Lectures Flashcards
1
Q
what is the difference between positive and normative engagement? which one is used in health economics?
A
positive = technical, fact based
normative = value judgement
both involved in health economics!
2
Q
briefly explain the “economic problem”
A
resources (land, labour etc) are limited, but our ‘wants’ are unlimited.
choices need to be made which involve foregoing another alternative (opportunity cost).
3
Q
define opportunity cost
A
the cost of the alternative options that could have been funded
e.g. if we fund exercise classes for over 65s, opportunity cost may be free swimming lessons for kids
(or literally anything else that money/resources could have been used for)
4
Q
define technical efficiency
A
assesses the best way of achieving an objective - minimises the amount of resources required to achieve a particular objective
e.g. what is the best way to increase breastfeeding rates?
5
Q
define allocative efficiency
A
concerned with decisions about whether to allocate resources to a programme i.e. is decision worth doing for society
e.g. should we fund incentives for breastfeeding?
6
Q
what are markets?
A
forum where consumers and producers interact to exchange goods and services with minimum government intervention
7
Q
explain how price and quantity act as signals in markets
A
as price falls the quantity consumers want to buy increases, but the quantity sellers want to sell will decrease.
inverse occurs when price rises.
8
Q
define utility
A
term for overall wellbeing or satisfaction or preference
9
Q
what assumptions are made about consumers and suppliers within markets/
A
consumers = utility maximisers
suppliers = profit maximisers
10
Q
define marginal benefit in the context of markets
A
the value gained from the last unit - this will be reflected in the price consumers are willing to pay
11
Q
define marginal cost in the context of markets
A
the cost of the last unit to the supplier - will be reflected in the price suppliers are willing to accept
12
Q
when does a market ‘clear’
A
when marginal social benefit = marginal social cost = price.
gains to society (utility and profit) are at their optimum.
aka market equilibrium.
13
Q
list factors that affect demand
A
price income price of other goods tastes and trends population size and composition
14
Q
list factors that affect supply
A
expectations of future price changes costs of production technology government regulation profitability of alternative products the supplier can produce profitability of goods in joint supply (e.g. butter and skimmed milk) random shocks (e.g. weather) number of suppliers
15
Q
list the 7 assumptions of a perfectly competitive market
A
- rational, self-interested consumer and profit maximising sellers
- numerous small sellers and buyers with no market power
- sellers can easily enter and leave the market
- perfect knowledge
- certainty
- homogeneous products
- no externalities or spillovers
16
Q
explain the concept of rational consumers and suppliers
assumptions of a perfectly competitive market
A
- buyers will buy something if it’s in their interests to do so (benefits outweigh the costs)
- sellers will provide what is required based on prices (which reflect what is valued)
17
Q
explain the concept of numerous, small sellers and buyers.
assumptions of a perfectly competitive market
A
lots of small sellers means they can’t agree to set price artificially high.
one (monopoly) or a few (oligopoly) sellers can artificially raise price.
this keeps costs minimised as producers are forced to set prices low to attract consumers.
18
Q
what are the fancy terms for one or a few buyers?
A
monopsony
oligopsony
19
Q
name two effects of competition on markets
A
- keeps price low
- improves quality as buyers can go elsewhere
20
Q
explain the concept of free entry and exit of markets
assumptions of a perfectly competitive market
A
if sellers can easily join and leave the market:
- ensure competition, new entrants where demand is growing
- poor quality will result in market exit
21
Q
explain the concept of perfect knowledge.
how does this apply to the healthcare market?
(assumptions of a perfectly competitive market)
A
consumers know:
- what’s available
- at what quantity
- and at what price
- and how much benefit they will derive from the good
allows them to choose what maximises their utility, and know where to find it.
sellers also need perfect knowledge of market.
in healthcare market - consumers generally DON’T have perfect knowledge of treatments etc. available, or which is the right one.
22
Q
explain how the concept of certainty works as an assumption of a perfectly competitive market.
how does this apply to the healthcare market?
A
buyers need to know when they will need goods/services and what price it will be at.
can’t plan health events in advance! but some aspects you can - e.g. can plan when you’ll next need to buy a new pack of contact lenses.
23
Q
explain the concept of a homogenous product
assumptions of a perfectly competitive market
A
in a perfect market, product is identical making it is easy to substitute between sellers.
24
Q
define product differentiation
A
techniques such as branding that allow sellers to make more profit even when selling a homogenous product
e.g. a million types of nurofen, all just ibuprofen!
25
# define externalities
(assumptions of a perfectly competitive market)
an externality is a benefit or cost from consumption or production that affects others (apart from the buyer or seller, so it's not reflected in price).
can be positive or negative.
if they aren't taken into account, market will produce too little (if positive externality) or too much (if negative externality).
