SlideShare a Scribd company logo

Technology Assessment/Outcome & Cost-Effectiveness Analysis 2016

Download as PPTX, PDF
E
evadew1

This document discusses technology assessment, outcomes research, and economic analyses in healthcare. It provides background on rising healthcare costs in the US without clear improvements in health outcomes compared to other countries. The rationale for assessing new technologies and their impact is described. Key aspects of technology assessment are outlined, including technical efficacy, diagnostic accuracy, diagnostic impact, therapeutic impact, patient outcomes, and societal outcomes. Challenges with randomized controlled trials in assessing technologies are reviewed. The National Lung Screening Trial is presented as an example. Finally, computed tomography for appendicitis is analyzed as a hypothetical example of how modeling could be used to assess a technology when a randomized trial may not be feasible.

Presentations & Public Speaking
1 of 81
Downloaded 14 times
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
Technology Assessment, Outcomes Research and Economic Analyses Deborah A. Baumgarten, M.D., M.P.H. Emory University Hospital Department of Radiology
Rationale • Cost of medical care increasing faster than cost of living • 5.6% of GDP in 1965; 17.6% 2012; 19.8% 2020 • $27.5 B 1960; $2.71 T 2011; $4.64 T 2020 • $8649 per person in 2011; $13,709 2020 • US spends more than 33 others in Organization for Economic Co-operation and Development • But fewer MDs, hospital beds per capita
OECD spending on health care OECD Health Data 2012
Health care expenditure as % GDP OECD Health Data 2012
0 2 4 6 8 10 12 14 16 18 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010 2013 US (17.1%) FR (11.6%) SWE (11.5%) GER (11.2%) NETH (11.1%) SWIZ (11.1%) DEN (11.1%) NZ (11.0%) CAN (10.7%) JAP (10.2%) NOR (9.4%) AUS (9.4%)* UK (8.8%) Health Care Spending as Percentage of GDP, 1980–2013 Percent OECD Health Data 2015
Rationale • Workers now paying at least $1,094 more in premiums annually for coverage than in 2000 • Still waiting to see what ACA does… • Despite expenditures, can we document significant improvement in health status? • Life expectancy • Perinatal mortality
Life expectancy
Infant mortality rate
Our poor health • Life expectancy 78.8 vs. 81.2 (2013) • Infant mortality 6.1/1000 vs. 3.5 (2011) • 68% adults over 65 had at least 2 chronic conditions (33% UK, 56% Canada; 2014) • Obesity rate 35.3% (3.7% Japan, 30.6% NZ; 2014) • Smokers 13.7 (12.8 Australia, 24.1 France; 2013)
US does more of… United States Rank compared with OECD countries OECD median MRI units 35.5 per 1,000,000 2nd (Japan 1st) 11.4 per 1,000,000 MRI exams 106.9 per 1000 50.6 per 1000 CT units 43.5 per 1,000,000 3rd (Japan 1st, Australia 2nd) 17.6 per 1,000,000 CT exams 240 per 1000 136 per 1000 Tonsillectomy 254.4 per 100,000 1st 130 per 100,000 Coronary bypass 79.0 per 100,000 3rd 47.3 per 100,000 Knee replacement 226 per 100,000 1st 121.6 per 100,000 Caesarian section 32.9 per 1000 live births 6th 26.1 per 1000 live births OECD Health Data 2012 and 2015
Why? • Current payment structure means MDs get paid more to do more interventions • ACA has not had effect yet… • Fear of litigation • Patients ask for more—and usually do not pay out of pocket (insurance covers)
Why? • MDs in US paid more but seen less • 4 visits/year (OECD average 6.5 visits/year) • US uses more pharmaceuticals • 2.2 drugs per adult (1.2 Netherlands) • Cost more (US=100; Australia=49; Germany=95) • Procedures cost more • Bypass 75,345 vs. 42,130 in Australia • Appendectomy 13, 910 vs 4,995 Netherlands
Technology Assessment
Bit of background • Technology assessment hierarchy • Based on efficacy (benefit under ideal circumstances) • Differs from effectiveness (benefit in real world) • Differs from efficiency (sufficient value? adds economic component) Fryback D, Thornbury J. The efficacy of diagnostic imaging. Med Dec Making 1991;11:88-94
Technology assessment hierarchy • Technical efficacy • Diagnostic accuracy • Diagnostic impact • Therapeutic impact • Patient outcomes • Societal outcomes Fryback D, Thornbury J. The efficacy of diagnostic imaging. Med Dec Making 1991;11:88-94
Technical efficacy • Ability of technology to produce image • Measured by image characteristics • Signal to noise, resolution of line pairs, etc. • Developmental: prototype and phantom, safety • Exploratory: possible use, wide range of patients, situations, conditions
Technical efficacy • Ideas of what falls here?
Diagnostic accuracy • Distinguish between disease, non-disease, or normal, abnormal • Establish diagnosis if symptoms • Screen in asymptomatic patient • Quantify severity or extent of disease • Prognostic information (staging) • Monitor therapy • Need gold standard; measure of truth
Diagnostic accuracy • Sensitivity, specificity, positive and negative predictive values • 2 x 2 table to display comparison to gold standard or reference • Area under ROC curve • Most radiology research falls into this category
Diagnostic accuracy • Ideas?
Diagnostic impact • Which imaging tests have greatest impact on clinical decisions? • Before test need to know • Differential diagnosis • Degree of certainty of diagnoses • Probability of each consideration • Reassess after test performed • Best done prospectively
Diagnostic impact • Ideas?
Therapeutic impact • Which imaging tests have greatest impact on therapeutic decisions? • Percent procedures avoided due to imaging • Percent imaging changed plan • Retrospectively inferred from records • Prospectively assessed using clinicians’ stated plans
Therapeutic impact • Ideas?
Patient outcomes • Traditionally, morbidity or mortality • Change in quality adjusted life years (QALY) • Can be difficult to prove link as imaging study several steps removed from patient outcome
Societal outcomes • Cost analyses from societal perspective • Usually reported as cost per QALY • Must be below some threshold to be accepted (? $50,000/QALY) • Example, head CT to screen for brain mets in lung cancer patients or only if symptomatic? • $70,000/QALY deemed too high to adopt
Outcomes research
Outcomes research—goals • Use imaging wisely • Avoid over-diagnosis and over-treatment • Risks of treatment in early or mild disease may outweigh benefits • Data that will permit maintenance of quality at reduced cost • Increasingly important • Radiology needs to prove worth
Outcomes research—getting started • Choose technology to be evaluated • Define specific clinical indication • Define target population • Choose alternative techniques for comparison (may include no imaging) • Define outcomes affected by technology
Randomized controlled trial • Gold standard of outcomes research • Patients assigned to different groups based on imaging (or no imaging) to be performed • Measure appropriate outcomes • Confounders and biases accounted for in design
RCT—limitations • Cost • Size of population • Power calculation • Estimate expected differences • Time to complete • Technology obsolete by time outcomes assessed? • Technology as moving target • Results may only apply to narrow spectrum of patients; issue of generalizeability
RCT—unnecessary when? • Anecdotal evidence of benefit dramatic; observational studies suffice • Example: head CT in trauma • New technology: • Has same or fewer side effects • Has same or better accuracy • Is less expensive
RCT—necessary when? • Really don’t know answer • New technology: • More expensive • May result in more morbidity • Very controversial
ACRIN—NLST • National Lung cancer Screening Trial • 25,000 patients randomized; low dose CT vs. CXR • Baseline, years 1 and 2 • Q6 month health status and medical intervention questionnaire • Very detailed description of what constitutes positive screen and what to do • Outcomes • Primary: Lung cancer specific mortality • Secondary: huge list
ACRIN—NLST • Low dose CT more sensitive than radiography • Two annual incidence screenings with low- dose CT resulted in • Decrease number of advanced-stage cancers • Increase number of early-stage lung cancers NEJM 2013;369:920-921
Example to work through • CT for appendicitis: improve outcomes? • Technology:
Example to work through • CT for appendicitis: improve outcomes? • Technology: MDCT
Example to work through • CT for appendicitis: improve outcomes? • Technology: MDCT (?oral, ?IV, ?rectal) • Indication:
Example to work through • CT for appendicitis: improve outcomes? • Technology: MDCT (?oral, ?IV, ?rectal) • Indication: RLQ pain, fever, etc. • Population:
Example to work through • CT for appendicitis: improve outcomes? • Technology: MDCT (?oral, ?IV, ?rectal) • Indication: RLQ pain, fever, etc. • Population: patients presenting to ER • Alternative:
Example to work through • CT for appendicitis: improve outcomes? • Technology: MDCT (?oral, ?IV, ?rectal) • Indication: RLQ pain, fever, etc. • Population: patients presenting to ER • Alternative: best clinical judgment plus lab • Outcomes:
Example to work through • CT for appendicitis: improve outcomes? • Technology: MDCT (?oral, ?IV, ?rectal) • Indication: RLQ pain, fever, etc. • Population: patients presenting to ER • Alternative: best clinical judgment plus lab • Outcomes: morbidity, mortality, perforation rate, length of hospital stay, surgery rate, % normal appendices removed, cost to hospital, etc.
Example to work through • CT for appendicitis: improve outcomes? • Technology: MDCT (?oral, ?IV, ?rectal) • Indication: RLQ pain, fever, etc. • Population: patients presenting to ER • Alternative: best clinical judgment plus lab • Outcomes: morbidity, mortality, perforation rate, length of hospital stay, surgery rate, % normal appendices removed, cost to hospital, etc. • Is this feasible?
What if RCT not feasible? • Modeling: Use of decision analysis software • Need information from other studies • Accuracy, sensitivity, specificity • Course of disease, natural history • Effectiveness of therapy • Patient utilities and costs • Meta-analyses for combining results of studies
