poly: Why We Die

Health specialists are still relatively ignorant about some really
big factors that cause some people to die, and others to live.
Sociologists and other social scientists have long studied some
very strong correlations between social status and health, but
medical science has been slow to accept that wealth causes health.

The June 3, 1998 Journal of the American Medical Association leads
with a study that medical researchers consider careful enough to
command their attention. Here is a June 3 news article on it:

------------------------------------------------------------------
ACCESS/QUALITY/COST - MORTALITY: RISKS DON'T EXPLAIN HIGHER RATES
AMONG POOR

      A study in this week's Journal of the American Medical
 Association finds that low-income people "have a death rate as
 much as three times higher than that of other groups," even when
 risky health behaviors are taken into account. Led by University
 of Michigan School of Public Health's Paula Lantz, Ph.D., the
 researchers "looked at 3,617 Americans and their living habits"
 over a 7.5-year period. When "smoking, drinking, overeating and
 lack of exercise" were adjusted for in looking at the study
 population, the researchers found that these factors
 "account[ed], at most, for 13% of the" mortality gap between poor
 people and other income groups (AP/Arizona Daily Star, 6/3).
 Lantz said, "Among people with lower incomes, they did have a
 higher rate of risky health behaviors. But in looking at who
 lived and died over the follow-up period of our study, we found
 that the increased risky health behaviors among the poor people
 was not what was explaining the higher mortality rate" ("All
 Things Considered," NPR, 6/2). The study found that "Americans
 with income below $10,000 a year had a death rate of 3.22 times
 that of people making $30,000 or more." After taking into
 account health risks, "the death rate among the poor was still
 2.77 times higher" (AP/Daily Star, 6/3).
      PUBLIC POLICY IMPLICATIONS
      Based on their findings, the study authors write that
 "public health policies and interventions that exclusively focus
 on individual risk behaviors have limited potential for reducing
 socioeconomic disparities in mortality. ... Increasing health
 promotion and disease prevention efforts among the disadvantaged
 is not a 'magic policy bullet' for reducing" the mortality
 differences (Lantz et al, JAMA, 6/3 issue). Commenting on the
 study, Tom Burke, a specialist in risk sciences at the Johns
 Hopkins School of Public Health, said: "A big part of being poor
 is lack of access to medical care or perhaps not feeling
 culturally comfortable with the medical care providers, so things
 that are little health problems become bigger health problems,
 and you have people presenting themselves at a much more advanced
 stage of disease. On the other hand, in urban areas where
 there's a higher concentration of pollution and environmental
 poor quality -- for instance, air quality -- you clearly see
 asthma rates in children and issues like that being very
 important."
      MORE STUDY NEEDED
      NPR's Brenda Wilson reported, "Research is beginning to
 study how attitudes affect health, the stress that comes with
 having too often to make do without the expectation that things
 will get better. An accompanying editorial argues that the
 children of the poor learn from adults around them that the world
 is a frightening and hostile place -- attitudes that contribute
 to stress, poor choices and early death" ("All Things
 Considered," 6/2). In the JAMA editorial, Dr. Redford Williams,
 chief of behavioral medicine at Duke University, writes:
 "Instead of simply targeting risky health behaviors, any
 effective intervention to ameliorate the impact of lower
 [socioeconomic status] on health and disease will need to also
 reduce hostility, depression and social isolation -- and perhaps
 to correct autonomic imbalance as well." Williams suggests
 further areas of research, including "whether harsh and adverse
 conditions during early childhood are indeed responsible for the
 clustering of health-damaging biologic, behavioral and
 psychosocial factors among adults with lower [socioeconomic
 status]." This research, he states, could be used to design
 interventions for young children that, "[based] on available
 evidence," would "ensure that the environments in which children
 live are not dominated by harsh, punitive overseers, but rather
 by competent caregivers ... who combine warm, positive
 communications with reasonably high expectations of competence on
 the child's part" (6/3 issue). Click here to see an abstract of
 the JAMA study.
------------------------------------------------------------------

Here are the death rate ratio parameters from their model:

Age 25-34 35-44 45-54 55-64 65-74 75+
            1.0 2.66 3.46 9.30 16.78 40.00
Sex Male Female
           1.0 .41
Race NonBlack Black
               1.0 1.19
Residence Rural Suburban City
           1.0 1.16 1.52
Education 16+yrs 12-15 0 -11
           1.0 .95 .90
Income 30K$+ 10-29K$ <10K$
            1.0 2.14 2.77
Smoking Never current former
            1.0 1.26 1.28
Alcohol drinks/mo. Moderate None Heavy
                         1.0 1.13 .85
Body Mass Normal Underweight Overweight
             1.0 2.03 .94
Physical Activity Quintiles
    5(high) 4 3 2 1(low)
       1.0 1.46 1.60 2.25 2.91

The media & article focused on the income parameters, which
are indeed striking. But other parameters are also striking.

While smoking gets tons of media attention, it only raises
mortality rates by ~27%. And being black is only a 19% hit.
In contrast, being a women lowers your mortality by 60%, and
living in a city raises mortality by 50%, relative to being rural.
Drinking alcohol and being overweight don't hurt, but physical
activity matters for a factor of 3, as much as income does.

Robin Hanson
hanson@econ.berkeley.edu http://hanson.berkeley.edu/
RWJF Health Policy Scholar, Sch. of Public Health 510-643-1884
140 Warren Hall, UC Berkeley, CA 94720-7360 FAX: 510-643-2627
Received on Mon Jun 29 19:01:42 1998