The quality of medical care, behavioural risk factors, and longevity growth

Frank R. Lichtenberg 27 June 2009



The cost of medical care continues to rise rapidly in the US and other industrialised countries. According to a report from consulting firm PricewaterhouseCoopers, US employers who offer health insurance coverage could see a 9% cost increase between 2009 and 2010, and their workers may face an even larger increase.

Some observers argue that rapidly increasing health care expenditure is due, to an important extent, to medical innovation – the development and use of new drugs, diagnostics, and procedures. For example, the Kaiser Family Foundation (2007), citing Rettig (1994), claims that “advances in medical technology have contributed to rising overall US health care spending.”

Other observers argue that most medical innovations do not improve people’s health. Lexchin (2004), for example, claims that “at best one third of new drugs offer some additional clinical benefit and perhaps as few as 3% are major therapeutic advances.”

If both of these claims were true, medical innovation would result in the worst of both worlds – a large increase in cost and little or no increase in benefit (in the form of improved health outcomes). However, a study that I have recently performed casts considerable doubt on both of these claims. My findings indicate that medical innovation has yielded significant increases in life expectancy without increasing medical expenditure.

My study (Lichtenberg 2009) examines the effect of the quality of medical care, behavioural risk factors, and other variables on life expectancy and medical expenditure using longitudinal state-level data. As shown in Figure 1, the rate of increase of longevity has varied considerably across US states since 1991.

Figure 1. Increase in life expectancy at birth 1991-2004, by state

I examined the effects of three different measures of the quality of medical care. The first is the average quality of diagnostic imaging procedures, defined as the fraction of procedures that are advanced procedures. The second is the mean vintage (FDA approval year) of outpatient and inpatient prescription drugs. The third is the average quality of practicing physicians, defined as the fraction of physicians that were trained at top-ranked medical schools.

I also examined the effects on longevity of three important behavioural risk factors – obesity, smoking, and AIDS incidence – and other variables – education, income, and health insurance coverage – that might be expected to influence longevity growth. My econometric approach controlled for the effects of unobserved factors that vary across states but are relatively stable over time (e.g. climate and environmental quality), and unobserved factors that change over time but are invariant across states (e.g. changes in federal government policies).

The gains from medical innovation

The indicators of the quality of diagnostic imaging procedures, drugs, and physicians almost always had positive and statistically significant effects on life expectancy. Life expectancy increased more rapidly in states where (1) the fraction of Medicare diagnostic imaging procedures that were advanced procedures increased more rapidly, (2) the vintage of self- and provider-administered drugs increased more rapidly, and (3) the quality of medical schools previously attended by physicians increased more rapidly.

Between 1991 and 2004, life expectancy at birth increased 2.37 years. The estimates imply that, during this period, the increased use of advanced imaging technology increased life expectancy by 0.62-0.71 years, use of newer outpatient prescription drugs increased life expectancy by 0.96-1.26 years, and use of newer provider-administered drugs increased life expectancy by 0.48-0.54 years. The decline in the average quality of medical schools previously attended by physicians reduced life expectancy by 0.28-0.47 years.

The availability of data from Australia’s universal health care system, Medicare Australia, allowed me to provide some additional evidence about the impact of advanced imaging technology on mortality. I estimated difference-in-difference models of the effect of advanced imaging innovation on age-specific mortality rates. Demographic groups that had above-average increases in the number of advanced imaging procedures per capita had above-average declines in mortality rates, but changes in mortality rates were uncorrelated across demographic groups with changes in the number of standard imaging procedures per capita. Estimates of the effect of diagnostic imaging innovation on longevity based on Australian data are quite consistent with estimates based on US data.

The increased fraction of the population that was overweight or obese, rising from 44% to 59%, reduced the increase in life expectancy by .58-.68 years. The decline in the incidence of AIDS is estimated to have increased life expectancy by .18-.20 years. The small decline in smoking prevalence may have increased life expectancy by about 0.10 years.

Growth in life expectancy was uncorrelated across states with health insurance coverage and education, and inversely correlated with per capita income growth. The 19% increase in real per capita income is estimated to have reduced life expectancy by .34-.43 years. The sum of the contributions of all of the factors to the increase in life expectancy is in the 0.85-1.32 year range. Consequently, between 1.05 and 1.52 years of the 2.37-year increase in life expectancy is unexplained.

Greater coverage, lower costs

Although states with larger increases in the quality of diagnostic procedures, drugs, and physicians had larger increases in life expectancy, they did not have larger increases in per capita medical expenditure. This may be the case because, while newer diagnostic procedures and drugs are more expensive than their older counterparts, they may reduce the need for costly additional medical treatment. The absence of a correlation across states between medical innovation and expenditure growth is inconsistent with the view that advances in medical technology have contributed to rising overall US health care spending. Increased health insurance coverage is associated with lower growth in per capita medical expenditure.


Kaiser Family Foundation (2007) “How Changes in Medical Technology Affect Health Care Costs,” March.