Modeling • Assumptions about test parameters (sensitivity, specificity) based on literature • Then vary assumptions • Assumptions about frequency of outcomes based on personal experience or literature (mortality and morbidity rates, etc.) • Allows variation of these
Modeling—literature example Cervical spine screening with CT vs. plain films Blackmore CC et al. Radiology 1999;212:117-125
CT in appendicitis example • Retrospective study of 651 patients treated before use of CT for appendicitis (1992-1995) • 52% went to surgery; 24% for observation then surgery; 24% observation to discharge • Initial surgical group, 81% appendicitis • Observation to surgery group, 80% appendicitis • Observation to D/C group, assumed no appendicitis (no returns to their hospital) Rhea JT et al. AJR 1997;169:113-118
CT in appendicitis example • Apply sensitivity and specificity of CT from literature to 100 theoretical patients • Use morbidity, mortality, perforation, correct diagnosis rates from study of 651 patients • Model impact of CT on outcomes • Decrease in nontherapeutic surgery (13 patients), observation days (1 day less on average), perforation rate, etc. • Alternative diagnoses also confirmed Rhea JT et al. AJR 1997;169:113-118
Intermediate outcomes • Choose levels lower in hierarchy • Diagnostic impact • Change in diagnostic possibilities and confidence • Before and after test ask referring MD: • Differential diagnosis • % certainty for each diagnosis • Degree of satisfaction with test
Intermediate outcomes • Seattle Low Back Pain Imaging Project (SLIP) • RCT of plain films vs. MR as initial imaging • Diagnostic impact • 15% new diagnosis with plain film • 30% for MRI (spinal stenosis, HNP, etc) • Change in probability for most likely diagnosis for HNP (p<.002) Jarvik JG et al. Radiology 1997;204:447-454
Intermediate outcomes • Therapeutic impact • Before test, ask MD for treatment plan • Observe if plan changes after test • If temporally related, infer causal relationship between test and plan change • SLIP: 38% avoided additional test if plain film first, 64% if MRI first (p<.001)
Intermediate outcomes • Advantages • Cheaper and easier than full RCT (RCT may show intermediate outcomes if assessed as in SLIP) • No one denied potentially good test • Works well for “add-on tests” • Tests that would not necessarily replace existing work-up
Intermediate outcomes • Disadvantages • Stated plans do not necessarily equal clinical action • If treatment invasive, difficult, expensive or risky, easier to hypothetically choose • RCT still better for “replacement test”
Patient outcomes • Reduced mortality • Reduced morbidity • Decrease complications from other tests • Avoid invasive procedures • Faster return to normal (or pretest) state
Patient outcomes • Quality of life: ability to undertake rewarding and enjoyable activities • How person feels, functions • Quality adjusted life years (QALYs) • Patient satisfaction, reassurance • Physiologic marker improvement • Improved function
Functional status • General and specific measures • SF-36 (Short Form 36) • 36 questions, 8 domains • General health, pain, physical functioning, social functioning, mental health, etc. • Roland score (back pain specific index) • Sciatic frequency index
SLIP follow-up • RCT ultimately enrolled 380 patients • Roland score at 12 months did not differ between 2 groups (8.75 plain film vs. 9.34 MRI, p = .53) • p >.75 for all domains of SF-36 • Patient reassurance measure • 1 month p=.09 • 3 month p=.08 • 12 month p=.002 (58% plain film vs. 74% MRI) • Increased reassurance with MRI over time • 10 in MRI vs. 4 in plain film had surgery (p =.09) Jarvik JG et al. JAMA 2003;289:2810-2818
Outcomes research—limitations • Tenuous link between test and patient outcome • Test usually steps removed from outcome • Best situation: proven link between therapy and outcome  link diagnosis to therapy to outcome • Example: detection of cerebral aneurysms with MRA  treatment decreases morbidity and mortality  link MRA to reduced morbidity/mortality
Outcomes research—limitations • If no proven impact on outcome, may be: • Diagnostic test (inaccurate, insensitive) • Referring clinician (inappropriate use of info; info does not reach) • Therapy (ineffective or unavailable) • Patient (lack of compliance) • System (lack of patient access)
Alternative outcome • Therapeutic value of diagnostic test • Diagnostic test can have direct impact on quality of life • Reassurance important to patients and referring clinicians • If consider only positive tests fail to consider that negative test has value
Alternative outcome • Example: nonspecific chest pain • 176 patients randomized to receive no studies or EKG and CPK level • Test group had less short term disability (20% versus 46%; P=.001) • Use of diagnostic test independent predictor of recovery in logistic regression • Test group felt care “better than usual” Sox HC, et al. Ann Intern Med 1981;95:680-5
Economic Analyses
Economic evaluations • Must establish viewpoint of analysis • Society, insurer, provider, patient • All have different inputs and outcomes; different costs to consider • Example: patient may value increased quality of life at any cost while HMO may value decreasing overhead costs
Costs • Value of resources • Charges as estimate • Medicare/Medicaid reimbursement as estimate • Micro-costing techniques—all resources identified and quantified • Consider cost of missed abnormalities and superfluous work-up of false positives
Costs—examples • Medical • Diagnostic test and interpretation • Nursing care • Nonmedical • Time (travel, tests and treatment) • Travel • Support with daily activities • Lost productivity
Types of economic analyses • Cost identification or cost minimization • Compares alternative health care strategies assumed equally effective
Cost identification—example • CT vs. US plus plain film after ESWL • 25 patients underwent all 3 • Assumed equal accuracy • Exams timed; direct technical costs calculated using procedural-based accounting system • CT 15.3 minutes; US + plain film 37.2 minutes • CT $36.86; US + plain film $57.60 • Sensitivity analysis on time; within reasonable range CT always cost less Remer EM et al. Radiology 1997;204:33-37
Types of economic analyses • Cost benefit • Costs and benefits assigned dollar values • Comparison of different technologies • Limited use in medicine due to difficulty assigning monetary values to health states • Cost effectiveness leaves outcomes in natural, objective units • # lives saved, # cancers detected, etc.
CEA—example • What gain from sixth stool guaiac? • Progression from fifth to sixth stool guaiac costs $47 million/cancer detected • Can use intermediate outcomes (length of stay, readmission rates, number of unnecessary surgeries prevented) Neuhauser D, Lewicki AM. N Engl J Med 1975; 293
Cost utility analysis • Similar to CEA but uses subjective outcomes that explicitly incorporate patient preferences • QALY: basic outcome measure • Year of life with significant morbidity less desirable than year with excellent health • Full health equals 1 • Death equals 0 • Morbidity somewhere between 0 and 1 • Preferred method for economic evaluations
Utility values • Side effects from interferon—.93 • Mild angina—.90 • Prophylactic mastectomy, oophorectomy—.86 • Moderate angina—.70 • Hospitalization for tuberculosis—.60 • Severe angina—.50 • Moderate to severe stroke—.39 • Recurrent stroke—.12
Components of economic analysis • Most with modeling software • Reference case (defines typical patient) • Strategies (set of alternative care pathways) • Time horizon (period to consider risks and benefits) • Perspective (viewpoint of decision maker) • Effectiveness measure (outcome) • Decision tree allows varying components
Decision tree Singer ME and Applegate KE. Radiology 2001;219:611-620
Components of economic analysis • Probabilities (all events in model have one) • Estimated from historical data; literature, pilot study or expert opinion • Utilities (assign utility to every health condition in model) • Costs (identify all relevant) • Assumptions (due to limitations imposed by available data; must explicitly state)
Economic analysis—example • Whole body CT screening • Base-case analysis: from cohort of 500,000 50 year old men • Sex, age varied in analysis • Benefits of screening assumed due to earlier detection of disease, improvement in survival • 8 conditions included in model: ovarian, pancreatic, lung, liver, kidney and colon cancer; aortic aneurysm, coronary artery disease • Costs of screening, follow-up tests, patient care Beinfeld, M. T. et al. Radiology 2005;234:415-422
Beinfeld, M. T. et al. Radiology 2005;234:415-422 Schematic of decision-analytic model with life expectancy and cost outcomes used in the cost effectiveness analysis
Economic analysis—example • Compared to routine care • 6 days of life gained at $2513/patient • Incremental cost: $151,000/life-year gained • 90.8% had finding; only 2% had disease • 32.3% of total cost = work-up of FPs • With favorable assumptions, one time screening not cost effective; would add burden to healthcare system Beinfeld, M. T. et al. Radiology 2005;234:415-422
Economic analysis—limitations • Not all agree with strategies chosen • Was comparison to standard of care? • Were all reasonable options considered? • Was reference case typical of patient population? • Fears of bias if study funded by drug company or equipment manufacturer • If too early, data scant; if late, decisions made and model obsolete • Noncontrast CT for renal colic: accepted standard without study of cost
Conclusions • Need to be familiar with concepts • Way to inform health care policy makers, guide research • Goal to improve quality in rational manner, decrease waste and unnecessary expenditure • Enhance Radiology community’s ability to participate in health care decision-making locally and nationally • Preserve our specialty!
Thanks!
US does more of… OECD Health Data 2015

