Cost Utility Analysis

Cost Utility Analysis (CUA) Define CUA Advantages and disadvantages of CUA Quality-adjusted life years (QALY) Methods for determining utilities (RS, SG, TTO) Steps in utility analysis and estimation of QALYs How much is a QALY worth? Rascati (Chapter 6) 1 Cost-Utility Analysis Quentitity lifegar • Costs are measured in dollars • Outcomes are measured in utility ―The most common outcome unit is called the quality adjusted life year (QALY) ―Other less frequent outcome units →Disability-adjusted life years (DALY) • Number of years lost due to illness, disability, or premature death →Healthy-year equivalents (HYE) • Number of years spent in perfect health 3 Utility • General economics terminology ―Utility is the satisfaction a person obtains from purchasing or consuming a good or service (Short utility video refresher) • Cost-utility analysis (CUA) terminology ―Common terms for utility →patient preferences →preference weight →preference value ―Most common outcome unit for CUA is the QALY →QALY incorporates both the quality of life (utility, morbidity, diseases) and quantity of life (mortality, life-year gained) into clinical outcomes 4 Utility vs. QALY • QALY = quality of life (utility) × quantity of life (length of life) →1 QALY = 1 year of life in a perfect health state (where utility = 1.0) → 0.75 QALY = 1 year of life in a less than perfect health state (where utility = 0.75) → 0.75 QALY = 9 months (or 0.75 years) of life in a perfect health state (where utility = 1.0) → 0 QALY = dead 5 When should CUA be used? 1. When QoL is an important outcome measure of interest in a study ―Hearing loss, seasonal allergies, erectile dysfunction 2. When the outcomes of interest include both morbidity (diseases, complications, side-effects) and mortality (death) 3. When treatments and outcomes that are being compared are different ―Antihypertensives vs. diabetes medication →Natural outcomes for both treatments are different (mmHg vs. A1c) →With CUA, the different outcomes can now be converted into a common unit – the QALY 6 3 Utility Convention • Perfect health utility = 1 Is • Morbidity (any disease state) Eton utility range from >0 to <1 • Mortality (death) utility = 0 7 Methods to estimate utility values • There are two broad methods used to elicit utility ‘weights’ for various health condition or ‘health states’ 215WI method 1. Direct elicitation 1. Rating scale (RS) 2. Standard gamble (SG) 3. Time trade-off (TTO) 2. Indirect elicitation method 1. Use of standardized survey instruments (EQ-5D, SF-6D) →(these will be covered under HRQoL) 9 Steps in utility analysis and the estimation of QALY 1. Describe the health state or health condition (e.g., hospitalbased kidney dialysis) to the respondents or subjects 2. Choose the instrument to measure utility (RS, SG, TTO) 3. Choose subjects who will determine the utility measure 4. Administer the instrument to them 5. Calculate/estimate utility 6. Calculate QALYs 10 Step 1: Describe the health state • Health state: hospital-based kidney dialysis • Description of health state: ―You often feel tired and sluggish. A piece of tubing has been inserted into a vein in your arm or leg. This may restrict your movement. There is no severe pain, but rather chronic discomfort. You must go to the hospital 2-3 times per week (8 hours per visit). You must follow a strict diet (low salt, little meat, small amount of fluid, no alcohol). Many people become depressed because of the nuisances and restrictions. Some feel they are being kept alive by a machine. 11 Step 2: Choose the instrument Three common methods (direct elicitation) 1. Rating scales (RS) 2. Standard gamble (SG) 3. Time trade-off (TTO) good sebataltnet apply forRisotto calenteQIb 1911Based underestendteprecess KER Dost Dgality 5000 3000 1428.57 s.gs 14299Aly III orb AS 9QAy 12 C BCO.RS 4 Step 2: Choose the instrument 1. Rating Scale (RS) • There are two endpoints on the rating scale → perfect health (utility=1.0) and dead (utility=0.0) ―RS may look like a thermometer KER • Other health states on the scale are explained to the subjects • Subjects are then asked to “rate” their current health state between the two endpoints ―Rating depends on individual preferences content fall SCEACLYIM boost fality 5 3000 20001913 Rating Scale (RS) 14 Step 2: Choose the instrument 2. Standard Gamble (SG) • Patient has two options: ―Option 1: patient receives treatment/intervention →However, receiving the treatment/intervention can cause immediate death (failure) or helps return the patient to normal health (success) ―Option 2: Patient declines treatment/intervention →However, the patient is sure to live in the