The concepts of insurance and insurance underwriting are not new; insurance for commercial ventures existed in some form by the Middle Ages, and life insurance had appeared by the 17th century. Private medical insurance, usually for catastrophic illnesses, was available in the 1800s.1 Within the past 100 years, there has been an explosion in the amount of life and health insurance available and in the diversity of insurance products.

Ultimately, the purpose of any type of insurance is to provide for financial relief in the event of significant economic loss. In this chapter, we will be referring to those types of insurance products related to life and health insurance (Table 113–1). Insurance, essentially, is a contract, a legal agreement between insurer and insured, specifying those losses that are to be covered and the agreed-upon benefit or reparation. To classify an event as an insurable risk demands specific requirements (see Table 113–1). Once satisfied, though, and assuming that a large number of insurable individuals are being considered, these requirements permit the actuarial prediction of the total amount of loss over some defined period of time.2 Employer-sponsored health insurance relies on this “law of large numbers” to determine group premiums without the need for medical evaluation. For life insurance and private health insurance, however, an analytic process termed insurance underwriting serves to transform these large-group data to identify the need for insurance as well as the potential risk of loss for each individual to be insured. The ultimate premium charged is a complex calculation that varies by company and insurance product but will be proportional to the risk assumed by the insurer, allowing for an equitable distribution of economic spread over large groups of people (see Table 113–1).

Physicians serve as resources for and, ultimately, rely on a variety of insurance programs (see Table 113–1) for personal and professional financial security. Basic whole life insurance, the most common type of life insurance in the past, often required a physician examination to confirm insurability. More recently, this product has been supplanted by a wide variety of term and variable financial offerings as individual policies or, more commonly, as part of employer-sponsored group life insurance (see Table 113–1).

Unique to the administration of health insurance is the concept of shared risk, whereby the insurer, the insured, and the insured’s employer (or the government) all participate in payment of premiums and claims. There can be a copayment for routine office visits or a residual amount due over and above a reasonable and customary reimbursement schedule. The goal is to meet expense and utilization goals that allow for maximal distribution of insurance at an affordable price.

Health care in the United States is at a crossroads. Health care–related costs are rising at an alarming rate, currently consuming 17% of the US gross national product (GNP), but these costs may reach 20% of US GNP within the next 10 years.3 Despite these expenditures, life expectancy in the United States lags behind that of other developed nations. Millions of Americans have no medical insurance coverage. Currently, employer- and government-sponsored fee-for-service or managed care organizations provide health, disability, and long-term care insurance for the majority of the US population. The next few years will likely see radical changes in the way health care is paid for and delivered in the United States. Within the medical community, the impact of this changing insurance climate will almost certainly be enormous.4,5

Application and Medical Requirements

Most applicants for insurance are in excellent health and are quickly offered policies at standard or even “preferred” premium rates. When medical impairments are revealed, however, each must be correlated with long-term mortality data relevant to that impairment. The process begins with an insurance application that often contains important past medical history and review of systems information. Authorized query to one of the national informational database exchanges can provide additional leads. The next step, an insurance medical examination with laboratory assessment, usually includes comprehensive medical history and physical examination review. Noninvasive cardiac testing can also be considered; stress testing, in particular, is often requested if large amounts of insurance are applied for (age and amount guidelines) or if the initial impression suggests intermediate or high cardiac risk (for cause guidelines). At each step in the process, the cardiology consultant serves as a member of the medical underwriting team, reviewing all of the cardiac information available, including electrocardiograms and stress test tracings.

The Attending Physician Statement

Underwriting risk assessment guidelines often direct the patient’s physician to submit medical information to the insurance company in the form of the attending physician’s statement (APS). This can include an outline of recent medical history and will often contain office and hospital records for review. Clinical problems identified in the APS are analyzed for severity of disease, extent of clinical evaluation, and thoroughness of clinical follow-up to provide data for mortality analysis.

Mortality Analysis

Each medical impairment identified during the medical underwriting evaluation must be correlated with long-term population survival statistics relevant to that disease process. The number of excess deaths attributable to that impairment is used to derive a mortality ratio (the observed deaths in a population of impaired individuals compared with the expected deaths for an otherwise comparable standard population).6-8 Published data relevant to mortality assessment derive from several sources, including insured experience studies, clinical studies, and national databases. The mortality ratio serves as a useful measure for comparing mortality projections among various medical impairments. In general, the higher the mortality ratio calculated, the greater the relative risk assumed by the company to provide life insurance for individuals affected by that impairment. The mortality ratios calculated for various medical conditions are integrated into a table of risk classes or ratings. Applicants within a rating class are grouped together to be assessed for similar mortality costs of insurance. The mortality cost is an important element in premium computation.

