Medical Decision Modeling Inc. - MBSW Online

Medical Decision Modeling Inc MBSW Online-ppt Download

  • Date:04 Aug 2020
  • Views:26
  • Downloads:0
  • Size:717.00 KB

Share Presentation : Medical Decision Modeling Inc MBSW Online

Download and Preview : Medical Decision Modeling Inc MBSW Online

Report CopyRight/DMCA Form For : Medical Decision Modeling Inc MBSW Online


Transcription:

presentation to 2012 Midwest BiopharmaceuticalStatistics WorkshopMay 22 2012by Harry J Smolen President and CEO.
Medical Decision Modeling Inc Indianapolis IN USA Discussion Agenda General background onpharmacoeconomic PE models.
Types of models commonly used inhealthcare technology assessment Methods for selecting modeling Cost effectiveness analysis General PE Modeling Background.
What is a model A model is a hypothetical description of General PE Modeling BackgroundWhy not experiment with the actual If it is possible to experiment with the.
actual system do it However especially in healthcare itis frequently too expensive toodisruptive too slow and or unethicalto experiment with the actual system.
General PE Modeling BackgroundExample experiment How many life years wouldbe saved if every other person in the US 65years had a one time colonoscopy withpolypectomy and surveillance every three to five.
years for positive findings Expensive 1 2 x 40 3M persons x 1 714 34 537 100 000 follow up costs Disruptive a day of inconvenience per patientfor prep and exam also disruption to.
gastroenterologists is there enough capacity Slow Many years to get results Possibly Unethical though uncommon perforations occur 0 9 1000 colonoscopies withunknown benefit to patients.
Computational vs Physical Models When most people think of models they envisionphysical models e g clay cars in wind tunnelsor life size layouts of an emergency room forworkflow analysis.
The types of models used to evaluate theeffectiveness and cost effectiveness ofhealthcare technologies are computational Computational models represent a system interms of logical and quantitative relationships.
that are manipulated to examine how the modelreacts and thus how the system would react ifthe model is valid11 Law and Kelton Simulation Modeling and Analysis 2 nd edition 1991 Simulation vs Analytical Solution.
A simple computational model d rt whered distance traveledr rate of travelt time spent traveling.
In this simple model it is possible to get an exact analytical solution This equation may apply to a single car on a testtrack but Is of little use to determining distance traveled on a.
busy highway with varying speeds and levels ofcongestion Simulation vs Analytical SolutionIn this simple d rt model it is possible to get an exact analytical solution.
Some analytical solutions can become extraordinarilycomplex e g inverting a non sparse matrix and requiresubstantial computing power If an analytical solution to a computational model isavailable and computationally efficient study the model.
in this manner However many systems are highly complex so that validcomputational models of them are themselves complex precluding the possibility of an analytical solution Such models must be analyzed by simulation i e .
numerically varying the relevant model inputs todetermine how they affect the output measures of1 Law and Kelton Simulation Modeling and Analysis 2 nd edition 1991 Evidence from RCTs Evidence from randomized controlled.
trials RCTs remain the highest quality data source for evaluating theefficacy of health care interventions However evidence from RCTs alonecan be uninformative if the RCT.
endpoints are not translated intomeasures that are valued by patients providers insurers and policy1 Weinstein MC et al Value Health 2003 Jan Feb 6 1 9 17 Evidence from RCTs.
For example in patients withosteoporosis fracture risk evolves over alifetime from the development of peakbone mass to subsequent bone loss andthe age related increase in the likelihood.
of falling among other factors1 It may takes decades to determine theeffects of osteoporosis interventions Hence the most informativeosteoporosis intervention RCTs would.