More Related Content

PPTX
Technology Assessment, Outcomes Research and Economic Analyses
evadew1
 
PPTX
Predictors of MDT review and the impact on lung cancer survival for HNELHD re...
Cancer Institute NSW
 
PPT
Introducing VESPIR: a new open-source software to investigate CT ventilation ...
Cancer Institute NSW
 
PPTX
Defining and assessing a delineation uncertainty margin for modern radiotherapy
Cancer Institute NSW
 
PPTX
One example of how Clinical Cancer Registry level data can review practice va...
Cancer Institute NSW
 
PPTX
A distributed data mining network infrastructure for Australian radiotherapy ...
Cancer Institute NSW
 
PPTX
Cardoso C - AIMRADIAL 2014 - Radial operators and femoral
International Chair on Interventional Cardiology and Transradial Approach
 
PPT
Hydrogel use in prostate cancer radiation therapy
Matthew Katz
 
Technology Assessment, Outcomes Research and Economic Analyses
evadew1
 
Predictors of MDT review and the impact on lung cancer survival for HNELHD re...
Cancer Institute NSW
 
Introducing VESPIR: a new open-source software to investigate CT ventilation ...
Cancer Institute NSW
 
Defining and assessing a delineation uncertainty margin for modern radiotherapy
Cancer Institute NSW
 
One example of how Clinical Cancer Registry level data can review practice va...
Cancer Institute NSW
 
A distributed data mining network infrastructure for Australian radiotherapy ...
Cancer Institute NSW
 
