Cost-Benefit Analysis
Jamie C. Barner, PhD
1
Objectives
• Students should be able to:
– Compare and contrast CEA and CBA
– Understand how to monetize healthcare benefits
– Understand how to calculate and interpret cost-
benefit ratios and net values
– Discern biases in methods to monetize healthcare
benefits
– Compare and contrast methods for eliciting
willingness to pay values
2
Cost-Benefit Analysis
Value of resources consumed (costs) from program/intervention
Value of outcome (benefits) from program/intervention
• Costs are measured in dollars
• Benefits are measured in dollars
• With vs. Without
3
Questions CBA Address
• Will the benefits of a program exceed the
cost of implementing it?
– Can assess a single program
• Which program will produce the greatest
net benefit?
– Can examine multiple programs with similar or
unrelated outcomes
• Weakness of CEA
4
Criteria for
Cost-Beneficial Programs
Benefit Cost
>1 OR
<1
Cost Benefit
5
Which one would you choose?
Programs CE B/C
AIDS prevention $230,000/case prevented 8.4
Vaccinations for children $104,000/case prevented 0.3
Diabetes management $140/normoglycemic 15.1
patient
Breast cancer screening $50,000/life year saved 2.4
Medication therapy $450/medication-related 2.5
management problem resolved
6
School Based Screening for TB
• Outbreak of TB in LA,
Houston, Boston, NYC Strategy Group B/C
• Public health officials in All K .31
LA ordered a massive
HS .76
screening of all HS and
K students Both .58
• Very expensive so Target K .97
conducted a CBA to
determine most HS 1.08
efficient method to use Both 1.06
in the future.
JAMA, 1995 7
Cost Benefit Analysis
Medical Non-medical
Costs ($)
Benefits ($)
Direct Benefits $ Indirect Benefits $ Intangible Benefits $
Medical Non-medical
Productivity Patient Preferences
Pain
Suffering
Human Capital
WTP (Willingness to Pay) WTP 8
Cost-Benefit Analysis in the Literature
Costs
Direct Benefits Indirect Benefits Intangible Benefits
Consensus and debate
Costs Direct Indirect Intangible True CBA
Benefits Benefits Benefits (according to
economists)
No, cost
comparison
Yes
Yes
9
Typical CBA studies in the Literature
Least utilized method of all 4 PE methods
30 CBA/yr vs. 1000s CEA/yr
Most are conducted in other countries
Most focus on health care programs or
interventions
Usually do not focus on comparing specific drug
therapies
10
Typical CBA studies in the Literature
• Types
– Preventative strategies (58%)
• vaccination programs (influenza, hepatitis)
– Pharmaceuticals (25%)
• usually vaccines
– Education and counseling (21%)
• depression, diabetes, asthma
– Screening strategies (18%)
• prenatal, cancer, HIV/AIDS, gene therapy
11
Study Examples
• Cost-Benefit Analysis of Vaccination against Tick-Borne
Encephalitis among French Troops
• Cost-Benefit Analysis of a Haemophilus Influenzae
Type B Meningitis Prevention Programme in The
Philippines
• Cost-Benefit Model Comparing Two Alternative
Immunisation Programmes Against Serogroup C
Meningococcal Disease: For Quebec Residents Aged 2
Months to 20 Years
• Bedside Rapid Flu Test and Zanamivir Prescription in
Healthy Working Adults: A Cost-Benefit Analysis 12
Study Examples
• Cost-Benefit Analysis of Active Vaccination
Campaigns Against Hepatitis A Among Daycare
Centre Personnel in Israel
• A Model to Estimate the Cost Benefit of an
Occupational Vaccination Programme for Influenza
with Influvac® in the UK
• Cost-Benefit Analysis of Risperidone and Clozapine in
the Treatment of Schizophrenia in Israel
13
Other Studies in the Literature
• Focus on measuring indirect and intangible costs
– Human Capital (HC)—productivity
– Willingness to Pay (WTP)
14
Study Examples
• HC
– A Review of Health-Related Workplace Productivity Loss
Instruments
– Workplace Productivity: A Review of the Impact of
Migraine and its Treatment
• WTP
– Measuring Treatment Preferences and Willingness to Pay
for Docetaxel in Advanced Ovarian Cancer
– Willingness to Pay for Inhaled Insulin: A Contingent
Valuation Approach
15
16
Journal of Rheumatology 2010: 37:1805-1814
Clinical Therapeutics 2010:32(8) 18
American Journal of Managed Care 2010: 16(8) 19
Health Expectations 2015; 18(6): 2252-2265. 20
21
Steps in Conducting CBA
1. Identify program or intervention
2. Identify alternatives
3. Identify and measure costs
4. Identify and measure benefits
5. Calculate results
22
Cost-Benefit Analysis
Case Study
• HealthSouth is an ambulatory care family practice
clinic located within a 150-bed county hospital.
