|
July
2003
Good At-Bats
It’s July, the midway point of the major
league baseball season, and one might expect that teams like the
New York Mets, which spend huge amounts of money for the best talent,
will be playing a lot better than teams like the Oakland A’s,
who spend among the least.
One would be wrong.
Why am I talking about baseball? A colleague brought to my attention
a best-selling book about the subject that draws parallels between
the field of dreams and the fields of health. Both rely on evidence
to explain and predict outcomes, but neither uses it particularly
well.
The book is “Moneyball” by Michael Lewis, and it explains
how teams with little cash to spend on players can consistently
beat teams that spend a lot. The difference lies in better evaluation
of talent through use of a wide range of different types of evidence;
sort of a “research to practice” issue.
Most people
know what a batting average is. It is simply the number of safe
hits divided
by times at bat — the single statistic
that becomes a batter’s identity. But actually, batting average
is a highly overrated statistic and tells a manager very little
about how the batter contributes to his team winning more games.
It is other numbers hidden in plain sight — on-base percentage,
slugging average, number of walks — that more reliably predict
and produce the desired outcome of winning.
A good at-bat
is not necessarily one in which the batter gets a hit. It often
consists
of a batter refusing to swing at bad pitches
and fouling off marginal pitches until the opposing pitcher is
exhausted and frustrated and begins to miss his target. The result
can be a boring, but effective, base on balls — or, better
yet, a home run.
In health, a good at-bat can mean finding cures, but it also means
using what we already know about prevention, epidemiology and clinical
practice to answer questions that are often dismissed as boring
science: Which people are most at risk for disease? What policies
most effectively prevent it? How do we ensure that people take
needed drugs as recommended?
For example, finding a cure for diabetes is critical, but it is
just as critical to understand how to target effective prevention
measures, how to help people manage mild disease and how to increase
the appropriate use of drugs and technologies so that the millions
of people who have diabetes will live long and productive lives.
In both baseball
and health, the best results come when we look systematically
at
the many diverse factors that contribute to the
goal. To be sure, baseball and medicine offer inexact parallels,
but our ultimate goal is the same as those who are picking the
players for a winning team — using all the evidence.
|