Cardoso C - AIMRADIAL 2014 - Radial operators and femoral
International Chair on Interventional Cardiology and Transradial Approach
 
Hydrogel use in prostate cancer radiation therapy
Matthew Katz
 

What's hot (20)

PPT
Limited three slice head CT protocol for monitoring VP shunts
Yasser Asiri
 
PPTX
Proposal presentation
Ashenafi Kaltro
 
PPT
Stereotactic radiation
Matthew Katz
 
PPT
Low Dose Radiation Therapy (LDRT) for COVID-19 Pneumonia
Matthew Katz
 
PDF
ASTUTE: Acute Stroke Telemedicine
NHS Improving Quality
 
PDF
Best Practices for a Data-driven Approach to Test Utilization
Viewics
 
PDF
Revolutionizing Renal Care With Predictive Analytics for CKD
Viewics
 
PPTX
Journal club-Determination of surgical priorities in appendicitis
Youttam Laudari
 
PPTX
Dr. Davy Cheng
ichil
 
PDF
R0907 pa dosect_pr
Guillermo Alberto López
 
PPTX
Implementing American Heart Association Practice Standards for Inpatient ECG ...
Allina Health
 
PPT
POC Breast 1 | 2007 - Adjuvant Chemotherapy
rtp
 
PPTX
FAST Forward Trial breast cancer
Kanhu Charan
 
PDF
2018 preceptorship-lung-stereotactic-ablative-rt-early-stages-nsclc-suresh-senan
Rubén Quenhua
 
PPTX
Accuracy of nexus ii
abdullah alzahrani
 
PDF
Patient Blood Management: Impact of Quality Data on Patient Outcomes
Viewics
 
DOCX
New tif results
Ben Gurion University of the Negev
 
PDF
Tackling the U.S. Healthcare System’s Infectious Disease Management Problem
Viewics
 
PPTX
ICN Victoria: Gantner on "Translating Research into Practice"
Intensive Care Network Victoria
 
PPTX
MedicalResearch.com - Medical Research Week in Review
Marie Benz MD FAAD
 
Limited three slice head CT protocol for monitoring VP shunts
Yasser Asiri
 
Proposal presentation
Ashenafi Kaltro
 
Stereotactic radiation
Matthew Katz
 
Low Dose Radiation Therapy (LDRT) for COVID-19 Pneumonia
Matthew Katz
 
ASTUTE: Acute Stroke Telemedicine
NHS Improving Quality
 
Best Practices for a Data-driven Approach to Test Utilization
Viewics
 
Revolutionizing Renal Care With Predictive Analytics for CKD
Viewics
 
Journal club-Determination of surgical priorities in appendicitis
Youttam Laudari
 
Dr. Davy Cheng
ichil
 
R0907 pa dosect_pr
Guillermo Alberto López
 
Implementing American Heart Association Practice Standards for Inpatient ECG ...
Allina Health
 
POC Breast 1 | 2007 - Adjuvant Chemotherapy
rtp
 
FAST Forward Trial breast cancer
Kanhu Charan
 
2018 preceptorship-lung-stereotactic-ablative-rt-early-stages-nsclc-suresh-senan
Rubén Quenhua
 
Accuracy of nexus ii
abdullah alzahrani
 
Patient Blood Management: Impact of Quality Data on Patient Outcomes
Viewics
 
New tif results
Ben Gurion University of the Negev
 
Tackling the U.S. Healthcare System’s Infectious Disease Management Problem
Viewics
 
ICN Victoria: Gantner on "Translating Research into Practice"
Intensive Care Network Victoria
 
MedicalResearch.com - Medical Research Week in Review
Marie Benz MD FAAD
 
Ad

Viewers also liked (9)

PPT
Assessment & Feedback Literature Review
Morse Project
 
PPTX
Probit and logit model
Jithmi Roddrigo
 
PDF
Strategic Environmental Assessment at the Policy Level: Recent Progress, Curr...
zubeditufail
 
PPTX
Environmental impact assessment methodology
Justin Joy
 
PPTX
Environmental Impact Assessment and Environmental Audit- Unit III
GAURAV. H .TANDON
 
ZIP
Environmental Impact Assessment
Nigel Gardner
 
PPT
Environmental impact assessment in nepal
Bhim Upadhyaya
 
PPT
Summarization of Environmental Impact Assessment Methodology by Dr. I.M. Mis...
Arvind Kumar
 
PPTX
L 17 Assessment in a Constructivist ,Technology-Supported Learning Ed Tech 1
lovely perez
 
Assessment & Feedback Literature Review
Morse Project
 
Probit and logit model
Jithmi Roddrigo
 
Strategic Environmental Assessment at the Policy Level: Recent Progress, Curr...
zubeditufail
 
Environmental impact assessment methodology
Justin Joy
 
Environmental Impact Assessment and Environmental Audit- Unit III
GAURAV. H .TANDON
 
Environmental Impact Assessment
Nigel Gardner
 
Environmental impact assessment in nepal
Bhim Upadhyaya
 
Summarization of Environmental Impact Assessment Methodology by Dr. I.M. Mis...
Arvind Kumar
 
L 17 Assessment in a Constructivist ,Technology-Supported Learning Ed Tech 1
lovely perez
 
Ad

Similar to Technology Assessment/Outcome & Cost-Effectiveness Analysis 2016 (20)

PPT
Patient Selection for Primary Prevention Implantable Cardioverter Defibrillat...
HMO Research Network
 
PPT
Patient Selection for Primary Prevention Implantable Cardioverter Defibrillat...
HMO Research Network
 
PPTX
13-Teleorthopaedic explaining benefits challenges n limitation.pptx
LailaRahmah7
 
PPTX
Outcome Measures in Cancer: Do disease specific instruments offer greater sen...
Office of Health Economics
 
PPTX
Driving progress in healthcare through NHS research
Health and Care Innovation Expo
 
PPTX
MLS13 QI Workshop
Steven Kinnear
 
PPTX
Cost Effectiveness in Radiology - By Jeffrey Shyu
Brigham and Women's Hospital
 
PPTX
Quality Measurement in Cardiac Surgery
Nora Albogami
 
PPT
Cost utility analysis of interventions to return employees to work following ...
ScHARR HEDS
 
PPTX
NVTC Capital Health Tech Summit: Dr. Shannon Keynote
Alexa Magdalenski
 
PPTX
Implementation science tailored to precision prevention
Graham Colditz
 
PPTX
Practical Points in Emergency CT for Emergency Physicians
Rathachai Kaewlai
 
PDF
Prof. Todor (Ted) A. Popov - 6th Clinical Research Conference
Starttech Ventures
 
PDF
Hugh Gravelle: The impact of care quality on patient choice
Nuffield Trust
 
PPTX
Theranosis_LS_StartHealth_Pitch_Deck.pptx
Ashish86763
 
PPTX
Radiology Cost Effectiveness - By Jeffrey Shyu
Brigham and Women's Hospital
 
PPTX
Trial of Decompressive Craniectomy for Traumatic Intracranial Hypertension
MQ_Library
 
