Why It's  Legal To Charge Smokers More
Than Nonsmokers For Health Insurance


For many years its has been customary and generally accepted to charge smokers more than nonsmokers for life insurance (and in some cases for automobile insurance), just as it is to charge people who drive dangerously more for automobile insurance.  However, until about 1984, it was unheard of to charge smokers more than nonsmokers (or nonsmokers less than smokers) for health insurance.

The idea of using differential health insurance rates -- e.g. charging smokers and nonsmokers different rates -- really begin in 1984 when the National Association of Insurance Commissioners [NAIC, the organization made up of state insurance commissioners] noted that (even at that time) some 60%-80% of all health insurance and medical care costs were directly caused by factors over which patients had control: e.g., smoking, overeating, insufficient exercise, alcohol abuse, hard drug abuse, unsafe sex practices, failure to use a seatbelt, etc.

They asked whether it was therefore time to abandon the practice - which went back to the birth of the pre-paid health insurance concept - of charging everyone the same rate.  This practice was adopted because, at that time, most of the causes of health care expenses were things over which patients had little control (e.g., bacteria, viruses, etc.), and therefore it seemed reasonable to charge the same rates to everyone.

But, since by 1984 most health care costs grew directly out of personal health habits, the NAIC wondered if it would be fair, possible, effective, and legal (i.e., not unlawful discrimination) to charge different rates for people who engaged in a variety of unhealthy behaviors. ASH agreed to study the problem and report to the NAIC. 

In our report, we identified three criteria under which such differential health insurance rates based upon various risk factors (e.g., activities) such as smoking would be clearly proper and clearly legal:
  1. The risk factor must cause substantially increased health care costs
  2. The risk factor must be easily and objectively verifiable
  3. The risk factor must be subject to change by the insured and, if changed, result in lower health care costs.

The NAIC adopted our report, and agreed to recommend higher insurance rates based upon only three factors; two of which were smoking and obesity.
SEE PAGES 1 AND 2 OF THE SMOKING AND HEALTH REVIEW: http://tobaccodocuments.org/pm/2500082869-2876.html


Based upon this analysis, and upon the NAIC's position, law professor John Banzhaf, Executive Director of Action on Smoking and Health (ASH), presented arguments for providing discounts to nonsmokers at the Second Annual Blue Cross and Blue Shield Health Risk Management Conference in Washington D.C. held on May 7-9, 1987. The conference title was "Controlling Costs Through Prevention: Creating Incentives for Better Health."

Shortly thereafter, and apparently as a result, the Department of Health and Human Services' [HHS] Health Care Financing Administration was asked to approve a proposal that “certain lifestyle factors be considered the basis for community rating by class"  [i.e., provide a basis for different health insurance rates].

In a letter dated July 6, 1987, Dan Kollmorgen, Director, Division of Compliance, Office of Prepaid Health Care, Health Care Financing Administration, wrote:

“In my letter of June 2, I indicated that the Office of Prepaid Health Care could not approve lifestyle factors as predictors of the utilization of health care services. After discussing the issue in depth, I am prepared to approve the single lifestyle factor of smoking as a basis for community rating by class. The other two proposed lifestyle factors of hypertension and weight remain unapprovable at this time.”


Although the federal government had ruled that smoking - but not obesity - could lawfully be the basis for higher health insurance rates (without any percentage limitations), former HHS Secretary Tommy Thompson adopted the concept and began suggesting its use regarding obesity as well as smoking.  Then it was brought to his attention by a reporter that charging the obese more for health insurance was contrary to existing department policy. http://www.theperfectworld.us/thread.php?id=143&postNum=688

This prompted Professor Banzhaf to file a petition with HHS asking them about permitting higher rates based upon obesity as well as smoking.  A copy of their letter ruling in response to Prof. Banzhaf's petition appears below.  In essence the Department ruled that:
1. Since smoking was a "behavior" rather than a health status, smokers could be charged higher rates than nonsmokers without any limitations.
2. However, because obesity was a "health status" rather than a behavior, higher rates for the obese were limited to a certain percentage, and only under other restrictions.