26
describe some benefits of perfectly competitive markets
1. Efficiency (technical and allocative) due to profit/sustainability motive.
2. Competition - improved quality in response to consumers
3. More choice - providers may be more responsive to consumer needs
4. Funding for innovation more readily available
5. Less impact caused by government changes
27
when does market failure occur?
when marginal social benefit ≠ marginal social cost ≠ price
28
list 3 causes of market failure.
| what do they all have in common?
1. doctors as agents
2. failures in voluntary insurance (public or private) - moral hazard / adverse selection
3. externalities
they're all related to information asymmetry!
i.e. either consumers or suppliers have more information
29
explain how "doctors as agents" causes market failure
when Drs decide what a patient needs, they become both demanders and suppliers of health care.
this can lead to supplied induced demand (SID) - drs may encourage pts to consume services they wouldn't have consumed if they had the same info.
can be linked to incentives (provider moral hazard).
30
define moral hazard
change in the attitude of the 'insured' people and health care providers because of the use of either private or public insurance - there is potential for "excess demand"
31
define consumer moral hazard
0 or low price of treatment to the individual makes people less active in preventing poor health and encourages higher usage
32
define provider moral hazard
fee-for-service can generate incentives to over- or under-provide certain aspects of care - drs are paid a fee for items of service provided, so there's an incentive to over treat (SID!)
or - drs may not be aware/interested in costs so may over-provide e.g. order more diagnostic tests
33
what can insurance companies do to reduce moral hazard?
- co-insurance: individual shares the insured loss with the insurer by paying a % of the loss
- deductibles/excess amount that the individual pays when a claim is made, regardless of the size of the claim
- no-claims bonuses payments made by insurer to discourage claims
34
what can government funded insurance health systems do to reduce moral hazard?
- use primary care as a gateway to secondary services
- waiting lists
- copayments/user charges (e.g. prescription fee)
- medical savings accounts where tax contributions are set aside for medical expenses
35
list some options for dealing with provider moral hazard
- reward for good practice
- salaried service
- financial limitations - fixed budgets
- government regulation e.g. limited lists of drugs that can be prescribed, clinical guidance
36
explain the process of adverse selection
- insurance companies don't know about individual risk, so premiums are set at an average community level
- with 'community rating', low risk people are subsidising high risk people
- info asymmetry: consumers know more of their risk status than insurance company, so will be aware if they are getting a bad deal (e.g. very healthy in an unhealthy population)
- this leads to low risk people opting out, which increases the average risk level, increasing premiums
- this repeats itself as now those at lower risk still paying for insurance become aware that they're overpaying - cycle continues
- results in low risk people being uninsured
this is market failure! people who would want to buy insurance at a profitable price don't as the insurance companies can't identify them.
37
give some options to deal with adverse selection
- premiums to reflect risk profile (but - high risk often poor/old/sick and can't afford without govt subsidies)
- govt intervention:
- compulsory insurance purchase, companies must provide affordable insurance
- progressive tax-based system with exemptions
38
explain externalities
spillovers from A's consumption/production of goods that impact on B - but A doesn't take B into consideration in their decisions!
e.g. vaccination - positive impact on others if you get vaccinated (herd immunity), but people often don't take this into account
39
explain the processing of 'skimming' in the context of healthcare insurance
when insurance companies become aware that there is a group of low risk individuals in a community, they may tailor premiums to individual risk (experience rating rather than community rating).
high risk groups will end up uninsured as they can't afford their really expensive premiums!
but low-risk people have nice low premiums, and insurance company doesn't have to pay out often so they're happy.
this ISN'T MARKET FAILURE - I can't afford to buy a rolex, but that doesn't mean the rolex market is broken!
40
what is an economic evaluation?
compares costs and consequences of two (or more) health care interventions
41
why do we need economic evaluation in healthcare?
because the markets don't provide efficient solutions.
need to intervene, making decisions about what to fund given the scarce resources available.
overall aim is to maximise benefits given the resources available.
42
list the 5 types of economic evaluation
1. cost-minimization (CMA)
2. cost-consequence (CCA)
3. cost-effectiveness (CEA)
4. cost-utility (CUA)
5. cost-benefit (CBA)
43
which type of economic evaluation uses QALYs/DALYs?
CUA
44
which type of economic evaluation expresses both costs and consequences in monetary terms?
CBA
45
which type of economic evaluation produces a list of individual consequences?
CCA
46
what is a CMA and when might it be used?
compares only the costs of different interventions, without giving any information on the consequences.
used when outcomes are known/assumed to be the same.
subset of CEA.