chronic disease state for the rest of their natural life 17 SG process 1. Person with a health condition has two options: ―Take the intervention with a certain probability of success/death, or ―Live with the condition without taking the intervention 2. Person is asked whether they would have the intervention or not have the intervention (i.e., live with the condition) 3. A base (initial) probability of success/failure is offered to the patient if they chose to undergo intervention ―Initial probability of success is set high 4. Probability of success is decreased until person is indifferent (i.e., the two options of either receiving or not receiving the intervention – are difficult to choose between) 5. The probability (of success) at this indifference point is the ‘utility’ associated with the condition 18 SG example: patient on chronic dialysis • Option 1: ―Patient receives kidney transplant (treatment/intervention) →90% chance of success and patient returning to normal life (or 10% probability of death) • Option 2: ―Patient did not receive kidney transplant →Patient continues on their dialysis for the rest of their lives 19 SG example: patient on chronic dialysis • Patient chooses intervention if p(success)=90% ―In order words, the p(death)=10% • P(success) is then decreased, or p(death) is increased continually until patient reaches a point of indifference ―If point of indifference = 50% p(success) →This means that at p(success)=50%, the two options of receiving the transplant (and returning to normal life), or not receiving the transplant (and living with dialysis for additional 40 years=life expectancy) are nearly equal and patient cannot make a decision • This point of indifference (i.e., 50% or 0.5) is the utility of the chronic dialysis health state →The life expectancy of the patient in this disease state (e.g., living for additional 40 years with dialysis) is not changed during this iteration process, only the probability of success is decreased until the point of indifference is reached 20 Standard Gamble 21 ex wouldyourathertolive 5,0ur w thediseaseOR 3. Time Trade-offyoLive agrinperfetheath • Time trade-off is the third method of direct elicitation to measure utility • Here, subjects are offered two options: ―Stay in disease state for a longer time 𝑡, and followed by death, or ―Stay in healthy state for a lesser time 𝑥, and followed by death • Time 𝑥 is varied until the subject is indifferent between the two alternatives ―only time 𝑥 (time that can be lived in healthy state) can be varied ―time 𝑡 (time expected to live in the disease state, or the life expectancy for the disease state) is constant and cannot be varied 25 Time Trade-off (TTO) • Subject is offered two alternatives: ―Option 1: →Patient can choose to continue to live in the disease state for a specific length of time 𝑡 (usually the life expectancy for a person with the disease) and dies ―Option 2: →Patient can choose to live in a healthy state for a shorter time 𝑥, and dies • Time 𝑥 is varied until the respondent is indifferent between the two options • Utility for the disease = ⁄ • TTO essentially means trading off a longer time you are supposed to live with the disease for a shorter time to live healthy 26 TTO example: patient with kidney disease • Assume a 35 years old patient with chronic kidney disease has a life expectancy of 40 years • Patient has two options: ―Live with the kidney disease for 40 years (𝑡) and die, OR ―Live a healthy life (without kidney disease) for 20 years (𝑥) and die • Scenario 1: Patients says he would rather live with the disease for 40 years and die than live healthy for 20 years and die ―Then, the number of years to live in healthy state is increased (maybe from 20 to 25 years) until a point of indifference occurs → Note that life expectancy (40 years) is constant, cannot be varied • Scenario 2: If patient would rather live healthy for 20 years and die than live in disease condition for 40 years and die ―Then, the number of years to live in the healthy state is decreased (maybe from 20 to 15 years) until a point of indifference is reached • Assuming point of indifference = 30 years of healthy state and 40 years of disease state Utility = ⁄ = ⁄ = 0.75 27 TTO 28 Advantages and disadvantages of rating scale • Advantages ―Easiest to administer and least time consuming →Can be conducted through questionnaires without a need for face-to-face interaction →People are familiar with similar scales such as visual analog scale • Disadvantages of ―Time element is not incorporated into utility score calculation ―Respondents are not asked to make preference choices between two interventional options 