Scandinavian health care system: focus on preventive medicine & primary careSlide12
Can we do more?Surgical procedures
Arthroscopic washout of advanced OA of the knee Arthroscopic debridement of degenerative meniscal tears Arthroscopic shoulder procedures: NICE 2007,
RCT sham surgery, Moseley 2002 BJJ 96-B, 2014, editorial BJSM, 2015Mid shaft clavicle fractures (ORIF): 24.6% repeat surgery Leroux
2014Childrens’ distal radial fractures: > 90% should be treated conservativelyLocking plates in adult distal radial fractures?Slide13
Can we do more?
Abandonment of Orthotics in Paediatric OrthopaedicsInsoles in treatment of asymptomatic flat feetOrthotics for physiological intoeing &
strapping for curly toesBracing for physiological genu varum and valgumSlide14
Screening in OrthopaedicsSlide15
Screening in Orthopaedics
Ideal screening programme:Condition must be importantEffective & accepted treatmentTreatment & diagnostic facilities availableRecognisable latent & early symptomatic stageOpinions on who should be treated agreed
Guaranteed safety, sensitivity & specificityExamination & treatment acceptable Natural history knownTests inexpensive & simpleCost effectiveAudited Wilson & Jungner 1968Slide16
Screening in Orthopaedics
Clinical & ultra-sonographic hip screening in DDHScoliosis screeningMRSA pre-operative screeningVitamin D deficiencySlide17
Screening in Orthopaedics
Screening in DDH: ClinicalSurveillance not screeningStatistics: neonatal screening
Clinical sensitivity: 66%Clinical specificity: 99.77%
Clinical PPV: 27.97%U/S sensitivity: 72.1%U/S specificity: 99.0%U/S PPV: 67.8%
Slide18
Screening in OrthopaedicsScreening in DDH: Clinical
Surveillance not screeningStatistics: GP 6 – 8 week (clinical)
Sensitivity: 13.34%Specificity: >99.9%PPV: 4.65%Slide19
Screening in OrthopaedicsScreening in DDH:
‘At Risk’Surveillance not screeningStatistics: ultra-sound ‘at risk’
Sensitivity: 100%Specificity: 94.2% (93.3%)PPV: 20.5% (7.0%)Statistically high sensitivity & specificity (accuracy) combined with a low prevalence can result in a low PPVSlide20
Screening in Orthopaedics
Scoliosis (National screening Committee 2010): Adolescent idiopathic typeScreening asymptomatic adolescents did not detect important degrees of idiopathic scoliosis earlier (than without screening)Most cases detected through screening will not progress to a clinically important form of scoliosis
Scoliosis requiring aggressive treatment (surgery) likely to be detected without screeningEffects of diagnosing scoliosis: potential harmSlide21
Screening in Orthopaedics
MRSA Pre-operative screening (East Lancashire, UK):8,867 patients elective & emergency T & O surgery (2010)Incidence: 0.47% (42)WHO screening criteria: 4 out of 9
33/42 decolonisation therapy, 18 successfullyNo surgical site infection Barkatali et al 2013 Slide22
Screening in Orthopaedics
Vitamin D deficiency (East Lancashire, UK):88% below normal Vitamin D threshold (winter/spring levels statistically lower) Low levels Vitamin D rarely associated with biochemical or radiological rickets (2.61
% with radiological rickets) Foley et al 2012Vitamin D thresholds and radiological rickets Vitamin D < 20mcg/l Vitamin D < 10mcg/lSensitivity (%)
100 66.7Specificity (%) 34.8 76.8PPV (%) 4.0 7.2 Slide23
Under-diagnosis in Orthopaedics
SUFESLESlide24
Summary
Avoid over-diagnosis & over treatment (Dreyfus 5, higher order thinking)Do not classify ‘healthy’ individuals as ‘sick’!
(‘First do no harm’)‘Control’ screening, surveillance is not screeningProduce high quality evidence or be dictated to!Slide25
Thank youSlide26
Can we do more?
Surgical proceduresVertebroplasty vs. sham surgery for pain in vertebral compression fractures Buchbinder
2009, Kallmes 2009Early anterior cruciate reconstruction in young individuals
structured training programme vs. early / delayed optional reconstruction 31% of delayed group ACL reconstruction, 61% no surgery no differences in function at 2 years Frobell
2010
Slide27
Role of classifications in Orthopaedics:Validation?
Useful: Salter- Harris (Childrens’ fractures) Vancouver (periprosthetic #s) Gartland
(supracondylar Humerus) Gustilo & Anderson (tibial fractures)Over-complicated:
Frykman (wrist fractures) Acromio-clavicular joint (Type I to VI) A/O fracture classification
Questionable:
Mangled extremity severity score (MESS)
Milch
(Lateral condylar mass #) Slide28
The role of evidence based medicine (EBM)
Achievements: Cochrane collaboration Higher standards of primary & secondary research
National & International clinical practice guidelines Teaching & development of critical appraisal
Building of knowledge baseSlide29