47
which types of economic evaluation assess technical efficiency?
CMA, CCA, CEA
| CUA (does both technical and allocative)
48
which types of economic evaluation assess allocative efficiency?
CUA and CBA
49
what is the most frequently used type of economic evaluation? what is this type's one limitation?
CEA - limited in that it only focuses on a single outcome, so can't compare across programmes that affect different outcomes
50
what two types of ratio can be calculated from any economic evaluation?
ACERs - average cost effectiveness ratios
ICERs - incremental cost effectiveness ratios
51
how do you calculate average cost effectiveness ratios (ACERs)?
total costs of the intervention / total effects of the intervention
52
how do you calculate incremental cost effectiveness ratios?
total costs intervention A / total effects intervention A
minus
total costs intervention B / total effects intervention B
53
what is the basic task of an economic evaluation?
identify, measure, value and compare the costs and consequences of the alternatives under consideration
54
explain the difference between marginal and incremental
marginal - very small or one unit of change. not realistic in context of healthcare decisions as they work in larger units!
so we use incremental - the added cost/benefit of choosing one option over the other
55
how are outcomes measured in a CEA?
natural units or patient/clinician reported outcomes
e.g. life years, cases averted.
can only be used across diseases/interventions where there is a common outcome e.g. kidney transplantation vs heart surgery IF outcome was life years saved
56
how are outcomes measured in a CUA?
QALYs or DALYs
57
how are outcomes measured in a CBA?
monetary
58
how are outcomes measured in a CMA?
outcomes are known to be equal in value for alternatives
59
how are outcomes measured in a CCA?
natural unit disaggregated outcomes - health and non-health outcomes reported individually and separately from the costs - results are not combined into a single unit.
60
for a CEA, what makes a good measure of benefits/consequences?
- related to health outcomes
- has interval properties (with respect to health outcomes)
- meaningful to decision makers
61
explain what interval properties are
the effect of a given change in the measure (e.g. -5mmHG BP) produces the same change in health regardless of pts baseline level of BP
this is more informative than if the health effect of a 5mmHG drop in BP are different depending on baseline (which they probs are?)
62
give some advantages of natural units as used in CEA
- relatively simple (no valuation as in CBA/CUA)
| - readily understood by clinicians and decision makers
63
give some disadvantages of natural units as used in CEA
- uni-dimensional measure of outcome - might miss out on important outcomes!
- comparisons can be difficult if outcomes are different
- doesn't include value (i.e. what would be preferred because it has a bigger impact on wellbeing)
- limited use for allocative efficiency questions
64
explain the use of measures in CEA
e.g. SF-36
useful if we think CEA may ignore some important outcomes, use a set of questions that evaluate a variety of health outcomes into a single outcome
65
give some advantages of self-reported measures in CEA
- based on info reported by the patient
- can be multi-dimensional (incl different aspects of QoL)
- widely used in clinical trials anyway
- there are lots out there, already validated, incl for tricky things like behaviour change!
66
give some disadvantages of self-reported measures in CEA
- difficult to assess overall change (what if you improve in some areas be deteriorate in others?)
- difficult to compare if different measures have been used
- changes may not be informative to decision-makers
- do not have interval properties
- does not include value
- typically do not incl outcomes related to QoL
- limited use for allocative efficiency questions
67
what two parts make up QALYs and DALYs?
1. quality/disability adjustment - reflects morbidity/QoL
| 2. length of life - mortality/quantity of life
68
how to the 0-1 scales work for QALYs and for DALYs
QALYs - 1 = perfect health, 0 = death or equivalent, <0 = worse than death
DALYs - 1 = death or equivalent, 0 = perfect health
(roughly) QALYs gained = DALYs averted
69
explain the difference in how QALYs and DALYs are calculated
QALYs are calculated using generic or condition specific measures completed by the patients, which are then weighted to reflect value on the 0-1 scale. or might be direct valuation on that scale by pts, valuation of vignettes, expert values.
DALYs were developed to assess global burden of disease. disability weights are specific to different conditions.