Insured Populations

Insurance policies are generally issued for durations (terms) that can continue for 20 to 30 years. The most valuable prognostic data, then, would necessarily derive from studies with extended periods of follow-up. Excellent long-term follow-up data are available for insured populations based on medical conditions, laboratory abnormalities, demographic characteristics, avocations, and personal habits identified at the time of original insurance application (Table 113–2). Results are usually expressed as mortality ratios or excess death rates in the impairment group compared with a standard population. As might be expected, however, these data become limited where significant medical advances or transformations in demographics or lifestyles occurred during the period of study.

Clinical Studies and National Databases

Important predictive data can also be derived from long-term clinical and epidemiologic studies published in the clinical literature. Particularly useful for mortality assessment are reports derived from the nationally maintained databases, such as the Surveillance, Epidemiology, and End Results (SEER) and United Network for Organ Sharing (UNOS) databases. Data from these databases, in which interval survival data are usually reported (eg, 5-year Kaplan-Meier survival curves), can be extrapolated to provide actuarial information useful to the calculation of mortality risk. At times, however, the experience reported in clinical studies must be interpreted cautiously because the focus or inclusion criteria of a particular study may not allow extrapolation to an insured population. Iacovino8 discusses such a situation in a review of mortality analysis methodology. Clinical investigators followed 48 patients with a mean age of 36 years for approximately 6 years and noted a favorable observed mortality of 10% for the entire period.9 However, reference to the US Standard Life Tables (1979-1981) revealed a much lower expected mortality (∼1.46%) at this age for the same length of follow-up. The estimated mortality ratio of 685% (10%/1.46% × 100) represented a highly substandard risk level for life insurance purposes, even though it may represent good clinical results in young patients with severe cardiac disease.10

Health Insurance Risk Assessment

Most applicants for health insurance apply as members of a group, usually through the workplace. The basic insurance coverage is often an important part of the employee benefits package, with costs borne in large part (or entirely) by employers. Government-sponsored insurance such as Medicare also involves groups identified by age, income, or handicap. Medical risk assessment plays little role in the underwriting of these policies. In private health insurance, premium determination is based on an experience rating determined through historical review of medical expenses most recently incurred by the members of the group. Government-sponsored insurance premiums represent a complex interplay of mortality projection, budgetary constraints, and political considerations.

Individuals who desire additional coverage over and above that available at work or those without a workplace benefit program can choose from a variety of individual health insurance plans. On completion of the usually mandatory medical questionnaire, the risk assessment process becomes quite similar to that of life insurance underwriting.

Certain underwriting factors, such as admitted medical impairments or claims history, may call for exclusionor limitation of health insurance coverage. The limitation can apply to all related claims or only to those over and above a specified claim frequency or duration. Although seemingly restrictive, the current system allows for health insurance coverage for some that would otherwise be denied any coverage at all. Advocates of health care reform have called for a health care system that would provide universal coverage independent of income or preexisting condition. Such a system has been developed in Massachusetts11 and is being monitored closely. Successful reform will most likely require innovation and compromise and may ultimately lead to fundamentally new approaches to health care delivery.12

In the past, most life insurance policies were issued on a “standard” basis, excluding most serious medical impairments but disregarding factors that might identify improved longevity. Beginning in the mid-1980s, innovation in actuarial product design led to the concept of the “preferred” insurance risk. The criteria that define these preferred risk policies have been shown to predict life expectancies longer than the general population when large numbers of individuals are considered. At the other end of the spectrum, an applicant with mild coronary disease may well be an insurable risk, but at a “substandard” rating, requiring an additional mortality cost added to the base premium.

To categorize risks as “preferred,” “standard,” or “substandard,” special attention is directed to the presence of known atherosclerosis risk factors such as high blood pressure, diabetes mellitus, hyperlipidemia, smoking history, and obesity. In addition, a strong family history of cardiovascular disease has been confirmed in studies in insured as well as in other broad-based epidemiologic populations to be an independent risk factor for coronary heart disease (CHD), with mortality ratios in insureds of 189% and 121% for men and women, respectively.13