PPTX
Cardiac rehab, telehealth, the evidence for alternatives for ACOs
Maxpowerjr
 
PPT
Professor Richard Beale @ MRF's Meningitis & Septicaemia in Children & Adults...
Meningitis Research Foundation
 
PPTX
Student By ASTRO Radiation Oncology Education.pptx
ehabshaltout2
 
Patient Selection for Primary Prevention Implantable Cardioverter Defibrillat...
HMO Research Network
 
Patient Selection for Primary Prevention Implantable Cardioverter Defibrillat...
HMO Research Network
 
13-Teleorthopaedic explaining benefits challenges n limitation.pptx
LailaRahmah7
 
Outcome Measures in Cancer: Do disease specific instruments offer greater sen...
Office of Health Economics
 
Driving progress in healthcare through NHS research
Health and Care Innovation Expo
 
MLS13 QI Workshop
Steven Kinnear
 
Cost Effectiveness in Radiology - By Jeffrey Shyu
Brigham and Women's Hospital
 
Quality Measurement in Cardiac Surgery
Nora Albogami
 
Cost utility analysis of interventions to return employees to work following ...
ScHARR HEDS
 
NVTC Capital Health Tech Summit: Dr. Shannon Keynote
Alexa Magdalenski
 
Implementation science tailored to precision prevention
Graham Colditz
 
Practical Points in Emergency CT for Emergency Physicians
Rathachai Kaewlai
 
Prof. Todor (Ted) A. Popov - 6th Clinical Research Conference
Starttech Ventures
 
Hugh Gravelle: The impact of care quality on patient choice
Nuffield Trust
 
Theranosis_LS_StartHealth_Pitch_Deck.pptx
Ashish86763
 
Radiology Cost Effectiveness - By Jeffrey Shyu
Brigham and Women's Hospital
 
Trial of Decompressive Craniectomy for Traumatic Intracranial Hypertension
MQ_Library
 
Cardiac rehab, telehealth, the evidence for alternatives for ACOs
Maxpowerjr
 
Professor Richard Beale @ MRF's Meningitis & Septicaemia in Children & Adults...
Meningitis Research Foundation
 
Student By ASTRO Radiation Oncology Education.pptx
ehabshaltout2
 

More from evadew1 (19)

PPT
The New "3R's": Radiology Resident Research
evadew1
 
PPTX
How to Write a Career Development Grant
evadew1
 
PPT
Introduction/Housekeeping
evadew1
 
PPT
Quality and Safety Research
evadew1
 
PPT
Research Ethics Garcia
evadew1
 
PPT
Motivation for Research in Radiology
evadew1
 
PPTX
Research During Residency
evadew1
 
PPT
Introduction/Housekeeping 2016
evadew1
 
PDF
Principles of Diagnostic Testing and ROC 2016
evadew1
 
PPTX
Motivation for Research In Radiology 2016
evadew1
 
PPT
Introduction/Housekeeping 2016
evadew1
 
PPTX
Study Design Tridandapani 2016
evadew1
 
PPT
Garcia Ethics 2016
evadew1
 
PPT
Radiology Residents Involvement in Research 2016
evadew1
 
PPT
Grant Writing 2016
evadew1
 
PPTX
IRB Approval Process 2016
evadew1
 
PPT
Abstract Writing & Oral Presentations 2016
evadew1
 
PPT
Clinical Trial Statstics 2016
evadew1
 
PPT
Decision Trees & Bayes Theorem 2016
evadew1
 
The New "3R's": Radiology Resident Research
evadew1
 
How to Write a Career Development Grant
evadew1
 
Introduction/Housekeeping
evadew1
 
Quality and Safety Research
evadew1
 
Research Ethics Garcia
evadew1
 
Motivation for Research in Radiology
evadew1
 
Research During Residency
evadew1
 
Introduction/Housekeeping 2016
evadew1
 
Principles of Diagnostic Testing and ROC 2016
evadew1
 
Motivation for Research In Radiology 2016
evadew1
 
Introduction/Housekeeping 2016
evadew1
 
Study Design Tridandapani 2016
evadew1
 
Garcia Ethics 2016
evadew1
 
Radiology Residents Involvement in Research 2016
evadew1
 
Grant Writing 2016
evadew1
 
IRB Approval Process 2016
evadew1
 
Abstract Writing & Oral Presentations 2016
evadew1
 
Clinical Trial Statstics 2016
evadew1
 
Decision Trees & Bayes Theorem 2016
evadew1
 

Recently uploaded (20)

PPTX
Primary and secondary sources, and history
valenciamarcchristia
 
PPTX
AcademyNaturalLanguageProcessing-EN-ILT-M02-Introduction.pptx
nguyenhung549091
 
PPTX
Human Mind & its character Characteristics
peterjoekari
 
PPTX
Role and Responsibilities of Bangladesh Coast Guard Base, Mongla Challenges
Tasnim Alam Safin
 
PPTX
An Unlikely Response 08 10 2025.pptx
FamilyWorshipCenterD
 
PDF
oil_refinery_presentation_v1 sllfmfls.pdf
TusharPrajapati65
 
PDF
Why Top Brands Trust Enuncia Global for Language Solutions.pdf
enuncia-global
 
PPTX
Relationship Management Presentation In Banking.pptx
ssuser1d0512
 
PPTX
Intro to ISO 9001 2015.pptx wareness raising
nadianisar5
 
PPTX
INTERNATIONAL LABOUR ORAGNISATION PPT ON SOCIAL SCIENCE
AbinayaaAnbazhakan
 
PPTX
Effective_Handling_Information_Presentation.pptx
HarisKaimKhani
 
PPTX
Introduction to Effective Communication.pptx
AnithaT18
 
PDF
Nykaa-Strategy-Case-Fixing-Retention-UX-and-D2C-Engagement (1).pdf
VaibhavBansal926958
 
PPTX
nose tajweed for the arabic alphabets for the responsive
Hijabi1
 
PPTX
fundraisepro pitch deck elegant and modern
aishanmims
 
PPTX
Understanding-Communication-Berlos-S-M-C-R-Model.pptx
JommelVienne
 
PPTX
Tablets And Capsule Preformulation Of Paracetamol
sgsayan31
 
PPTX
Learning-Plan-5-Policies-and-Practices.pptx
judekimwellnombre15
 
PPTX
Hydrogel Based delivery Cancer Treatment
sgsayan31
 
PPTX
The Effect of Human Resource Management Practice on Organizational Performanc...
wemasterconstruction
 