70
how do you calculate QALYs and QALY gain?
utility value (0-1) of health state x current life expectancy
QALY gain will be the difference in QALYs with vs without the intervention
71
how do you calculate DALYs and DALY gain?
disability weight x life expectancy at that weight
DALY gain will be the difference in DALYs with vs without the intervention
72
give some advantages of CUA
- single measure of outcome
- measures quality and length of life
- incorporate preferences/values
- utility values so they have interval properties
- can be used to compare and allocate resources across disease areas
- easily incorporated into economic modelling
- can address allocative efficiency if a threshold is known
73
give some disadvantages of CUA
- difficulties in deriving health state preferences
- may not be sensitive to changes due to treatment
- may be difficult to link to intermediate public health outcomes
- limited to HEALTH benefits
74
give a basic overview of CMA
- it's a comparison based on incremental cost only
- the outcomes of the alternative treatments being compared are equivalent
- might have discovered this equivalence through a study, or known in advance (e.g. branded v generic drug)
- concerned with technical efficiency
75
give a brief overview of cost consequence analysis
- only really a partial analysis
- costs and benefits of alternative programmes are estimated and listed
- no attempt at aggregating these into cost effectiveness ratios or utility benefit
76
when are CCAs useful?
for detailed breakdown of costs and consequences, but less useful for contributing towards actual decision making.
generally an add on! either to another econ evaluation or just on the end of a study
77
give some advantages of CCA
- potential to include all health and non-health benefits (incl those with no values attached)
- meaning for stakeholders
78
give some disadvantages of CCA
- no overall outcome measure
- can't inform on either technical or allocative efficiency
- risk of cherry picking!
79
for the different types of economic evaluation, how do we determine that an intervention is cost effective?
CBA - net benefit
CMA - cost savings
CEA/CUA - generate more benefits and cost less. can use ICER if there's a threshold in place.
80
how does a CBA work?
takes measured outcomes for health and non-health things (e.g. life years gained, ability to go to work outcomes), plus costs like resource use (e.g. nurses, drugs etc) and converts them all into monetary terms.
can then identify the net social benefit of the program - i.e. the difference between incremental benefits and incremental costs
if positive, new intervention is cost effective.
technical AND allocative efficiency.
81
give 3 methods of valuing health and non-health outcomes in monetary terms (for a CBA)
1. human capital approach - value of paid work missed
2. revealed preferences - values implied through our behaviour e.g. what is the premium paid for risky jobs? this tells us how much we value saving a life!
3. stated preferences - values based on willingness to pay around hypothetical questions e.g. how much are you willing to pay for a flue vaccine to avoid flu symptoms?
82
give some issues with using money as an outcome (in CBA)
- it's difficult to value all health outcomes
- market wage may not be a good reflection of productivity
- what about unpaid work, gender pay gap etc
- median income in UK varies with age and gender - do we value loss of life in 22yo female LESS than a 55yo man??
83
give some advantages of CBA
- addresses both allocative and technical efficiency
- multi-dimensional - can incl all health outcomes and non-health benefits - means it's particularly useful for some public health interventions!
- incl values or preferences for outcomes
- allows comparison across different programmes and sectors (because outcomes are monetary!) - again, good for public health
84
give some disadvantages of CBA
- controversial to give monetary values to health/nonhealth benefits
- issues with relationship between wages, age and gender
- some stakeholders consider it 'unacceptable'
85
what 3 things are needed to generate utility values?
1. description of what is being valued
2. a method that allows us to value on a 0-1 utility scale (for QALYs/DALYs)
3. a group of people to do the valuation
86
where can descriptions of health states come from?
what two forms can they take?
patients
experts (clinicians, carers)
a combo!
summarised in vignettes or standardised questionnaires.
NB - can just ask patients to value their health state without providing a description!
87
what are vignettes, and what are their pros and cons in valuing health states?
little paragraphs that patients/experts have written describing their health state.
pros - more relevant to the condition than a questionnaire, can be linked to study specific effects (e.g. rare side effects).
cons - time consuming/expensive, lack of comparability between studies
88
what are preference based measures?
standardised questionnaires that have already been valued.
| may be generic or condition specific.
89
list some valuation methods
- visual analogue scale
- time trade off
- standard gamble
- discrete choice
- person trade off
- willingness to pay (problematic in context of health!)
90
how does the visual analogue scale method of valuing health states work?
- ask people to rate how good or bad a health state/description is, normally from 0-100
- might be valuing their own health state, or a few different health states based on vignettes etc
- if they value a few different ones then can get them to place health states on a line to reflect relative difference in health state value
NB - sometimes you have to convert VAS values to utility using power functions
91
how does the time trade off (TTO) method of health state valuation work?
- people choose between two options: full length of life (or shorter, set time period) in the health state vs a shorter period in full health.
- you then vary the period in full health until the person choosing is indifferent between full health for X years vs health state for Y years
e.g. 10 yrs in health state = full health for 5.5. yrs
so each yr in health state = 5.5/10 = 0.55 of a full health year
92
briefly explain the standard gamble method of valuation
ask people to choose between living in health state vs uncertain option with <100% chance of being restored to full health or dying immediately
e.g. living with health state vs option w/90% of restoring full health but 10% chance of immediate death
93
will utility values generated from each different method of valuation be the same?
generally no!