32 Advantages and disadvantages of standard gamble • Advantages ―It incorporates some element of time in its calculation while the RS does not • Disadvantages of us ghost ―The process is cognitively difficult for the respondent to comprehend ―It is resource intensive →Requires face-to-face administration or using computerized iterative process 33 Advantages and disadvantages of time trade-off • Advantages ―It incorporates time in the disease state more directly than the SG method • Disadvantage ―Like SG, it is more cognitively demanding and resource intensive than RS 34 Comparing the 3 methods • RS is conceptually the easiest and scores are consistently lower than both SG or TTO 8 t • TTO is conceptually easier than SG • TTO values is sometimes lower than SG • Some consider the SG to be the “gold standard” but much research still left to be done to answer the question of “which is best among the three methods?” 35 Step 3: Choose subjects who will determine utilities 1. People from the general public? ―Recommended if taking a societal perspective ―Issues →General public may not fully grasp the severity of the condition compared to patients with the disease →Disease condition may be difficult to communicate using short descriptions • Too much information will be cognitively demanding to the average public 37 Step 3: Choose subjects who will determine utilities 2. People with the disease? ―Patient may understand the effects of the condition better than the general public unfired ―Opinion may be biased →Patient may report higher utility (adaptation to disease?) →Proxies are needed for some patients (e.g., young child, dementia patients) G I 38 Step 3: Choose subjects who will determine utilities 3. Health professionals or disease experts? ―May provide good estimates because of understanding of the disease ―Opinion may be biased →May underrate discomfort and disability compared to patients ―This is the most common method used in the literature Exam →Have better understanding of the disease →Are more accessible to researchers 39 Steps in utility analysis and estimation of QALY 1. Describe the health state or health condition (e.g., hospital-based kidney dialysis) 2. Choose the instrument (RS, SG, TTO) 3. Choose subjects who will determine utilities 4. Administer the instrument to them • • questionnaire (paper-based, electronic) face-to-face 5. Calculate utility 6. Calculate QALYs 40 Step 5: Calculate utilities • Assume the following utilities were determined from a rating scale 1.0 = Completely healthy 0.84 = Kidney transplant 0.58 = Hospital dialysis (patients) 0.56 = Hospital dialysis (public) 0.33 = Hospital confinement 0.0 = Dead 41 Step 6: Calculate QALYs • QALY = Quality Adjusted Life-Years • QALY = Quality (utility) × quantity (life-years) • Quality = utility of health state • Life-Years or Years of life saved = the length of life permitted by an intervention • Equation for QALY: QALY = Utility × Life-Years 42 Examples: PET Calculate QALY 4010.587 5 8 QAlgoutilityxly • Example 1: If dialysis extends the life of a person with renal failure by 10 years at a utility of 0.58 on a rating scale. What is the QALY of the patient? QALY = 0.58 × 10 = 5.8 QALYs (it is assumed that utility per year was constant at 0.58) Interpretation: Even though the patient lived 10 years extra as a result of being on dialysis, when the quality of living those 10 years was accounted for by the utility of the disease/condition, they essentially lived 5.8 years (quality-adjusted) 43 Examples: Calculate QALY Eat • Example 2: Treatment A cost $5,000 and extends life for 6 years at a quality of 0.8, while Treatment B costs $4,000 and extends life for 10 years at a quality of 0.3. • Which is the preferred Treatment between A and B? ―Based on CEA unitlifegear ―Based on CUA Conte Abmy decision ―I will need ICER to make A B o 8 6 48 O 3 90 3 44 Which is preferred between Treatment A and Treatment B? Treatment A : Cost = $5,000 LY = 6 years (outcome for CEA) QALY = 0.8U × 6 years = 4.8 QALYs (outcome for CUA) Treatment B: Cost = $4,000 LY = 10 years (outcome for CEA) QALY = 0.3U × 10 years = 3.0 QALYs (outcome for CUA) 45 Based on CEA (outcome=LY), which treatment would you pick? toomint cheap iless e B A. Treatment A is dominant B. Treatment B is dominant C. I need ICER to make decision Noneedtoconate ICED Treatment Cost LY A $5,000 6 years B $4,000 10 years 46 Based on CEA (outcome=LY), which treatment would you pick? • You will pick Treatment B ―It is dominant →i.e., less costly and more effective (higher LY) ―There is no need to calculate ICER 47 Based on CUA (outcome=QALY), which treatment would you pick? A. Treatment A