Lexchin, Joel (2004), “Are new drugs as good as they claim to be?,” Australian Prescriber.

Lichtenberg, Frank, “The Quality of Medical Care, Behavioral Risk Factors, and Longevity Growth,” NBER WP 15068.

Rettig, Richard A.(2007), “Medical Innovation Duels Cost Containment,” Health Affairs (Summer 1994): 15.



Topics:  Health economics

Tags:  life expectancy, Medical care, medical innovation, healthcare costs


I suggest taking a different perspective by replacing the term cost with investment. I will not attempt to place a monetary value on "quality of life". At one point in history "capital" was virtually synonymous with land. In the "information age" education and health probably contribute more to an individual's productivity than land does. It is easy to conceive of tuition fees as an investment in human capital. If an individual pays the fees it is a private investment. If society pays the fees it is a public investment. In either case there would be no shortage of ways to examine the return on investment in education is. For example one could compare incomes of educated and uneducated people in a given region. National productivity measures could be compared to national education levels. It makes intuitive sense that a more educated society should be a more productive society.

It also makes intuitive sense that in order to exploit one's human capital one needs to be healthy. The same would hold true for an aggregate measure of health; thus health is a measure of a society's human capital. The challenge then is to measure return on investment in health care.

Through most of history very little was known about health or disease. What was known could be learned by an individual in a few years at a medical school and a few more at a hospital. Doctors are trained to be human databases. This model has not changed in hundreds of years. It is no longer reasonable to presume that a doctor knows more than a small portion of what there is to know. Recently the medical profession has been criticized for not having an adequate grounding in science. "Science" is the major source of new treatments based on improved understanding of disease models. Drugs are the prime treatment tools available to doctors yet few if any have ever studied pharmacology. As a consequence doctors are being reduced to technicians who must follow recipe treatments because they do not understand the fundamental workings of medications.

The depth of aggregate human knowledge in many medical disciplines is so great that only a few doctors can truly achieve expert status. Moreover to do this they must pick a very narrow field to become expert in. A brilliant oncologist may not recognize illness or disability caused by anything other than cancer. A general practitioner has to "guess" which specialty has the particular expertise necessary to treat a patient presenting with a disorder that is complex or hard to recognize. It is very easy to guess wrong.

The practice of medicine is suffering from problems related to information management. Luckily other industries have addressed many of the complex information management issues. All experiments pitting a doctor using computer aided diagnostics against one that is not have demonstrated substantial improvements in accuracy. Raising the ratio of correct to incorrect diagnoses leads to asymmetrically better reductions in treatment costs. Less time/money is wasted consuming more diagnostic resources and proper treatment can occur earlier in the disease process. It is much cheaper to treat a disease in its early stages than in later stages.

Imaging technology is so powerful now that it is being used to interpret people's intentions -brain machine interfaces can interpret a persons mental activity to the point where a person can operate machinery by thought alone. Indeed, the sensing equipment is so sensitive that it provides the person with a severed spinal cord the same level of fine motor control that s/he would have if their spinal cord were restored.

By 2010 the cost of sequencing a complete human genome is likely to fall to the $100 range. There are large numbers of tests that could be performed today that would greatly improve the likelihood of prescribing the most effective medication on the first try.

These and many other extant technologies could be deployed quickly but for the fact that the medical community is not technically qualified. Much of what doctors memorize in medical school can be off-loaded onto computer databases. Medical education could then focus on learning the underlying science such that doctors could implement the technologies that would improve their productivity. Such an education would make it possible for them to keep abreast of new scientific developments throughout the course of their careers.

I suggest that the main reason the "cost" of health care is rising is that the relative productivity of the medical profession is falling. This is a problem that has a solution -increased investment in computer aided diagnostics, including knowledge bases, improved imaging, greater use of bioassays, imbedded sensors, etc. Doctors need a very different skill set than they currently graduate with (the exception being clinical experience). To change this would involve investments in human capital.

I close with the example of psychiatry. The world health organization reports that the greatest cause of lost productivity is mental illness, particularly mood disorders such as depression and bipolar disorder. The main treatment options available are medications. Rapid advances in neuroscience are leading to new treatments based on improved understanding of brain function, and neuropharmacology. Indeed the next "generation" of drugs developed to treat specific disorders may be completely unrelated to the medications available today. The drugs currently available were developed before science had unraveled many of the mechanisms of disease. Indeed the drugs were often approved without understanding how they worked. Much research has focussed on determining what the current drugs do. The next generation medications will be engineered. It is possible to some extent to match a medication to the specific brain dysfunction of a given patient today. This capability will improve significantly in the near future; however, psychiatrists will require an understanding of neurology and psychopharmacology.

Psychiatrists are not trained in either of these disciplines. This makes it very difficult for a psychiatrist to diagnose or select an optimal treatment. The economic burden of mood disorders is enormous. It should be relatively easy to measure the economic gains associated with more effective treatments. More effective treatments will result from greater investment.

Courtney C. Brown Professor of Business at the Columbia University Graduate School of Business