- different techniques ask different types of question, which influence the response
- biases in each technique influences values
- variation in same technique can also give different values!
94
what are some reasons for using the general public for valuation of health states?
- veil of ignorance (might mean they rate things worse!)
- no strategic behaviour
- tax payers
- pts might not be able to provide values (elderly, ill, kids)
NICE recommended! DALY's use general pop.
public values lower than pt for physical health, but higher for mental health
95
what are some reasons for using patients to value health states?
- better understanding of their condition and their own well-being
- give pts a voice
96
what are some reasons for using experts to value health states?
- if pts can't do it (elderly, ill, kids)
| - may have better understanding than general population
97
why might their be a difference in health state valuation results if you asked general population vs the public?
- different descriptions: vignettes/standard measures vs actual patient experience
- general pop may focus on different aspects to pts
- different measuring sticks (pts may assess their symptom severity differently from general pop)
- adaptation - pts may have adapted to their health state, which general pop usually don't take into account
98
what are generic preference-based measures?
questionnaires completed by pts - standardised description of health across different dimensions with levels to distinguish severity.
a tariff is then used to score the measure based on preferences given by a general population sample usually via TTO.
i.e. pt does questionnaire, the score from this is then converted to a utility value using a pre-set tariff
99
give some causes of different utility value results from using different generic PBMs for the same health state.
i.e. used SF-6D and EQ-5D-3L on same pts, but got different utility values - why?
- different dimensions (e.g. visual function not covered by EQ-5D)
- different sensitivity (e.g. 3 point vs 5 point scale)
- different severity range
- might have used different valuation methods
- variations in how each method is applied
- different valuation samples
100
give some advantages of using generic PBMs
- cheap and easy!
- allows comparison across conditions
- accepted by NICE
101
give some disadvantages of using generic PBMs
- may not be relevant or sensitivity to condition or treatment effects
- different measures produce different values
102
what are condition specific preference based measures (CS PBMs?
basically people have taken condition specific questionnaires and adapted them for econ evaluation.
reduced the no. questions and then used general population to generate a tariff.
then used the same as generic PBMs but for only one condition!
103
what is costing?
the act of measuring and valuing resources.
NB - true cost of those resources is their opportunity cost!
104
what are the three phases of costing?
1. identify relevant resource use
2. measures and record resource use
3. estimate value of resources - assign a unit cost/price
105
what are the three different perspectives that might be used for costing?
1. healthcare perspective
2. government perspective
3. society perspective
106
what costs will be included when costing from a healthcare perspective?
healthcare costs e.g. drug, hospital costs, tests etc.
| will also incl costs arising from consequences of treatment (side effects etc)
107
what costs will be included when costing from a government perspective?
healthcare costs and costs of other publicly provided services (e.g. social services)
108
what costs will be included when costing from a society perspective?
healthcare costs, costs of other public sector services, and direct costs to pts (e.g. transport) plus production loss from time off work
109
give some examples of primary data sources that might be used in costing
- trials (but population might not be representative)
- routine databases (usually no outcome data)
- clinical databases (for specific conditions)
- medical records
- ask experts
110
give some examples of secondary data sources that might be used in costing
- published studies
- do a systematic review
- useful when econ evaluation being done with a model rather than alongside a trial
- challenge of finding representative studies
111
are average or marginal costs used in costing?
- mostly average costs as an approximation for small changes in costs
- marginal costs may be more relevant locally
112
define fixed and variable costs
fixed = costs don't vary with the quantity of output
variable = costs that do vary with the level of output
113
define discounting
a means of calculating the present value of costs and benefits which arise in the future
114
how do you calculate present value of a future cost?
= future cost / ((1+discount rate)^t)
t = no. yrs in future
discount rate = 3.5% in the UK
115
define positive time preference
we value things that happen in the future less than we value them now - we discount things that are going to happen in the future
116
what are discount factors?
make calculating present values easier.
set numbers for a given discount rate and year, and you can just multiply the cost by the factor to get the present value
117
do we discount using the same rates in the short and long term?
no! if you used a 3.5% discount rate in the very long term, costs and benefits are really low.
UK treasury has different rates for however many years into the future.
118
briefly explain why we discount costs
1. opportunity cost to spending now - we prefer to defer costs, and we can use this money for something that benefits now
2. we should be richer in the future (based on past trends)
3. future is uncertain - why pay £100 now to benefit in 5 years time if there's a chance you'll get the benefit anyway, without paying
4. we're short-sighted! we just don't want to think about unpleasant things in the future (although govt shouldn't run like this!)