• The practice serves about 1000 patients with
diabetes yearly.
• A recent inspection by Medicaid officials revealed
that only 20 percent of the patients with diabetes
received the basic level of care as recommended by
the American Diabetes Association (ADA)
guidelines.
• For the organization to continue to receive federal
funding, it had to improve the outcomes of its
patients with diabetes.
23
Cost-Benefit Analysis
Case Study
• Recently, a pharmacist proposed the development
of a diabetes disease state management clinic, but
the organization’s management felt that it would
not be economically feasible.
• In light of the recent inspection, management was
willing to reconsider its position but they needed to
have “hard data” showing how the pharmacist
could provide benefit, to not only the patients, but
also the organization as well.
24
Cost-Benefit Analysis
Case Study
• The medical director approaches you, and she
would like a proposal of what type(s) of services
would be provided, expected outcomes of the
clinic, and the resources needed to run the
clinic.
• The director also has stated that because of
budgetary constraints, she can only support the
clinic if it can show that benefits outweigh the
costs.
25
Step 1. Identify Program
Who would you target and what would you do?
Think/Pair/Share
• Program
– What type of program?
• Target
– Which patients?
• Services
– What services will be offered?
• Outcomes
– What outcomes will be measured?
26
27
http://www.tdctoolkit.org/wp-content/uploads/2014/05/02MinPracticeRec.pdf 28
Step 2: Identify Alternatives
• Alternatives
– No diabetes clinic (“do nothing”)
– Target all diabetes vs. high risk
• 1000 patients vs. 400 patients
– Alternative service (e.g., lipid clinic)
29
Cost Benefit Analysis
Medical Non-medical
Costs ($)
Benefits ($)
Direct Benefits $ Indirect Benefits $ Intangible Benefits $
Medical Non-medical
Productivity Patient Preferences
Pain
Suffering
Human Capital
WTP (Willingness to Pay) WTP 30
Step 3:
Identify and Measure Costs
What are the medical and non-medical costs
with and without the program or intervention?
• Perspective • Direct non-medical
• Direct medical – Patient travel time,
– Pharmacist visits lodging
– Pharmaceuticals – Program
– Laboratory tests implementation
– Physician visits • Non-medical personnel
• Space
– Hospitalizations
• Advertising
– Emergency • Supplies
department visits 31
Calculate the incremental costs (difference between with vs. without).
Table 1. Direct costs with versus without (n=400)
Direct Costs With Without Difference
Medical
Pharmacist 41600 0 41600
Medication (50% ↑)
Oral agents 359640 239760 119880
Insulin 106680 71120 35560
Lab Tests (1/yr-4/yr) 80000 20000 60000
Non-medical
Travel 2400 0 2400
Time 21600 0 21600
Administrative 520 0 520
Space 5000 0 5000
Advertising 200 0 200
Supplies 600 0 600
Total direct costs 287360
$718.40/pt/yr 32
Assumptions
MEDICAL
• RPh $50/hr, 2days/wk, 8hrs/day, 52 weeks
• Medication assume 50% increase
– oral agents $666/yr 90% of patients (n=360)
– insulin $1778/yr 10% of patients (n=40)
• Labs $50/lab
– without—1/yr hba1c
– with– 4/yr hba1c q6mos, lipid, kidney fxn
NON-MEDICAL
• Travel (patients)
– .30cents/mile X 2 visits @ 10miles
• Time (patients)
– 2 visits/yr travel +wait time=1.5 hrs/visit-- $18/hr
• Administrative
– $10/hr—1 hr/wk
33
Step 4:
Identify and Measure Benefits
• Direct
• Indirect
• Intangible
34
Cost Benefit Analysis
Medical Non-medical
Costs ($)
Benefits ($)
Direct Benefits $ Indirect Benefits $ Intangible Benefits $
Medical Non-medical
Productivity Patient Preferences
Pain
Suffering
Human Capital
WTP WTP 35
Benefits
Direct Indirect Intangible Benefits
Direct
• “Potential savings” in the avoided use of
health services
– Pharmacist visits
– Pharmaceuticals (DUR vs. Adherence programs)
– Laboratory tests
– Physician visits
– Hospitalizations
– Emergency department visits
36
Calculate the cost savings or direct benefit (difference between with vs. without).
