This preview shows page 1. Sign up to view the full content.
Unformatted text preview: Health Insurance: the Basics
PAM 4350 February 10, 2009 Last Thursday's Quiz Grades are available on Blackboard through the grade book function. Mean score was 8.0 out of 10. You can also pick up your copy from Anna and Christy, the teaching assistants, after class, or from me during office hours. 2/19 Midterm Covers material through 2/12 Last year's midterm on Blackboard. "NC" = material not covered this year. Last year's answer key on Blackboard. Extended office hours next week (Blackboard announcement). Sloan Program in Health Administration 5 Year Accelerated BS/MHA Program For Cornell undergraduates, accelerated 5 year BS/MHA Program. Senior year in Sloan followed by one extra year for Masters degree. Deadline for Applications: March 15th Application, GRE or GMAT scores, Statement of Purpose, 2 letters of reference, transcripts from all colleges Tuesday, February 17thOpen House 5:307:00 MVR 114 For more information or to get an application for this or our regular 2 Year MHA Program please contact: Geysa Smiljanic in MVR 122, email@example.com or Cathi Calori, Associate Director MVR 118, firstname.lastname@example.org Website: www.sloan.cornell.edu Quality and Costs of Care for Medicare Patients Hospitalized for Heart Attacks, Hip Fractures, or Colon Cancer, by Hospital Referral Regions, 2004
Median relative resource use = $27,499
1.20 Quality of Care* (1-Year Survival Index) 1.10 1.00 0.90 0.80 $0 $5,000 $10,000 $15,000 $20,000 $25,000 $30,000 $35,000 $40,000 Relative Resource Use** * Indexed to risk-adjusted 1-year survival rate. ** Risk-adjusted spending on hospital and physician services using standardized national prices. Source: E. Fisher, J. Sutherland, and D. Radley, Using a 20% national sample of Medicare beneficiaries. The Dartmouth Explanation Lack of scientific evidence to guide many clinical decisions creates uncertainty for physicians When in doubt, patients and physicians believe that more care means better care Physicians and hospitals are rewarded for being busy In areas where there are plenty of specialist physicians and hospital beds, primary care physicians are more likely to refer patients to specialists; specialists then use expensive medical technology Physicians Differ Greatly in How They Respond to Patients Experiencing Chest Pain A physician at Cook County Hospital in Chicago showed 20 case histories of emergency room patients complaining of chest pain. "We asked the doctors to estimate on a scale of zero to 100 the probability that each patient was having a heart attack and the odds that each patient would have a major life-threatening complication in the next 3 days. In each case, the answers we got pretty much ranged from zero to 100. It was extraordinary." Source: Gladwell, Blink, 2005. Medical Care is Fragmented in High-Cost Areas
Spending on Medicare Patients w/ a Chronic Condition in Their Last 2 Years of Life
MDs Hospital MD visits Involved days/pt per pt
Mayo Clinic Cleveland Clinic Mass General Johns Hopkins UCLA $53,400 $55,300 $78,700 $85,700 $93,800 20.3 26.1 29.5 25.7 38.5 12.0 14.8 17.3 16.5 18.5 23.9 33.1 39.5 28.9 52.8 Source: 2008 Dartmouth Atlas. Dartmouth Conclusion: Some Regions Have Gone Past the "Flat Part of the Curve" Source: Feldstein, 2007. 3 Key Questions
1) Why does spending differ so much across different regions? 2) Why don't regions that spend more have higher quality of medical care? 3) Does this mean the huge increase in medical spending in the U.S. since 1950 was not worth it (i.e., Cutler is wrong)? It's Still Likely That Increases in Medical Spending Have Been Worth It, on Average Cutler was looking at the returns to large increases in spending (e.g., from A to D on slide #8). Possible to have large positive returns to large changes in spending (e.g., from $0 to $30,000 for heart attack patients), but small/no returns to small changes (e.g., from C to D on slide #8). Some people are skeptical about the Dartmouth method. The ideal study would randomly assign people to regions with low versus high medical spending and measure the difference in health outcomes. What Can We Learn From This Figure? Source: Almond et al., 2008. And This One? Source: Almond et al., 2008. Hospitals Treat Babies Who Weigh Just Under 1500 Grams Much More Intensively Than Those Just Over Source: Almond et al., 2008. Conclusions for 2/5 Topic There are large differences in