119
briefly explain why we discount benefits
1. human behaviour prefers benefits now
2. we can use and enjoy health benefits now
3. uncertainty - might not believe the benefits will materialise in future (at individual level, at societal level we know that someone will get the benefit)
4. we care more about people alive now than those not born yet (but govt should consider future gens.)
5. we're short sighted
6. inconsistent to discount costs but not benefits
120
what are purchasing power parities?
rates of currency conversion that equalize the purchasing power of different currencies
121
give a simplified formula for ICER
difference in cost/difference in QALYs
122
what is dominance?
- simple decision rule for determining which intervention is cost effective (when comparing two) or which should be eliminated from consideration (for 2+ interventions)
- one intervention dominates the other if it is MORE EFFECTIVE and LESS COSTLY
123
explain the cost effectiveness plane
4 quadrants.
incremental cost goes up and down vertically, incremental effect horizontal.
bottom right quadrant = new treatment dominates (more effective, less cost)
top left quadrant = existing treatment dominates as new treatment is less effective and more expensive
124
how do we decide which intervention is cost effective when there's no dominance?
use thresholds or QALY league tables
125
how do you use thresholds to assess cost effectiveness?
compare ICER to decision maker's threshold - if it's lower, then yay, fund it!
e.g. NICE £20-30k
126
how do QALY league tables work?
- interventions are ranked by incremental cost per QALY (lowest at top)
- rough idea is you fund from the top down
127
give some limitations of QALY league tables
- differences between studies - discount rates, methods for utility estimates, costing perspective etc etc
128
what are the three main types of uncertainty in economic evaluation?
methodological, parameter and structural
129
what is methodological uncertainty?
issues with which modelling method is best, how do we measure and value utilities, which resources have been costed, which discount rate
130
what is parameter uncertainty?
uncertainty surrounding the true mean parameter values.
how does the data we have relate to the population we are interested in?
131
what is structural uncertainty?
thinking about whether we have the correct comparators, does the model include all relevant events, how are those events modelled?
132
what three main sources are there for uncertainty?
1. assumptions and methods used
2. data - sampled vs unsampled
3. generalizability of the findings to other settings
133
what are the two key methods of dealing with uncertainty?
use of a reference case (tackles methodological uncertainty)
sensitivity analysis
134
what is sensitivity analysis?
a type of analysis done to test the impact that different choices within an economic evaluation have had on the results
135
what is the NICE reference case?
basically a statement of what methods should be used for public health economic evaluations.
helps tackle methodological heterogeneity - if all the economic evaluations are carried out in the same way, even if there is uncertainty the results should still be comparable
136
describe one-way sensitivity analysis
you vary one variable.
e.g. make a Tornado diagram by changing values one at a time and keeping everything else constant
OR threshold analysis - vary one variable to see when A becomes more cost effective than B
e.g. above what price is a new drug no longer cost effective (using ICER threshold £20k/QALU)
137
is one-way sensitivity analysis likely to over or underestimate uncertainty and why?
underestimate - doesn't take into account interactions between variables.
this is where multi-way sensitivity analysis comes in
138
what is scenario analysis?
type of multi-way analysis.
assumptions specified for a no. variables e.g. extreme case scenario, reference case scenario
139
how does extreme scenario analysis work?
1. set each variable to most optimistic value for the new intervention - best case results
2. then set them all to the worst case
likely to overstate uncertainty, likely to be unrealistic.
140
what is probabilistic sensitivity analysis?
- samples all uncertain parameters from their distribution and re-calculates results
- does this a load of times then calculates mean and uncertainty
- addresses parameter uncertainty
141
give some advantages of simple sensitivity analysis
- easy to implement
- might let you identify key factors of the decision problem which determine the cost-effectiveness of the technology
- can be used as a means of validating/verifying the underlying logic model
142
give some disadvantages of simple sensitivity analysis
- extreme scenarios not that helpful for decision makers
- multi-way scenario analysis can be complicated to present
- all parameters are uncertain so varying a selection of these doesn't account for all uncertainty
143
when is economic modelling useful?
- when trials don't give a clear decision, or some data is missing
- need to extrapolate beyond time period of trials
- to link intermediate trial end-points to final outcomes (e.g. trial outcome of cancer detection link to QALYs/years lived)
- generalising results to other settings
144
give three common types of models
1. decision trees - basically a flow diagram
2. Markov models
3. Discrete Event Simulation (DES)
145
what are Markov models?
- follow a group of pts over a no. time periods
- each pt assigned to a health state at each period of time, with a set likelihood of moving to a different health state in the next time period
146
what is discrete event simulation (DES)?