potentially last for decades21 NIH Osteoporosis Prevention Diagnosis and Therapy Consensus Statement 2000 2 Vanness DJ Osteoporos Int 2005 Apr 16 4 353 8 Evidence from RCTs A similar situation exists for many.
other chronic diseases such asdiabetes colorectal and prostatecancer and Alzheimer s disease The lengthy clinical trials needed toproperly evaluate interventions for.
these chronic diseases would requirefinancial and time opportunity coststhat would make them infeasible11 Vanness DJ Osteoporos Int 2005 Apr 16 4 353 8 Evidence from RCTs.
Very few epidemiological studies orclinical trials are able to measuredisease progression and the impact ofinterventions on costs quality of life and health outcomes over a lifetime1.
In the absence of such information themost practical method to evaluate thehealth outcomes and costs ofinterventions is to develop models thatintegrate relevant data and extrapolate.
to long term time horizons1 Br ndle M et al Curr Med Res Opin 2004 Aug 20 Suppl 1 S1 3 Evidence from Models vs RCTs Models that evaluate health careinterventions synthesize evidence on.
health consequences and costs frommany different sources including datafrom clinical trials observational studies insurance claim databases caseregistries public health statistics and.
preference surveys11 Weinstein MC et al Value Health 2003 Jan Feb 6 1 9 17 Evidence from Models vs RCTs Even when the disease does not require a long term evaluation period models can prove.
valuable by Using results from indirect comparisons ofindividual RCTs to compare treatments notstudied head to head and estimateoutcomes not consistently measured.
Allowing the extrapolation of effects topopulations not studied in a particular RCT Allowing sensitivity analysis of assumptionsaddressing treatment efficacy health stateutilities costs etc .
1 Weinstein MC et al Value Health 2003 Jan Feb 6 1 9 17 Decision Model DefinedA decision model is a structured representationof a decision process that allows a person toperform a decision analysis.
decision analysis is just the systematic articulation ofcommon sense Any decent doctor decision maker reflects on alternatives is aware of uncertainties modifiesjudgments on the basis of accumulated evidence balancesrisks of various kinds considers the potential consequences.
of his or her diagnoses and treatments and synthesizes allof this in making a reasoned decision that he or shedecrees right for the patient All that decision analysis isasking the doctor decision maker to do is to do this a lotmore systematically and in such a way that others can see.
what is going on and can contribute to the decisionprocess 11 Raiffa H Clinical Decision Analysis Philadelphia WB Saunders 1980 ix x Types of Modeling MethodsTypes of modeling methods frequently used in health.
technology assessment Decision trees Markov Cohort Monte Carlo.
Microsimulation Fixed time advance Discrete event Time to event without and withqueuing for resources Agent based.
Decision Trees A decision tree is a diagrammaticrepresentation of the possible outcomes andevents used in decision analysis The questions to be asked in an analysis of a.
question are arranged as a series of decision orchance nodes each node with resultingbranches creating a tree effect The sequential steps proceed with each stepdepending on the decision or probability.
outcome from the preceding step11 http medical dictionary thefre... Decision Tree ExampleSimple tree fragment modeling complications ofanticoagulant therapy1.
1 Sonnenberg FA et al Med Decis Making 1993 Oct Dec 13 4 322 38 Decision Tree ComponentsSimple tree decision trees embody the essentialparadigm of decision analysis Specifically alldecisions may be decomposed into three.
broadly defined components 1 Decision node point in time when a choice ismade among competing strategies2 Decision strategy set of actions or eventsconsequent to a decision.
3 Outcome nodes terminal branches of tree thatrepresent outcomes of a strategy 1 Multipleoutcomes payoffs may be assigned 1 Stahl JE Pharmacoeconomics 2008 26 2 131 48 Calculating Expected Value of a Decision Tree.
The expected value of a decision tree is calculatedby averaging out or folding back the branches of The value s of each strategy is path probability tothe terminal node multiplied by the payoff s at theterminal node.