the amount of medical care people receive and the cost of care between regions There is little evidence that health outcomes are better in high-cost regions, and it may even be worse Higher costs appear to be driven by: uncertainty regarding the best way to treat patients belief that more care is better availability of physicians and hospital beds fragmented care: "too many cooks spoil the broth" Ongoing debate regarding whether the Dartmouth results indicate that most expensive medical technologies have a low value The Rest of Today's Objectives 1) Describe the benefits of health insurance 1) Discuss how health insurance works 1) Discuss problems created by health insurance, and how those problems are addressed We're Shifting to the 2nd Part of the Course
Payers: consumers, taxpayers, businesses, government
$$$ Determine how much to spend on health care and what types of benefits to include in health plans. Insurers
$$ General contractors: negotiate w/ providers regarding how and how much they will be paid. "Manage" patients' medical care. Sub-contractors: combine medical inputs (e.g., drugs, medical devices, and labor) to provide medical care. Supply medical inputs or technologies (e.g., drugs, devices) to providers. Providers: MDs and hospitals
$ Suppliers Today is Your Lucky Day Imagine that you have just won a $1 million lottery prize. You have two payout choices: Option A: receive $1 million today Option B: return your $1 million in order to play a 2nd lottery where there is a 10% chance of winning $5 million 89% chance of winning $1 million 1% change of receiving nothing Which option would you prefer? Option B is Clearly Better... Isn't It? Expected Payout
Option A (1.0) X $1 million = Option B (0.10) X $5 million + (0.89) X $1 million + (0.01) X $0 million = $1 million < $1.39 million What Would This Woman Do if Health Insurance Policies Were Not Available?
Example: healthy 40-year old female Illness/Disease Heart Disease Diabetes Cancer Stroke Pneumonia ... None Expected Probability Treatment Cost Cost 0.01 X $20,000 = $200 0.02 $15,000 $300 0.002 $40,000 $80 0.001 $20,000 $20 0.005 $8,000 $40 0.30 $200 $60 $2,000 An Actuarially Fair (or "Pure") Premium Under a System of Experience Rating
Example: healthy 40-year old female Illness/Disease Heart Disease Diabetes Cancer Stroke Pneumonia ... None Expected Probability Treatment Cost Cost 0.01 X $20,000 = $200 0.02 $15,000 $300 0.002 $40,000 $80 0.001 $20,000 $20 0.005 $8,000 $40 0.30 $200 $60 $2,000 + 15% loading charge = $2,300 actual premium Health Insurance Premiums Actual premium paid by the consumer = Pure Premium (or actuarially fair premium) = expected medical spending for a group of people with the same risk level (i.e., same age, gender, ...) Loading charge = amount of money the insurance company charges in addition to the pure premium Covers administrative cost, claims processing cost, marketing cost, profit Pure Premium + Loading Charge Why would anyone pay more than the actuarially fair premium for health insurance? Why would anyone pay more than the pure premium for health insurance? Reducing risk is valuable to consumers If health risks are assessed accurately, a person will purchase insurance if his/her risk premium > loading charge Health insurance also allows people to receive medical care that they otherwise could not afford in years when they are very sick (e.g., cancer in the previous example) Risk Premium = the amount above and beyond the pure premium someone is willing to spend to get insurance Primary Benefit of Health Insurance Protects against financial risk Most people prefer to face a (relatively) small but certain loss (i.e., the health insurance premium) in exchange for avoiding the risk of a large financial loss Insurance shifts financial risk from an individual to the group by pooling resources; losses are shared by all members of group. It's difficult to predict whether/when a lowprobability but very expensive medical condition will occur Risk Spreading
Concentration of Health Spending in the U.S. Population, 2003 100% 80% 60% 40% 20% 0% Top 1% Top 5% Top 10% Top 15% Top 20% Top 50% Bottom 50% Population Percentile Ranked by Health Care Spending