Letter Ruling by HHS to Prof. John Banzhaf
Permitting Differences
in Health Insurance Rates
Based Upon Both Smoking and Obesity

Department of Health and Human Services

SEP 28 2004

Centers For Medicare and Medicaid Services
Medicare Plan Policy Group
7500 Security Blvd.
Baltimore, MD 21244

Professor John F. Banzhaf III
Professor of Public Interest Law
George Washington University Law School
2000 H Street, NW
Washington, DC 20006

Dear Professor Banzhaf:

We are writing in response to your letter of June 7, 2004 , regarding Secretary Thompson's statement that the Department of Health and Human Services has not prohibited health insurance companies from charging obese individuals higher insurance premiums. You also asked that we allow a differential health insurance rate factor for obesity as a community rating class under section 1308(8) (C) of the Public Health Service (PHS) Act [42 USC 300e-1]. Please forgive the delay in our response.

This letter is to confirm the Secretary's response in the article you referenced and that based on our research States are not bound by federal statute in the case of permissible underwriting practices in the health insurance market they regulate.

In light of the Surgeon General reports and other studies and your request, we have undertaken a re-examination of the position CMS took in 1987 on the question of whether obesity could "reasonably be used to predict the use of health services" (the standard for approval as a community rating factor under section 1302(8)(C). While no Federally-qualified HMO has submitted such a factor for approval, triggering the approval process under section 1302(8)(C), we could theoretically permit Federally-qualified HMOs to use such a factor. This would mean that the HMO would be permitted under the Title XIII community rating rules to charge a higher premium to members classified as obese than the premium charged to otherwise similarly situated non-obese individuals.

However, even if we changed our 1987 position on obesity under the Title XIII community rating provisions, this would have a very limited effect in light of a subsequently enacted provisions contained in the Health Insurance Portability and Accountability Act of 1996 (HIPAA). Most Federally-qualified HMOs would be subject, for most of their members, to the HIPAA requirements that apply to employment-related "Group Market" health insurance. These provisions are found in Part A of Title XXVII of the PHS Act, as added by HIPAA. Section 2702 of the PHS Act is titled "Prohibiting Discrimination Against Individual Participants and Beneficiaries Based on Health Status." While being a smoker is a behavior, not a health status, arguably the condition of obesity is a health status.

Section 2702(b) (1) provides that a covered health plan:

"...may not require any individual (as a condition of enrollment or continued enrollment under the plan) to pay a premium or contribution which is greater than such premium or contribution for a similarly situated individual enrolled in the plan on the basis of any health-status related factor in relation to the individual. . ."

If we determined obesity to be a "health status" for purposes of the requirement in section 2702(b), a decision to reverse the 1987 CMS decision on obesity arguably would only permit Federally-qualified HMOs to consider obesity as a factor with respect to establishing its rates for its "individual market" members, to whom the provisions in section 2702(b) would not apply.

We also note that section 2702(b)(2) clarifies that the prohibition on discriminating based on health status does not prevent an insurer from using premium discounts or rebates, or modifying copayments and deductibles, "in return for adherence to programs of health promotion and disease prevention." The original HIPAA title I interim regulations, issued April 8, 1997 (62 FR 16894) asked for comment on how to implement this exception. HHS, along with the Departments of Labor and the Treasury later proposed, in a Notice of Proposed Rulemaking published January 8, 2001 (66 Federal Register 1421-1435), to limit this exception to those incentives provided pursuant to "bona fide wellness programs." In order to qualify as "bona fide," a wellness program would have to meet four criteria set forth in the NPRM.

The first would be that the cost of providing the discount can not exceed a specified percentage (proposed to be between 10-20%) of the cost of providing the coverage. Second, it must be reasonably designed to promote good health (and not simply a subterfuge for raising rates for certain individuals based on health status). Third, the incentive must be available to all similarly situated individuals, and if the way to qualify for the incentive is unreasonable or medically precluded because of a health factor for a particular individual, the individual must be offered an alternate way to qualify for the incentive. Fourth, plan materials must describe the availability of such alternatives.

Please contact Frank Szeflinski of my staff on (303) 844-7119, if you have any questions.


                            Thomas Hutchinson Director
                            Medicare Plan Policy Group

Action on Smoking and Health (ASH)
701 4th St. NW / Washington, DC 20001 / (202) 659-4310
A national nonprofit, scientific and educational organization founded in 1967.
All donations are fully tax deductible.