Table 2. Direct medical benefits with vs. without (n=400)
Medical Service With Without Difference
Physician visits 192000 192000 0
ER visits 40200 107200 67000
Hospitalizations 155100 413600 258500
Total 387300 712800 325500
Total direct medical benefit 325500
Societal (includes patient cost: Organizational [excludes patient
travel & time) cost:
B/C = B/C =
37
Assumptions
• MD
– $160/visit 3 visits/yr
• ER
– $670/ER visit
• Without 20% of 400=80 patients with 2 visits/yr
• With 60 patients with 1 visit/yr
• Hospitalization
– $2585/hospital stay
• same as ER
38
Cost Benefit Analysis
Medical Non-medical
Costs ($)
Benefits ($)
Direct Benefits $ Indirect Benefits $ Intangible Benefits $
Medical Non-medical
Productivity Patient Preferences
Pain
Suffering
Human Capital
WTP WTP 39
Indirect Benefits
Methods to assess
1. Practitioners’ or policymakers’ views on what is
appropriate
2. Court awards
3. Individual expenditures on life and health
insurance
4. Market valuation (human capital) wage and
productivity losses due to illness, disability or
premature death
40
Human Capital
• Value of health benefits=the economic
productivity they permit
– Cost of disease=lost productivity
• Cost of a sick day=how much you earn that day
41
Human Capital
• Use discounted values of expected earnings
– Census estimates
• age, gender, education
– Gather data from individuals
– Labor income is estimated as before-tax income
• Non-labor income is excluded (interest, etc.)
– Use market values for non-market activities
• unpaid household work, child care, etc.
42
U.S. Census Bureau Income Estimates
Think/
Males Females
21-24 yrs $21,393 $19,108
< High school $17,170 $14,211
High school graduate
Some college
$21,078
$22,088 Pair/
$17,054
$19,104
Share
Bachelor degree $30,258 $25,881
Advanced degree $30,618 $25,998
25-34 yrs $32,085 $26,788 • What is your
< High school $21,470 $16,068
High school graduate $28,262 $21,083 human
Some college
Bachelor degree
$32,063
$41,854
$25,292
$32,860
capital?
Advanced degree $50,888 $38,530 • What about
45-54 yrs $44,836 $30,960 your
neighbor?
< High school $26,687 $17,671
High school graduate $36,634 $24,429
Some college
Bachelor degree
$42,852
$57,663
$30,534
$39,939
• What issues
Advanced degree $71,239 $48,652 can you
65-74 yrs $32,718 $23,261 identify
< High school
High school graduate
$20,450
$26,396
$15,887
$21,101
with this
Some college $32,524 $25,492 method?
Bachelor degree $49,109 $32,040
Advanced degree $65,264 $41,850 43
Human Capital
• Problems
– Biased against specific groups
– Earnings may not equal the value of outputs
• Football player vs. Teacher
– Does not include values for pain and suffering if the
disease state or condition does not impact
productivity
• E.g., Menopause, Impotence vs. Diabetes, Cancer
44
Calculating Productivity
Using Human Capital Approach
• Wage rate
• Yearly (income/year)
• Daily (income/ # days worked per year)
• Number of days worked/yr range 240-365
• 365-vacation (14)- weekends (104)- sick days (7)=240
• Example
• Daily wage rate X # missed days
45
Calculating Productivity
Missed Days
• Missed Work/Housekeeping Days
• Daily wage rate X no. of missed days
• Restricted Activity Days
• Daily wage rate X % time restricted/day
• Caregiver Time
• School absenteeism (children 5-17 yrs)
• Daily wage rate X no. of missed days from school
• Bed days (children 0-4)
• Daily wage rate X no. of bed days 46
Calculate Productivity-Missed Work
Average No. of Days Daily Wage No. Missed Value of
Income Worked/yr Rate Days/yr Productivity
W/O $36,000 / 240 = _____ X 10 = $_____ X 400 = _______
W $36,000 / 240 = _____ X 5 = $_____ X 400 = _______
Indirect Benefit per person $
Indirect Benefit (N=400) $______
Table 3. Indirect Benefits with vs. without (n=400)
Indirect Cost Without With Difference
Total 600000 300000 300000
Total Indirect Benefit 300000
Societal (includes patient cost: travel & time)
B/C = [_______ (Direct) + _________(Indirect)]/_________(cost) = ____ 47
Cost Benefit Analysis
Medical Non-medical
Costs ($)
Benefits ($)
Direct Benefits $ Indirect Benefits $ Intangible Benefits $
Medical Non-medical
Productivity Patient Preferences
Pain
Suffering
Human Capital
WTP WTP 48
Intangible benefits
Benefits
Direct Indirect Intangible
• Intangible costs/ benefits
• Costs of pain, suffering, anxiety or fatigue that
occur because of an illness
• Intangible benefits are the “cost-savings” that result
from a reduction in pain and suffering
49
Intangible benefits
• Valuation of goods/services are easier for marketed
vs. non-marketed goods/services
– Health care vs. coffee or pair of jeans
• Valuation of goods/services are based on:
– Need
• e.g., health care (pain/suffering, productivity, etc.)