Source: KFF calculations using data from Agency for Healthcare Research and Quality, Medical Expenditure Panel Survey, 2003. 97% 64% 49% 24% 3% 73% 80% The Person Below Wants to Buy Insurance From the Same Company. What Should His (and Her) Premium Be? Example: 60-year old male with diabetes Illness/Disease Heart Disease Diabetes Cancer Stroke Pneumonia ... None Expected Probability Treatment Cost Cost 0.05 X $20,000 = $1,000 1.00 $15,000 $15,000 0.01 $40,000 $400 0.01 $20,000 $200 0.02 $8,000 $160 0.00 $200 $0 $20,000 + 15% loading charge = $23,000 actual premium Health Insurance Premiums Premiums determined by 1 of 2 methods: Experience Rating = $2,300 and $23,000 Community Rating (CR) = $12,650 for both Based on a person's predicted costs based on their current age, gender, health conditions, health behaviors... Premiums differ between people due to different health levels Premium same for all people regardless of age, gender, whether smoke, health conditions... Financial risk spread among a larger community Healthy people subsidize sicker people (equity) Questions 1) Who is likely to buy health insurance in each premium pricing system? 2) Which of these two pricing systems is "fairer"? 1) How are health insurance companies likely to behave in a community-rated pricing system? Community Rating (CR) Can Cause Adverse Selection Charging everyone the same premium, regardless of health level, promotes equity If the premium covers the health insurer's expected costs, it exceeds the expected expenses for the young and healthy (YH), but falls short of the expected expenses for the older and sicker (OS) Results Some YH's may not want to buy health insurance. As they drop out, it raises premiums for those remaining. Insurers might not want to sell policies to OS's CR probably increases the number of uninsured while altering mix of insured people (sick people more likely to have health insurance than healthy people) Sicko Video Peter Jennings Video 13:00 15:40 Control of Adverse Selection Group Insurance: a group of people who come together for reasons other than buying insurance (e.g., employees at large companies) should contain both healthy and sick people, but the overall composition should be fairly representative of the general population Preexisting condition clause Waiting period Most People in the U.S. Receive Health Insurance from Their Employer
Source of Health Insurance Among the Non-Elderly, 2007 Uninsured
17.2% 100% = 261.4 million Public (Medicaid, Medicare, Military) Individually purchased 14.3% 62.9% 5.5% Employer-Provided ("group market") Source: U.S. Census Bureau. How Premiums Are Set for Most Americans Employers are typically experience rated: firms with relatively sick employees are charged higher premiums than firms with relatively healthy employees But most firms communityrate premiums within the firm: each employee is asked to contribute the same amount, regardless of their health/risk The Treatment Costs Below Describe How Much Medical Care This Woman Would Receive if She Were NOT Insured Example: healthy 40-year old female Illness/Disease Heart Disease Diabetes Cancer Stroke Pneumonia ... None Probability 0.01 0.02 0.002 0.001 0.005 0.30 Expected Treatment Cost Cost $20,000 $15,000 $40,000 $20,000 $8,000 $200 = $200 $300 $80 $20 $40 $60 $2,000 X Moral Hazard When the presence of insurance changes the behavior of the person being insured If you are at a party with an open bar, moral hazard is when you drink more than you would have if you had to pay full price for drinks If you have a nodeductible carinsurance policy, you may drive a bit more recklessly In health care, moral hazard exists if you choose the expensive prescription drug (for example) rather than the generic drug you would have chosen if you had to pay the full price Types of cost sharing: Deductible CoPayment Patient costsharing is a tool to reduce moral hazard Dollar amount per year (e.g. $1,000) that a patient has to pay before the insurer pays for any medical expenses Fixed dollar amount (e.g. $20) a patient pays per MD office visit or per prescription The insurer pays a percentage (e.g., 80%) of a bill and the patient pays the remainder (e.g., 20%); usually for hospital stays or MD fees Maximum outofpocket amount a patient is required to pay over the course of a year CoInsurance Stoploss provision Summary Main benefit of health insurance is protecting people from the financial risk associated with hard to predict, lowprobability, and expensive medical conditions. Are several problems related to health insurance: adverse selection: high-risk enrollees are unprofitable under community rating. moral hazard: low price spurs demand for medical care ...
View Full Document
- Spring '08