Branching probabilities can be deterministicrepresented by point values e g 0 6 or stochasticrepresented by probability distributions e g normal exponential Uncertainty around branching probabilities and.
terminal node values is examined with sensitivity1 Stahl JE Pharmacoeconomics 2008 26 2 131 48 Advantages of Decision Trees Graphical can diagrammatically representdecision alternatives chance events and possible.
outcomes visual approach assists withcomprehending decision sequences anddependencies Efficient can quickly express complex alternativesclearly and easily modify as new information.
becomes available Complementary can use in conjunction with othermethodologies e g append recursive methods toterminal nodes11 Olivas R http www stylusandslate com de... .
tree v5 1b pdf Disadvantages of Decision Trees Must assume population being examinedcan be modeled in the aggregate if beingapplied to an individual assumption is.
made that aggregate probabilities arerelevant to the individual1 Does not specify when events occur Assumes that each event can occur only Can address previous two disadvantages with a.
recursive tree see next slide 21 Stahl JE Pharmacoeconomics 2008 26 2 131 48 2 Sonnenberg FA et al Med Decis Making 1993 Oct Dec 13 4 322 38 Recursive Decision Tree1Previous terminal nodes of.
POST BLEED POST EMBOLUS AND NO EVENTare replaced by the chancenode ANTICOAG whichappears at the root of the.
Note that with only two timeperiods there are 17terminal nodes five periodswould have hundreds of1 Sonnenberg FA et al Med Decis Making 1993 Oct Dec 13 4 322 38 .
Markov Models Described as partially cyclic directed Particularly useful when a decisionproblem involves exposure to risks or events over time.
ongoing exposures or situations where thespecific timing of an event is regarded asimportant or uncertain or where describing the timing of events isnecessary for face validity1.
1 Stahl JE Pharmacoeconomics 2008 26 2 131 48 Markov Models Assumes that the patient is always in oneof a finite number of health states calledMarkov states.
All events of interest are modeled astransitions Each state is assigned a utility andpossibly a cost and the contribution ofthis utility depends on the length of time.
in the state11 Sonnenberg FA et al Med Decis Making 1993 Oct Dec 13 4 322 38 Markov Models Common representation of a simpleMarkov process called a state .
transition diagram Each state represented by a circle Arrows connecting different statesindicate allowed transitions States with arrows to itself indicate.
that patient may remain in thatstate in consecutive cycles Note no transition from DISABLED to WELL nor DEAD to any other Markov Models Time Cycles.
Time horizon of the analysisdivided into equal increments oftime called Markov cycles Assumed that a patient can onlymake a single state transition.
during a cycle Length of cycle chosen to representa clinically meaningful time interval If time horizon is patient lifetime then cycle is usually one year.
If events occur more frequently cycle can be a month or even a1 Sonnenberg FA et al Med Decis Making 1993 Oct Dec 13 4 322 38 Markov Models Incremental Utility Evaluation of a Markov process yields the.
average amount of time spent in each state patient is given credit for time spent ineach state e g life years Optionally each state can be associated witha numerical factor representing the quality of.
life in that state relative to perfect health e g WELL 1 0 DISABLED 0 7 DEAD 0 0 Utility associated with spending one cycle ina particular state is referred to as theincremental utility1.
1 Sonnenberg FA et al Med Decis Making 1993 Oct Dec 13 4 322 38 Markov Models Costs Analogous to utilities assigned to particularstates a cost may be specified for each state.
Types of Modeling Methods Types of modeling methods frequently used in health technology assessment: Decision trees Markov Cohort Monte Carlo Microsimulation Fixed-time advance Discrete-event, Time-to-event (without and with queuing for resources) Agent-based Decision Trees A decision tree is a diagrammatic representation of the possible ...

