Page 13 - Mecklenburg_Medicine_March-2018

Feature
Diabetes and Welfare
By Gordon Hull, Director, Center for Professional and Applied Ethics,
Associate Professor Philosophy and Public Policy, UNC Charlotte
T here is growing empirical evidence that confirms what
most of us know intuitively: The poor have a harder time
controlling their diabetes than those of greater means.
A couple of recent papers by Seth A. Berkowitz, MD,
serve to illustrate. In one, Berkowitz and his coauthors found that a
number of indices of “material insecurity” were associated with poor
diabetes control; of the various factors they studied, food insecurity
was most closely linked. In the other, Berkowitz and colleagues found
that unstable housing was associated with a staggering increase in
visits to the emergency room or hospitalization for diabetics (adjusted
odds ratio 5.17; 95 percent CI 2.08–12.87). These individuals are not
homeless — “housing unstable” means they can’t pay rent/mortgage;
they moved two or more times in the last 12 months; or they live in
a place they do not own/rent. Less than 1 percent of these patients
received help with their housing. In the meantime, many more people
might be called “housing precarious.” A Harvard report noted that a
quarter of all renters paid half or more of their income for housing,
and underscored that “much to their detriment, cost-burdened
households are forced to cut back on food, health care and other
critical expenses.” In Charlotte, a 2017 report estimated the city had a
deficit of 21,000 units of housing that would be affordable to someone
making less than 50 percent of the city’s median income.
This presents both a moral and an economic problem. The
economic problem should be clear enough and can be put this way:
It is almost certainly cheaper to assist with housing than it is to
pay for ER visits. As Berkowitz notes, diabetes cost the country
$217 billion last year, including more than 21 million ER visits and
hospitalizations. A Kaiser Health News report from 2010 found that
diabetes then cost $83 billion a year in hospital spending, or 20 percent certainly cheaper
of total hospital spending in the country. Research in diabetes care
estimated that, in 2012, “40 percent of all healthcare expenditures
attributed to diabetes came from higher rates of hospital admission and
longer average lengths of stay per admission, constituting the single
largest contributor to the attributed medical cost of diabetes.”
Behind the obvious fiscal reasons to trim diabetes-related
hospitalizations, however, lurk some difficult moral issues. It seems
to me that findings like these suggest we need to expand the social
safety net — welfare, if you prefer — dramatically, if we as a society
want to say that we care for low-income people with diabetes. On
welfare, we have been doing the opposite at least since the 1996,
Clinton-era welfare reform legislation. There are lifetime maximums
on TANF and three-year limits to SNAP for many people. These
recent studies remind us that social determinants of health matter,
and that one way of addressing those is with traditional welfare
programs that are in a 20-year decline. Of course, even locally
there are questions — the logical extension of providing housing
assistance to those with diabetes is further expansion of the safety
net, perhaps by ramping up food assistance.
That said, the difficult problem is not, I think, of deciding
how far to extend assistive welfare programs. If one frames the
problem as reducing noncompliance with treatment regimes,
it becomes apparent that there is another continuum, this one
based on how coercive the help with care management is.
Strictly assistive programs, such as housing or food subsidies,
pose no special problems here. Somewhat more contentious
would be efforts to make non-compliance more difficult, such
as a steep tax on sugary drinks. Our experience with cigarettes
shows that this sort of thing can make a difference. Much more
contentious would
The economic be efforts to directly
induce compliance;
problem should new technologies are
be clear enough rapidly making these
and can be put this more economically
way: It is almost viable. For example,
now there are pills
that know they have
been ingested and can
report this information.
to assist with Simpler devices like
housing than it is to fitness trackers also
could provide useful
pay for ER visits. information.
In my view, programs
in this last category are
morally questionable.
First, they don’t address
a fundamental problem: Being in unstable housing makes it more
difficult to comply with a medical routine. Second, they pose serious
questions about patient privacy and autonomy. Finally, they leave
unanswered the question of what to do if patients don’t follow their care
regimen. Would you deny them treatment? Hertfordshire in England has
started down this path, banning elective surgery for those who smoke
or fail to lose weight if they need to. This last point about privacy and
autonomy suggests an unusual wrinkle to old debates about welfare. In
the United States, it looks like housing assistance from the state is better
for individuals’ autonomy than other efforts to get them to maintain
treatment regimens.
However you resolve the moral issues, it seems to me that as long
as we accept a moral and legal obligation to treat those who arrive
at the emergency room with poorly-managed diabetes, the question
really is about what kind of dependence on the state is preferable and
how we want to pay for that dependence. At the end of the day, we
tend, as a society, to look at health care and healthcare expenses in a
vacuum. Berkowitz’s work provides an evidence-based reminder that
this is myopic.
Mecklenburg Medicine • March 2018 | 13