- patient-level simulation approach which models specific events experienced by individual patients
- flexible, event-oriented approach that follows the experience of entities over time
147
give some limitations of decision trees
- patients only move through the tree once, can't go between health states - what about chronic conditions, cancer relapse/remission?
- no temporal element - won't be able to discount etc
148
how do Markov models work?
- incl all possible health states relating to disease
- each time you run the cycle pts will either transition between states or remain in current state - can only be in one state at a time
- rates of moving between states governed by transition probabilities, which are set in the model
149
give some limitations of Markov models
- 'memoryless' - being in the 'sick' state once doesn't affect the chance of the pt being in it again - is this realistic?
- hard to determine what proportion will have had the event of interest at each time point
- application of underlying risks over time can be difficult
150
what are cost-effectiveness acceptability curvers (CEACs)?
- means of quantifying and representing uncertainty
- show probability that intervention is cost-effective over a range of cost-effectiveness thresholds
- decision still made on mean cost per QALY, but CEACs help interpret uncertainty
151
give some reasons why transferability of economic evaluations is an issue
- different countries have different treatment patterns, genetic pools, adherence, costs, values of health states, discount rates, funding thresholds etc etc etc
- can adjust for some of these, but some it's harder to adjust for!
152
what should be considered when transferring costs from one context to another?
- resources used/unit cost - currency conversion, differences in local costs
- time - inflation/discounting
- perspective of study
153
what should be considered when transferring outcomes from one context to another?
- how they've measured/valued health outcomes
| - effect in terms of absolute risk and relative risk reductions
154
how can you tell if it's appropriate to transfer?
use transferability checklists!
there's loads
155
give some different approaches for transferring results from one context to another
- simple adaptation - just adjusting unit costs, discount rates etc, if everything else matches up
- analyse individual patient data - if econ eval info is available and relevant but not comparable, and you have individual level data
- in ALL other cases - consider model adaptation (decision analytical modelling)
156
what are the different aspects to be considered in the financing of health care systems?
1. collecting revenue
2. pooling resources
3. purchasing goods and services
4. providing services
may be private or public or a mix
157
describe some different methods of collecting revenue in a health care system
1. general taxation - income tax, VAT etc
2. payroll tax - specific tax taken from each payslip (or earmarked section of income tax) for healthcare - often called social health insurance contributions
3. payment at time of use - out of pocket
- compulsory (taxation) vs voluntary pre-paid (private health insurance)
- system may have exemptions (e.g. elderly) that don't have to contribute
158
what is pooling funds in the context of health care systems financing?
any system may have some type of pooling funds to protect individuals from risk of large healthcare bills - i.e. insurance
159
who holds pooled funds for health care? how are these funds managed?
- may be national health ministries, local govts, social health insurance funds, private profit/non-profit insurance funds, community NGOs
- may be one or more pooled funds managed together or separately.
- may be independent from govt
- should be accountable and transparent?
160
how does purchasing services work in different healthcare systems?
- this is the 'buying' of service from providers
- could be public or private providers
- different suppliers could be selectively contracted
- can use the way providers are paid to incentivise e.g. to treat more, be cost-conscious, quality of care