Primary and secondary sources, and history
valenciamarcchristia
 
AcademyNaturalLanguageProcessing-EN-ILT-M02-Introduction.pptx
nguyenhung549091
 
Human Mind & its character Characteristics
peterjoekari
 
Role and Responsibilities of Bangladesh Coast Guard Base, Mongla Challenges
Tasnim Alam Safin
 
An Unlikely Response 08 10 2025.pptx
FamilyWorshipCenterD
 
oil_refinery_presentation_v1 sllfmfls.pdf
TusharPrajapati65
 
Why Top Brands Trust Enuncia Global for Language Solutions.pdf
enuncia-global
 
Relationship Management Presentation In Banking.pptx
ssuser1d0512
 
Intro to ISO 9001 2015.pptx wareness raising
nadianisar5
 
INTERNATIONAL LABOUR ORAGNISATION PPT ON SOCIAL SCIENCE
AbinayaaAnbazhakan
 
Effective_Handling_Information_Presentation.pptx
HarisKaimKhani
 
Introduction to Effective Communication.pptx
AnithaT18
 
Nykaa-Strategy-Case-Fixing-Retention-UX-and-D2C-Engagement (1).pdf
VaibhavBansal926958
 
nose tajweed for the arabic alphabets for the responsive
Hijabi1
 
fundraisepro pitch deck elegant and modern
aishanmims
 
Understanding-Communication-Berlos-S-M-C-R-Model.pptx
JommelVienne
 
Tablets And Capsule Preformulation Of Paracetamol
sgsayan31
 
Learning-Plan-5-Policies-and-Practices.pptx
judekimwellnombre15
 
Hydrogel Based delivery Cancer Treatment
sgsayan31
 
The Effect of Human Resource Management Practice on Organizational Performanc...
wemasterconstruction
 