Related Presentations

Roche Template MBSW

Midwest Biopharmaceutical Statistics Workshop. XiangyunWang, Statistical Programming and Analysis . Roland Morley & Chen Huang, PD Drug Safety. Genentech, Inc.

21 Views0 Downloads

ISPE OFFICE SPACE ANALYSIS MBSW

Comparable to AQL probability for single test. Bayesian approach also available2 * 1Bergum, J. and Li, H. “Acceptance Limits for the New ICH USP 29 Content-Uniformity Test”, Pharmaceutical Technology , Oct 2, 2007 2Leblond, D., and Mockus, L. “Posterior Probability of Passing a Compendial Test.” ... E. “Comparison of Tests for ...

4 Views0 Downloads

UDU iDay Project MBSW

How can one concisely and clearly summarize a product’s capability to meet complex criteria, such as that of ASTM E2810, from multiple Stage 1 (i.e., development) studies? 2015 MBSW, Tuesday, May 19, 2015. Jeff Hofer and Adam Rauk, Eli Lilly

2 Views0 Downloads

DCO capabilities SOSSEC Inc SOSSEC Inc

Mr. Howard will introduce the APM’s, CPS will go first and then Mr. Howard will come back and introduce CAD. The DCO Tools Suite RDP defines the required capabilities that allow CPTs, RCCs, and other cyberspace defenders to perform DCO and cybersecurity tasks based on mission, threat, and/or directives.

11 Views0 Downloads

MULTI CRITERIA DECISION ANALYSIS AND DECISION SUPPORT SYSTEMS

Quantify uncertainty. Quantify preferences. Combine uncertainty and preferences . Choice of A Decision Rule Depends On. Importance of decision consequences. Complexity of decision task. Time pressure. Similarity of alternatives. Rating on twodimensions: Value ofconsequences. Probabilityoftheiroccurence.

20 Views0 Downloads

Modeling Consumer Decision Making and Discrete Choice Behavior

Classical Inference. Population. Measurement. Characteristics. Behavior Patterns. Choices. Imprecise inference about the entire population – sampling theory and asymptotics

28 Views0 Downloads

Intelligent Modeling for Decision Making

Intelligent Modeling for Decision Making Katta G. Murty Industrial and Operations Engineering ... Storage and Retrieval of Import Containers Top View of a Block B1 Being Served by an RTGC Land Scarcity for Terminal Development in Hong Kong The Highest Land Utilization Terminal in the World Key Service Quality Metrics Objectives of the Study ...

22 Views0 Downloads

Chapter 2 Decision Making Systems Modeling and Support

In a DSS the what-if and the goal-seeking options must be easy to perform 2.10 The Implementation Phase There is nothing more difficult to carry out, nor more doubtful of success, nor more dangerous to handle, than to initiate a new order of things (Machiavelli [1500s]) *** The Introduction of a Change *** Important Issues Resistance to change ...

27 Views0 Downloads

Chapter 1 Introduction to Managerial Decision Modeling

Types of Decision Models Quantitative vs. Qualitative Data Spreadsheets in Decision Making Steps in Decision Modeling Steps in Modeling Example Model: Tax Computation Tax Brackets Percent of Taxable Income Taxable Income up to $14,600 10% $14,601 to $59,400 15% $59,401 to $119,950 25% Example Model: Break-Even Analysis Slide 11 Possible ...

28 Views0 Downloads

Spreadsheet Modeling amp Decision Analysis

Introduction to Mathematical Programming MA/OR 504 Chapter 7 Machine Learning: Discriminant Analysis Neural Networks 6-*

18 Views0 Downloads

Decision Trees and Modeling Approach Federico Girosi RAND

then the main points could come up one at a time in "build" fashion. Can we streamline the verbiage on this chart--maybe even make it into a kind of table with adult and child across the top and the important factors down the side? then the main points could come up one at a time in "build" fashion.

12 Views0 Downloads

Modeling Organizational Decision Making in Military

Outline Outline III Corps battle management organizational structure Summarize Army frustration with measuring the ROI of digitization Review of C2 decision modeling paradigms The multi-tiered decision process within a battle staff Coping with situational ignorance: the need for sensemaking System performance measurement: the need for a ...

71 Views4 Downloads