– Resources
– Preference
50
Willingness to Pay
• Determines how much people are willing to
pay to reduce the chance of an adverse
health outcome.
– Example: If a person was willing to pay $50 for
a ½ hour visit with a pharmacist to improve
their diabetes condition, then the intangible
benefit/person/visit would be $50.
51
Willingness to Pay
Contingent Valuation/
Hypothetical Market
• Respondents are being asked to
value a “contingent market”
• Does the respondent understand the
“market” being presented?
52
Hypothetical Market-Diabetes DSM
• Patients with diabetes have improved their condition by
learning more about their disease and by taking their
medications as directed. Pharmacists can help people with
diabetes understand their condition and the medications used
to treat it. In addition they can:
– Help you learn how to use a glucose meter
– Help you with dietary management techniques such as carbohydrate
counting
– Help you recognize and handle situations when diabetic crises occur
– Check your feet for signs of nerve damage
– Check your blood sugar and blood pressure
– Monitor your diabetes by keeping a record on file and following up
with you on a regular basis to assess your progress
– Contact your doctor and report any changes in your health
53
Willingness to Pay
Obtaining Bids
A. Open-Ended
B. Closed Ended/Take It or Leave It
C. Bidding Game
D. Payment Card
54
55
56
57
58
Payment Card Example
• Assume you had to pay for the initial assessment and
educational appointment with the pharmacist. This visit
would last approximately 1 hour. Your insurance would not
cover any costs. Assume that this consultation would result in
a 50% improvement in your diabetes. Below is a list of
possible fees for the initial consultation visit.
– What is the maximum amount that you would be willing to pay for a 1-
hour one-one consultation visit resulting in a 50% improvement in
your diabetes? Please circle your choice.
$150 $90 $30
$130 $70 $10
$110 $50 $0
59
Willingness to Pay
• Problems
– What people say vs. what they
will really pay
– Can the average person answer
questions
60
Willingness to Pay
• Bias Issues
–Compliance
–Strategic
–Starting Point
–Income
61
Assume that you have an uncontrollable cough that continues to keep you
awake at night. You have tried all forms of Robitussin, as well as cough
formulations containing codeine, and nothing works. A new product has
been developed that will cure your cough 50 percent of the time. This new
medication can be taken in a tablet once a day without any side effects. Your
insurance company will not pay any of the costs.
What is the maximum amount that you would be willing to pay
for a 30 day supply of this medication?
1. $100
2. $80
3. $60
4. $50
5. $40
6. $30
7. $20
8. $10
9. $0
62
Assume that you could hire someone to attend your least favorite class
this semester and take notes for you. This individual is an expert at
taking notes and can guarantee you 100 percent that you will get an A
in the course if you read and study his notes.
What is the maximum amount
per hour that you would be
willing to pay this notetaker?
1. $5.00/hr
2. $8.00/hr
3. $10.00/hr
4. $12.00/hr
5. $15.00/hr
6. $20.00/hr
7. $25.00/hr
8. $30.00/hr
63
Alternate form starting with $50
Assume that you have an uncontrollable cough that continues to keep you
awake at night. You have tried all forms of Robitussin, as well as cough
formulations containing codeine, and nothing works. A new product has
been developed that will cure your cough 50 percent of the time. This new
medication can be taken in a tablet once a day without any side effects. Your
insurance company will not pay any of the costs.
What is the maximum amount that you would be willing to pay
for a 30 day supply of this medication?