- could be linked to needs of population/performance of providers
161
give some different primary care purchasing options
1. fee for service (FFS) - (small) risk of induced demand
2. capitation - set amount for each patient on registration lists
3. salary GPs
162
give some different hospital purchasing options
1. FFS
2. Budgets agreed in advance - but what about quality?
3. global budget based on historical budget and population needs
4. prospective cost per case - use diagnostic related groups
163
give some pros and cons of capitation/prospective payment
good - can be used to control costs as amount paid is fixed
bad - reduced effort (getting paid anyway), cherry pick easy cases, over refer to other services
164
how can you manage the costs of healthcare systems
1. use primary care as gatekeeper
| 2. integrated approach where purchaser is the provider
165
explain how an integrated approach to healthcare provision works
- saves on costs of selecting providers, drawing up and enforcing contracts
- this is how the NHS worked prior to the internal market
166
list the 4 main types of healthcare system
1. private - out of pocket payments and voluntary private insurance
2. social health insurance - payroll taxation
3. national health service - general taxation
4. community health insurance funded by voluntary pre-paid contributions managed by community groups (less common)
most are a bit of 1-3
167
explain how market based (private) healthcare systems work
- insurance is funded by voluntary pre-paid contributions, or healthcare costs paid out of pocket
- consumer or employer pays premium or pays directly for services
- premium might be tax deductible
e.g. USA
- price at point of use may be 0 of co-payments/deductibles
- private ownership of health sector services
- no universal coverage - at risk groups excluded and need government cover
168
give some benefits of a market based healthcare system
- efficiency due to profit motive
- more choice - health care providers may be more responsive to health needs
- improved quality in response to consumers
- funding for innovation more readily available
- less impact from govt changes (in theory)
169
give some limitations of a market based healthcare system
- not efficient due to high costs of admin etc
- monopoly providers won't need to be responsive to consumers
- market failure due to asymmetric info
- all leads to distorted markets which can't signal which interventions are efficient
- equity issues
- hard to fund public health interventions
- competition may lead to fragmentation and poor care, or even poor quality (cost cutting to stay in market)
170
explain how a social health insurance healthcare system works
- compulsory contributions if in formal work - fees taken from payroll tax
- may have extra funds from general taxation / allow private insurance
e.g. germany, france, netherlands, japan, korea
- managed by non-profits with some independence from govt
- public and/or private ownership of services
- aims for universal coverage with varied success
- unemployed given access to other subsidised schemes
171
give some strengths of social health insurance
- higher levels of funding than general tax based system
- direct relationship between payment and benefits can act as incentive to consumers to pay into the system
- can also build in choice of insurer and provider
- independence from govt
- equity can be reflected based on progressive contributions
172
give some limitations of social health insurance
- universal coverage not possible without govt intervention
- problematic in context w/large informal work sector
- may be no competition and associated benefits
- might not achieve efficiency
- can lead to distortions in the labour market
173
explain how a govt funded NHS works
- funded via general taxation, with only nominal fees at point of use
- may be private insurance available as well
e. g. UK, canada, finland, NZ, australia
- aim for universal coverage on most care with no excl for citizens
174
give some strengths of govt funded health systems
- universal coverage - no excl based on pay or health
- contributions part of general taxation - progressive and affordable
- can have lower transaction costs if single system in place for collecting funds and paying out services
- integrated care and info sharing to improve care
- transparency and accountability
- easier to manage large public health initiatives
175
give some limitations of govt funded health systems
- no competition - no advantages in efficiency/choice/quality
- can be large costs associated with building these into the system
- subject to reform based on who is in power
- sufficient funding problems - long waiting times, no access to expensive technologies, low wages
176
explain how community based health insurance systems work
- seen in LMICs
- insurance scheme owned and run by community
- low premiums, regardless of health status
- possible with strong community cohesion
but - low uptake and renewal rates due to affordability and poor understanding of insurance. poor quality of care due to low bargaining power of each group.
177
list the seven economic objectives of healthcare that can be used to judge performance of a healthcare system
1. efficiency
2. equitable provision
3. equitably financed
4. financial protection (avoids households falling into poverty)
5. efficiency of revenue collection
6. cost containment
7. transparency and accountability
178
what is economic efficiency and how would we assess this when judging a healthcare system
- does it use resources to maximise benefits?
- are both technical and allocative efficiency being met?
- what interventions are being used and are they cost effective?
- have they got guidelines with cost effective treatment options - are they being adhered to?
179
what is equity?
= fairness of distribution of benefits and costs
horizontal equity = equal treatment of equals - system should provide in same way for those with equal need.
vertical equity = unequal treatment of unequals - different treatment for differing levels of need.
180
explain the concept of equitable financing of healthcare systems
payment based on ability to pay.
| use of a progressive system so that rich pay a higher proportion of their income.
181
what is financial protection and how would we assess this when judging a healthcare system
protection against health care related expenditure that leads to poverty e.g. uninsured face risk of bankruptcy/loss of livelihood.
assess based on catastrophic health expenditure / impoverishing health expenditure.
can be achieved through use of tax revenues or regulated health care insurance.
182
what is "efficiency of revenue collection" and how would we assess this when judging a healthcare system
- collection of all revenue with minimum admin costs
- assess based on ability of system to collect all revenues, costs of revenue collection and minimisation of fraud.
achieved via using general taxation, penalties for opting out and common revenue pools with common collection process
183
that is "cost containment" and how would we assess this when judging a healthcare system
ensuring healthcare expenditure is controlled while maintaining effectiveness.
- assess using healthcare spending per capita/region/service etc
but make sure you assess effectiveness as well!
184
give some cost containment strategies that might be employed by a healthcare system
- cost sharing to minimise consumer moral hazard
- payment mechanism to minimise provider moral hazard
- gate keeping
- price and non-price competition
- guidelines/packages of approved cost-effective treatment options
- prescription drug agreements
- single template for admin systems to reduce transaction costs
- monitoring and regulation
185
what is transparency/accountability and how would we assess this when judging a healthcare system
- decision making, responsibilities, accountability and outcomes should be transparent and clear
- assess availability of info on responsibilities and decisions, performance data, mechanisms to report failings
