-
steps of the outcomes research process
- identify
- measure
- value
- compare
-
steps of outcomes research taxonomy
- data collection & data estimation methods
- PE method; CA,CEA,CUA,CBA
- PE & quality of life research
- outcomes research
-
technical efficiency
when the maximum possible outputs are produced (over time) from a given set of inputs
-
allocative efficiency
when inputs are put to their best possible use so that no further gains are possible
-
what analysis is used when dealing with allocative efficiency
CBA
-
what analysis is used when dealing with technical efficiency
-
5 process' that are assisted with PE
- budget planning
- pricing decisions
- formulary decision making
- evaluation of clinical treatment guidelines
- evaluation of pharmaceutical care services
-
what are the outcome measurement units in CBA
monetary units
-
what are the outcome measurement units in CEA
natural units
-
what are the outcome measurement units in CMA
assumed to be equivalent in comparable groups
-
what are the outcome measurement units in CUA
Quality-adjusted life years or other utilities
-
cost-consequence analysis
lists the costs and outcomes associated with each Tx
-
cost of illness study
attempts to determine the monetary burden of a particular disease or condition on society as a whole
-
direct medical resources
- medications
- hospital days
- tests
-
other direct (non-medical) resources
- input into treatment
- other pt expenses
-
indirect resources
- unpaid assistance
- days lost from work
- decreased productivity
-
-
marginal cost
the cost to produce one more unit of outcome
-
total cost
sum of all health care expenditures for entire population
-
average cost
total cost of treatment / total # of pts treated
-
5 process' for cost determination
- specify inputs
- count the units
- assign dollar values
- adjusting for differences in the timing of cost
- allow for uncertainty
-
-
counting units
- hospital days
- single lab test
- 30 day supply of medication
-
primary data
- clinical trials
- naturalistic trial/ observation cost/ accounting system
-
secondary data
- administrative/ claims database
- literature/ published AWP
- expert opinion
-
model data
- decision analytical
- statistical
-
discounting equation
- sum F / (1+ r)tF - future cost
- r - discount rate 5%
- t - number of years from present time
-
sensitivity analysis
technique of systematically varying key variables in an analysis to determine if the results of the analysis change
-
direct costs
expenditures on tangible products or services, which contribute to the gross national product
-
direct benefits
measurable positive economic, clinical & humanistic outcomes that result from these products or services
-
indirect costs
include unpaid resource commitments and the cost of morbidity and mortality
-
indirect benefits
benefits derived rom avoiding indirect costs
-
intangible costs
costs of pain and suffering and take into account the preferences of patients/society
-
intangible benefits
benefits derived from avoiding the pain and suffering
-
human capital approach
- if medical care is seen as an investment in productivity
- value intervention using earning potential
-
revealed preference approach
value intervention/life based on wage differences between jobs with different levels of risk
-
willingness to pay / contingent valuation
value intervention based on preferences
-
cost analysis
- compares the cost of 2 or more alternatives
- use direct & indirect costs NOT outcomes
- useful for budget considerations
-
break even equation
sum (fixed costs) + (#of units (sum(variable cost/unit) = (# of units)x(price)
-
cost calculation of a CA
(direct + indirect costs) - (direct + indirect cost savings)
-
cost calculations of CBA
(direct costs + indirect costs + intangible costs)
-
when are modes used
to study a phenomena when complete data are lacking
-
decision analysis breaks down decisions into what
- alternatives
- probabilities
- outcomes
-
types of decision analytic models
- decision tree
- markov model
- they are deterministic models because they account for 100% of the variation of the subjects being modeled
-
what type of relationship is used in a decision analytical model
functional relationship - for every level o X there is only one value for Y
-
types of nodes in a decision tree
- decision node = square
- chance node = circle
- terminal nodes = triangle
-
6 recommendations of decision tree construction
- tree must have balance
- only 2 branches after each chance node
- no embedded decision nodes
- the branches must be linked
- the tree must have symmetry
- the order of the branches does not matter
-
example of when a CEA would be used
- to determine which alternative provides the most benefit for a given dollar amount
- or
- to determine which alternative can produce the desired benefit or effect at the lowest cost
-
measuring costs equation for CEA
(direct costs + indirect costs) - (direct + indirect costs savings)
-
how do you determine the cost/outcome in a CEA
- identifying and measuring outcomes
- valuing inputs
-
what does an ICER show
how much cost is increased when the marginally better treatment option is chosen for one patient
-
criteria for ICER pt groups
- mutually exclusive treatments alternatives
- not between independent treatment alternatives in different pt groups
-
formula for ICER's
change in cost / change in effectiveness
-
what makes a treatment dominated
less effective and more costly
-
what is extended dominance
- a Tx has a higher ICER than a more effective Tx alternative
- effectiveness is produced at a higher marginal cost than necessary
-
2 methods of prospective data collection
- TTO - time trade off
- SG - standard gamble
-
when would you do a cost minimization analysis (CMA)
when you have 2 treatments and the total effects are equal
-
models for decision analysis
- decision tree
- markov model
-
modesl for statistical analysis
regression model
-
how does a CUA work
it converts the effectiveness data to a common unit of measure, mortality or morbidity
-
if you are going to capture naturalistic outcomes what model do you use
CEA
-
what model would you use if the effectiveness data indicate that alternatives are equal
CMA
-
definition of utility
the level of subjective satisfaction, distress, or desirability that people associate with a particular health state
-
3 ways utility are determined
- judgement - educated estimates
- literature - published estimates
- measurement - prospective data collection
-
When you are doing standard gamble what is the point where you decide to do the treatment even at the expense of the risk
ambivalent point
-
what is rater in a TTO model
when a pt has knowledge of experience with the illness and it may influence their judgement
-
cost calculations for CEA
direct costs - direct cost savings
-
cost calculations for CUA
direct costs - direct cost savings
-
what type of benefit/outcome is associated with cost analysis
no outcomes measured, pick the cheapest product
-
what type of benefit/outcome is associated with CEA
natural units (life years gained, mm Hg blood pressure, mMol/L blood glucose)
-
what type of benefit/outcome is associated with CMA
assumed to equivalent in comparable groups
-
what type of benefit/outcome is associated with CBA
dollars or monetary units
-
what type of benefit/outcome is associated with CUA
quality adjusted life year
-
what are the aspects of health related quality of life
- physiological status
- functional status
- well being
- life satisfaction
-
what are the aspects of health status
- physiologic status
- functional status
- well being
-
what are the aspects of functional status
-
4 quality of life measures
- utility measures (SG, TTO)
- health profiles (SF 36)
- functional indices (index of ADL)
- disease specific measures (arthritis impact scale)
-
difference between generic and disease specific measurements
- generic has the ability to make comparisons between conditions
- disease specific has increased sensitivity
-
reliability of a model
- similarity of results provided by independent, but comparable measures of the same object.
- consistency
- a model can be reliable but not valid
-
validity of a model
- does the score measure what it is intended to measure
- accuracy
-
responsiveness of a model
sensitivity to clinical changes in disease over time
-
how do researchers minimize bias in PE models
- include all key characteristics of the intervention being modeled
- balance thoroughness with simplicity
- be transparent
|
|