is dominant B. Treatment B is dominant C. I need ICER to make decision Treatment Cost LY Utility QALY A $5,000 6 years 0.8 0.8× 6 = 4.8 QALYs B $4,000 10 years 0.3 0.3×10 = 3.0 QALYs 48 Based on CUA (outcome=QALY), which treatment would you pick? • Not clear, ICER needed to make decision: ―Treatment A ($5,000, 4.8 QALYs) has higher cost and higher effectiveness than Treatment B ($4,000, 3.0 QALYs) ―No single treatment is dominant (i.e., no treatment is both less costly and more effective) 5100091000 555.5 555.6 ―You need to calculate ICER spDqtpy 48 3.0 ―Then make a decision whether the incremental cost per QALY is worth it →Depends on the WTP 49 Calculate ICER • Treatment A (treatment of interest) vs. Treatment B (reference) • 𝐼𝐶𝐸𝑅 = ∆ ∆ = $ , . $ , . = $ , . KEYWIP NOT WPEN = $556 𝑝𝑒𝑟 𝑄𝐴𝐿𝑌 𝑔𝑎𝑖𝑛𝑒𝑑 KEMPT COSTE Interpretation: If the willingness to pay (WTP) is greater than or equal to $556, then the decision maker will choose Treatment A and consider it ‘cost-effective’ compared to Treatment B. If the WTP is less than $556, Treatment B will be chosen instead even though it is less effective than Treatment A. 50 Advantages of CUA 1. It includes patients’ preferences (i.e., utility) 2. It provides a single measure to incorporate morbidity and mortality ― Different types of health outcomes or treatment/diseases with multiple outcomes of interest can be combined into one common outcome unit – the QALY (unlike CEA) → For example, in cancer treatment, the impact of a drug on length of life, quality of life, and side-effects (toxicity) are important 3. No need to estimate the monetary values of the different outcomes (unlike CBA) 51 Disadvantages of CUA 1. It is time-consuming and difficult to accurately estimate utility values ― Results may vary depending on who assesses the conditions (public vs. patients vs. disease expert) and by what instrument is used (RS vs. SG vs. TTO vs. indirect measures such as EQ-5D and SF-6D) 2. Should you discount utilities just like costs are discounted? ― Many published literature discounts costs and utilities at the same rate (usually 5%) 3. Unanswered question remain ― How much is a QALY worth? → How much are you willing to pay for a gain in QALY? 52 53 How much is a QALY worth? • There is no single answer to the question ―Depends on who you ask • QALY as a unit of health outcomes is recommended and used by many government agencies ―US Public Health Service on Cost-Effectiveness in Health and Medicine ―UK’s NICE (National Institute for Health and Care Excellence) ―Australia’s Pharmacy Benefit Advisory Committee ―World Health Organization (value of QALY = 3 times a nation’s per capital GDP) 54 How much is a QALY worth? -United States • In the United States, the most frequently cited estimate for the value of a QALY is $50,000 as a base case ―sensitivity analyses is conducted for a range of values (e.g., $50,000 $100,000) ―figure has been cited since the ‘80s ―would be higher now if adjusted for inflation →2022 PV = $50,000 × 1.04 = $50,000 × 1.04 = $213,404 (Assuming an MCPI of 4% per year and reference year for citing the $50,000 QALY value is 1985) 55 How much is a QALY worth? • Value associated with a unit gain in QALY depends on who you ask: ―The public usually gives a higher value to QALYs than policy-makers do ―More wealthy countries put a higher value on a QALY than less wealthy countries ―Other considerations include political, equity, and budgetary factors 56 Summary 1. In CUA, costs are measured in dollars while outcomes are measured in QALY 2. QALY incorporates both quality of life (utility) and quantity of life (number of life-years permitted an individual because of an intervention) 3. Utility values usually range from 0 - 1 4. CUA is preferred to CEA because it permits comparison of different treatments with different outcomes that can be converted into a common unit called the QALY 57 Summary 5. Direct methods used to estimate utilities include RS, SG, and TTO 6. Indirect methods used to estimate utilities include use of standardized instruments such as the EQ-5D and SF-6D 7. RS is the easiest to administer and least time-consuming compared to SG and TTO 8. Both SG and TTO are more cognitively demanding and resource-intensive compared to RS 58 Summary 9. If taking a societal perspective, it is recommended that utilities be elicited from the general public 10. QALY = Utility × Life Years 11.There is no universally acceptable answer to the question: How much is a QALY worth? ―$50,000 is the often cited figure in the US ―Wealthier nations put a higher value on the QALY ―Politics and budget are other factors that may influence the value placed on QALY 59