Technology Assessment/Outcome & Cost-Effectiveness Analysis 2016

  • 1. Technology Assessment, Outcomes Research and Economic Analyses Deborah A. Baumgarten, M.D., M.P.H. Emory University Hospital Department of Radiology
  • 2. Rationale • Cost of medical care increasing faster than cost of living • 5.6% of GDP in 1965; 17.6% 2012; 19.8% 2020 • $27.5 B 1960; $2.71 T 2011; $4.64 T 2020 • $8649 per person in 2011; $13,709 2020 • US spends more than 33 others in Organization for Economic Co-operation and Development • But fewer MDs, hospital beds per capita
  • 3. OECD spending on health care OECD Health Data 2012
  • 4. Health care expenditure as % GDP OECD Health Data 2012
  • 5. 0 2 4 6 8 10 12 14 16 18 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010 2013 US (17.1%) FR (11.6%) SWE (11.5%) GER (11.2%) NETH (11.1%) SWIZ (11.1%) DEN (11.1%) NZ (11.0%) CAN (10.7%) JAP (10.2%) NOR (9.4%) AUS (9.4%)* UK (8.8%) Health Care Spending as Percentage of GDP, 1980–2013 Percent OECD Health Data 2015
  • 6. Rationale • Workers now paying at least $1,094 more in premiums annually for coverage than in 2000 • Still waiting to see what ACA does… • Despite expenditures, can we document significant improvement in health status? • Life expectancy • Perinatal mortality
  • 9. Our poor health • Life expectancy 78.8 vs. 81.2 (2013) • Infant mortality 6.1/1000 vs. 3.5 (2011) • 68% adults over 65 had at least 2 chronic conditions (33% UK, 56% Canada; 2014) • Obesity rate 35.3% (3.7% Japan, 30.6% NZ; 2014) • Smokers 13.7 (12.8 Australia, 24.1 France; 2013)
  • 10. US does more of… United States Rank compared with OECD countries OECD median MRI units 35.5 per 1,000,000 2nd (Japan 1st) 11.4 per 1,000,000 MRI exams 106.9 per 1000 50.6 per 1000 CT units 43.5 per 1,000,000 3rd (Japan 1st, Australia 2nd) 17.6 per 1,000,000 CT exams 240 per 1000 136 per 1000 Tonsillectomy 254.4 per 100,000 1st 130 per 100,000 Coronary bypass 79.0 per 100,000 3rd 47.3 per 100,000 Knee replacement 226 per 100,000 1st 121.6 per 100,000 Caesarian section 32.9 per 1000 live births 6th 26.1 per 1000 live births OECD Health Data 2012 and 2015
  • 11. Why? • Current payment structure means MDs get paid more to do more interventions • ACA has not had effect yet… • Fear of litigation • Patients ask for more—and usually do not pay out of pocket (insurance covers)
  • 12. Why? • MDs in US paid more but seen less • 4 visits/year (OECD average 6.5 visits/year) • US uses more pharmaceuticals • 2.2 drugs per adult (1.2 Netherlands) • Cost more (US=100; Australia=49; Germany=95) • Procedures cost more • Bypass 75,345 vs. 42,130 in Australia • Appendectomy 13, 910 vs 4,995 Netherlands
  • 14. Bit of background • Technology assessment hierarchy • Based on efficacy (benefit under ideal circumstances) • Differs from effectiveness (benefit in real world) • Differs from efficiency (sufficient value? adds economic component) Fryback D, Thornbury J. The efficacy of diagnostic imaging. Med Dec Making 1991;11:88-94
  • 15. Technology assessment hierarchy • Technical efficacy • Diagnostic accuracy • Diagnostic impact • Therapeutic impact • Patient outcomes • Societal outcomes Fryback D, Thornbury J. The efficacy of diagnostic imaging. Med Dec Making 1991;11:88-94
  • 16. Technical efficacy • Ability of technology to produce image • Measured by image characteristics • Signal to noise, resolution of line pairs, etc. • Developmental: prototype and phantom, safety • Exploratory: possible use, wide range of patients, situations, conditions
  • 17. Technical efficacy • Ideas of what falls here?
  • 18. Diagnostic accuracy • Distinguish between disease, non-disease, or normal, abnormal • Establish diagnosis if symptoms • Screen in asymptomatic patient • Quantify severity or extent of disease • Prognostic information (staging) • Monitor therapy • Need gold standard; measure of truth
  • 19. Diagnostic accuracy • Sensitivity, specificity, positive and negative predictive values • 2 x 2 table to display comparison to gold standard or reference • Area under ROC curve • Most radiology research falls into this category
  • 21. Diagnostic impact • Which imaging tests have greatest impact on clinical decisions? • Before test need to know • Differential diagnosis • Degree of certainty of diagnoses • Probability of each consideration • Reassess after test performed • Best done prospectively
  • 23. Therapeutic impact • Which imaging tests have greatest impact on therapeutic decisions? • Percent procedures avoided due to imaging • Percent imaging changed plan • Retrospectively inferred from records • Prospectively assessed using clinicians’ stated plans
  • 25. Patient outcomes • Traditionally, morbidity or mortality • Change in quality adjusted life years (QALY) • Can be difficult to prove link as imaging study several steps removed from patient outcome
  • 26. Societal outcomes • Cost analyses from societal perspective • Usually reported as cost per QALY • Must be below some threshold to be accepted (? $50,000/QALY) • Example, head CT to screen for brain mets in lung cancer patients or only if symptomatic? • $70,000/QALY deemed too high to adopt
  • 28. Outcomes research—goals • Use imaging wisely • Avoid over-diagnosis and over-treatment • Risks of treatment in early or mild disease may outweigh benefits • Data that will permit maintenance of quality at reduced cost • Increasingly important • Radiology needs to prove worth
  • 29. Outcomes research—getting started • Choose technology to be evaluated • Define specific clinical indication • Define target population • Choose alternative techniques for comparison (may include no imaging) • Define outcomes affected by technology
  • 30. Randomized controlled trial • Gold standard of outcomes research • Patients assigned to different groups based on imaging (or no imaging) to be performed • Measure appropriate outcomes • Confounders and biases accounted for in design
  • 31. RCT—limitations • Cost • Size of population • Power calculation • Estimate expected differences • Time to complete • Technology obsolete by time outcomes assessed? • Technology as moving target • Results may only apply to narrow spectrum of patients; issue of generalizeability
  • 32. RCT—unnecessary when? • Anecdotal evidence of benefit dramatic; observational studies suffice • Example: head CT in trauma • New technology: • Has same or fewer side effects • Has same or better accuracy • Is less expensive
  • 33. RCT—necessary when? • Really don’t know answer • New technology: • More expensive • May result in more morbidity • Very controversial
  • 34. ACRIN—NLST • National Lung cancer Screening Trial • 25,000 patients randomized; low dose CT vs. CXR • Baseline, years 1 and 2 • Q6 month health status and medical intervention questionnaire • Very detailed description of what constitutes positive screen and what to do • Outcomes • Primary: Lung cancer specific mortality • Secondary: huge list
  • 35. ACRIN—NLST • Low dose CT more sensitive than radiography • Two annual incidence screenings with low- dose CT resulted in • Decrease number of advanced-stage cancers • Increase number of early-stage lung cancers NEJM 2013;369:920-921
  • 36. Example to work through • CT for appendicitis: improve outcomes? • Technology:
  • 37. Example to work through • CT for appendicitis: improve outcomes? • Technology: MDCT
  • 38. Example to work through • CT for appendicitis: improve outcomes? • Technology: MDCT (?oral, ?IV, ?rectal) • Indication:
  • 39. Example to work through • CT for appendicitis: improve outcomes? • Technology: MDCT (?oral, ?IV, ?rectal) • Indication: RLQ pain, fever, etc. • Population:
  • 40. Example to work through • CT for appendicitis: improve outcomes? • Technology: MDCT (?oral, ?IV, ?rectal) • Indication: RLQ pain, fever, etc. • Population: patients presenting to ER • Alternative:
  • 41. Example to work through • CT for appendicitis: improve outcomes? • Technology: MDCT (?oral, ?IV, ?rectal) • Indication: RLQ pain, fever, etc. • Population: patients presenting to ER • Alternative: best clinical judgment plus lab • Outcomes:
  • 42. Example to work through • CT for appendicitis: improve outcomes? • Technology: MDCT (?oral, ?IV, ?rectal) • Indication: RLQ pain, fever, etc. • Population: patients presenting to ER • Alternative: best clinical judgment plus lab • Outcomes: morbidity, mortality, perforation rate, length of hospital stay, surgery rate, % normal appendices removed, cost to hospital, etc.
  • 43. Example to work through • CT for appendicitis: improve outcomes? • Technology: MDCT (?oral, ?IV, ?rectal) • Indication: RLQ pain, fever, etc. • Population: patients presenting to ER • Alternative: best clinical judgment plus lab • Outcomes: morbidity, mortality, perforation rate, length of hospital stay, surgery rate, % normal appendices removed, cost to hospital, etc. • Is this feasible?
  • 44. What if RCT not feasible? • Modeling: Use of decision analysis software • Need information from other studies • Accuracy, sensitivity, specificity • Course of disease, natural history • Effectiveness of therapy • Patient utilities and costs • Meta-analyses for combining results of studies
  • 45. Modeling • Assumptions about test parameters (sensitivity, specificity) based on literature • Then vary assumptions • Assumptions about frequency of outcomes based on personal experience or literature (mortality and morbidity rates, etc.) • Allows variation of these
  • 46. Modeling—literature example Cervical spine screening with CT vs. plain films Blackmore CC et al. Radiology 1999;212:117-125