1. $50
2. $40
3. $30
4. $20
5. $10
6. $0
64
Notetaker
10
9 8.6
7.91
8
7
5.91
6
Mean
5
$
SD
4
3.2
3
2
1
0
No Range Range 5-30
65
Cough Medicine
40
36.36
35 33.18
30 28.75
25
22
19.09 Mean
20
$
15.49 SD
15 12.86 12.21
10
5
0
No Range Range 0-50 Range 0-100 Classmate 100
Starting Point 66
Bias
Intangible Benefits-Diabetes DSM
• On average, patients were willing to pay
$50 for a 1-hour consultation and $25 for
follow-up visits with the pharmacist.
Table 4. Intangible benefit (n=400)
Service Estimates With Without
Initial consultation 50 20000 0
Follow-up 25 10000 0
Total 30000
Total intangible benefit 30000
67
HC vs. WTP
HC WTP
• Easier to measure • More difficult to
measure
• Only considers • Captures productivity,
productivity (in terms patient preferences
of earnings) (intangibles)
• Biases against specific • Biases may not give
groups accurate responses
68
Indirect /Intangible Benefits
Issues to consider
• What is the impact on productivity?
– Are there potential biases with any group?
• Prostate vs. breast cancer
• Diabetes vs. hypertension
• What is the impact on pain and suffering?
• Cancer, HIV/AIDS, arthritis vs. hypertension, cholesterol
• Will the additional information lead to a more informed
decision or change the results?
• Menopause, infertility, depression
• What resources would be needed to obtain the data?
– Are these costs affordable?
– Can the data be collected in a timely manner? 69
Step 5: Calculate Results of
Costs and Benefits
• Discounting
• Calculation methods (< 1 year)
– Ratio
• Cost/Benefit < 1
• Benefit/Cost > 1
– Net
• Net costs=Total costs-Total benefits (<0)
• Net benefits=Total benefits-Total costs (>0)
– Two methods may not lead to the same choice
70
Example
• Option A
– costs $1000: saves $2000
• Option B
– costs $5000: saves $7500
• Benefit-Cost
– A = $2000/ $1000 = 2:1
– B = $7500/ $5000 = 1.5:1
• Net Benefit (net benefits-net costs)
– A = $2000 - $1000 = $1000
– B = $7500 - $5000 = $2500 71
Calculate B/C and Net Benefit for all and direct benefits
Ratio and Net Calculation
• Using either B/C ratio or net benefit
calculation, is the program cost-beneficial?
Benefits Costs
Direct 325500 Societal 287,360
Indirect 300000
Intangible 30000
TOTAL 655500 All Benefits Direct Benefits
Benefit/Cost 2.28 1.13
Net Benefit 368140 38140
72
Step 5. Calculate Results of
Costs and Benefits
• Additional calculation methods (>1 year)
– Net present value (NPV)
– Internal rate of return (IRR)
73
Net Present Value (NPV)
• Present value of a stream of costs and benefits
• Accept projects with positive NPV
74
NPV Example
• Pharmaceutical care program costs
– $5000 per year for the next three years
– Discount rate is 5 percent
• Present value of costs
– $14,297 = [$5000 (yr 1) + $5000/1.05 (yr 2) + $5000/(1.05)2 (yr 3)]
• Program benefits
– $5,000 (yr 1), $10,000 (yr 2), $15,000 (yr 3)
• Present value of benefits
– $ 28,129 = [ $5000 (yr 1) + $10,000/1.05 (yr 2) + $15,000/(1.05)2 (yr 3)]
• Net (PV benefits – PV costs)
– NPV= $13,832 = [$28,129 - $14,297]
• Ratio (PV benefits/PV costs)
– Ratio NPV = 1.97 = [$28,129/$14,297]
75
Internal rate of return (IRR)
• Rate of return that equates the present value of
benefits with the present value of costs
• Computer programs and special functions on
calculators are available for determining the IRR
– difficult to calculate by hand
• Accept projects that are > “hurdle rate” or discount
rate (e.g., 5%)
– If the IRR is greater than 5% then pursuing the project will
bring a beneficial rate of return
76
IRR Example
• IRR for pharmaceutical care program described
above is 6.3 percent
• Market rate of return (hurdle rate) is 4.5 percent
• 6.3 percent > 4.5 percent
– Accept pharmaceutical care program
• will yield a greater return than investing it on the market
77
Cost Benefit Analysis
• Are you able to…?
– Compare and contrast CEA and CBA
– Monetize healthcare benefits
– Calculate and interpret cost-benefit ratios and net
values
– Compare and contrast methods for eliciting
willingness to pay values
– Detect biases in methods to monetize healthcare
benefits
– Determine when it may be important to
incorporate indirect and/or
78 intangible benefits