  • 47. CT in appendicitis example • Retrospective study of 651 patients treated before use of CT for appendicitis (1992-1995) • 52% went to surgery; 24% for observation then surgery; 24% observation to discharge • Initial surgical group, 81% appendicitis • Observation to surgery group, 80% appendicitis • Observation to D/C group, assumed no appendicitis (no returns to their hospital) Rhea JT et al. AJR 1997;169:113-118
  • 48. CT in appendicitis example • Apply sensitivity and specificity of CT from literature to 100 theoretical patients • Use morbidity, mortality, perforation, correct diagnosis rates from study of 651 patients • Model impact of CT on outcomes • Decrease in nontherapeutic surgery (13 patients), observation days (1 day less on average), perforation rate, etc. • Alternative diagnoses also confirmed Rhea JT et al. AJR 1997;169:113-118
  • 49. Intermediate outcomes • Choose levels lower in hierarchy • Diagnostic impact • Change in diagnostic possibilities and confidence • Before and after test ask referring MD: • Differential diagnosis • % certainty for each diagnosis • Degree of satisfaction with test
  • 50. Intermediate outcomes • Seattle Low Back Pain Imaging Project (SLIP) • RCT of plain films vs. MR as initial imaging • Diagnostic impact • 15% new diagnosis with plain film • 30% for MRI (spinal stenosis, HNP, etc) • Change in probability for most likely diagnosis for HNP (p<.002) Jarvik JG et al. Radiology 1997;204:447-454
  • 51. Intermediate outcomes • Therapeutic impact • Before test, ask MD for treatment plan • Observe if plan changes after test • If temporally related, infer causal relationship between test and plan change • SLIP: 38% avoided additional test if plain film first, 64% if MRI first (p<.001)
  • 52. Intermediate outcomes • Advantages • Cheaper and easier than full RCT (RCT may show intermediate outcomes if assessed as in SLIP) • No one denied potentially good test • Works well for “add-on tests” • Tests that would not necessarily replace existing work-up
  • 53. Intermediate outcomes • Disadvantages • Stated plans do not necessarily equal clinical action • If treatment invasive, difficult, expensive or risky, easier to hypothetically choose • RCT still better for “replacement test”
  • 54. Patient outcomes • Reduced mortality • Reduced morbidity • Decrease complications from other tests • Avoid invasive procedures • Faster return to normal (or pretest) state
  • 55. Patient outcomes • Quality of life: ability to undertake rewarding and enjoyable activities • How person feels, functions • Quality adjusted life years (QALYs) • Patient satisfaction, reassurance • Physiologic marker improvement • Improved function
  • 56. Functional status • General and specific measures • SF-36 (Short Form 36) • 36 questions, 8 domains • General health, pain, physical functioning, social functioning, mental health, etc. • Roland score (back pain specific index) • Sciatic frequency index
  • 57. SLIP follow-up • RCT ultimately enrolled 380 patients • Roland score at 12 months did not differ between 2 groups (8.75 plain film vs. 9.34 MRI, p = .53) • p >.75 for all domains of SF-36 • Patient reassurance measure • 1 month p=.09 • 3 month p=.08 • 12 month p=.002 (58% plain film vs. 74% MRI) • Increased reassurance with MRI over time • 10 in MRI vs. 4 in plain film had surgery (p =.09) Jarvik JG et al. JAMA 2003;289:2810-2818
  • 58. Outcomes research—limitations • Tenuous link between test and patient outcome • Test usually steps removed from outcome • Best situation: proven link between therapy and outcome  link diagnosis to therapy to outcome • Example: detection of cerebral aneurysms with MRA  treatment decreases morbidity and mortality  link MRA to reduced morbidity/mortality
  • 59. Outcomes research—limitations • If no proven impact on outcome, may be: • Diagnostic test (inaccurate, insensitive) • Referring clinician (inappropriate use of info; info does not reach) • Therapy (ineffective or unavailable) • Patient (lack of compliance) • System (lack of patient access)
  • 60. Alternative outcome • Therapeutic value of diagnostic test • Diagnostic test can have direct impact on quality of life • Reassurance important to patients and referring clinicians • If consider only positive tests fail to consider that negative test has value
  • 61. Alternative outcome • Example: nonspecific chest pain • 176 patients randomized to receive no studies or EKG and CPK level • Test group had less short term disability (20% versus 46%; P=.001) • Use of diagnostic test independent predictor of recovery in logistic regression • Test group felt care “better than usual” Sox HC, et al. Ann Intern Med 1981;95:680-5
  • 63. Economic evaluations • Must establish viewpoint of analysis • Society, insurer, provider, patient • All have different inputs and outcomes; different costs to consider • Example: patient may value increased quality of life at any cost while HMO may value decreasing overhead costs
  • 64. Costs • Value of resources • Charges as estimate • Medicare/Medicaid reimbursement as estimate • Micro-costing techniques—all resources identified and quantified • Consider cost of missed abnormalities and superfluous work-up of false positives
  • 65. Costs—examples • Medical • Diagnostic test and interpretation • Nursing care • Nonmedical • Time (travel, tests and treatment) • Travel • Support with daily activities • Lost productivity
  • 66. Types of economic analyses • Cost identification or cost minimization • Compares alternative health care strategies assumed equally effective
  • 67. Cost identification—example • CT vs. US plus plain film after ESWL • 25 patients underwent all 3 • Assumed equal accuracy • Exams timed; direct technical costs calculated using procedural-based accounting system • CT 15.3 minutes; US + plain film 37.2 minutes • CT $36.86; US + plain film $57.60 • Sensitivity analysis on time; within reasonable range CT always cost less Remer EM et al. Radiology 1997;204:33-37
  • 68. Types of economic analyses • Cost benefit • Costs and benefits assigned dollar values • Comparison of different technologies • Limited use in medicine due to difficulty assigning monetary values to health states • Cost effectiveness leaves outcomes in natural, objective units • # lives saved, # cancers detected, etc.
  • 69. CEA—example • What gain from sixth stool guaiac? • Progression from fifth to sixth stool guaiac costs $47 million/cancer detected • Can use intermediate outcomes (length of stay, readmission rates, number of unnecessary surgeries prevented) Neuhauser D, Lewicki AM. N Engl J Med 1975; 293
  • 70. Cost utility analysis • Similar to CEA but uses subjective outcomes that explicitly incorporate patient preferences • QALY: basic outcome measure • Year of life with significant morbidity less desirable than year with excellent health • Full health equals 1 • Death equals 0 • Morbidity somewhere between 0 and 1 • Preferred method for economic evaluations
  • 71. Utility values • Side effects from interferon—.93 • Mild angina—.90 • Prophylactic mastectomy, oophorectomy—.86 • Moderate angina—.70 • Hospitalization for tuberculosis—.60 • Severe angina—.50 • Moderate to severe stroke—.39 • Recurrent stroke—.12
  • 72. Components of economic analysis • Most with modeling software • Reference case (defines typical patient) • Strategies (set of alternative care pathways) • Time horizon (period to consider risks and benefits) • Perspective (viewpoint of decision maker) • Effectiveness measure (outcome) • Decision tree allows varying components
  • 73. Decision tree Singer ME and Applegate KE. Radiology 2001;219:611-620
  • 74. Components of economic analysis • Probabilities (all events in model have one) • Estimated from historical data; literature, pilot study or expert opinion • Utilities (assign utility to every health condition in model) • Costs (identify all relevant) • Assumptions (due to limitations imposed by available data; must explicitly state)
  • 75. Economic analysis—example • Whole body CT screening • Base-case analysis: from cohort of 500,000 50 year old men • Sex, age varied in analysis • Benefits of screening assumed due to earlier detection of disease, improvement in survival • 8 conditions included in model: ovarian, pancreatic, lung, liver, kidney and colon cancer; aortic aneurysm, coronary artery disease • Costs of screening, follow-up tests, patient care Beinfeld, M. T. et al. Radiology 2005;234:415-422
  • 76. Beinfeld, M. T. et al. Radiology 2005;234:415-422 Schematic of decision-analytic model with life expectancy and cost outcomes used in the cost effectiveness analysis
  • 77. Economic analysis—example • Compared to routine care • 6 days of life gained at $2513/patient • Incremental cost: $151,000/life-year gained • 90.8% had finding; only 2% had disease • 32.3% of total cost = work-up of FPs • With favorable assumptions, one time screening not cost effective; would add burden to healthcare system Beinfeld, M. T. et al. Radiology 2005;234:415-422
  • 78. Economic analysis—limitations • Not all agree with strategies chosen • Was comparison to standard of care? • Were all reasonable options considered? • Was reference case typical of patient population? • Fears of bias if study funded by drug company or equipment manufacturer • If too early, data scant; if late, decisions made and model obsolete • Noncontrast CT for renal colic: accepted standard without study of cost
  • 79. Conclusions • Need to be familiar with concepts • Way to inform health care policy makers, guide research • Goal to improve quality in rational manner, decrease waste and unnecessary expenditure • Enhance Radiology community’s ability to participate in health care decision-making locally and nationally • Preserve our specialty!
  • 81. US does more of… OECD Health Data 2015