Sirs:
Zupancic, Richardson, O'Brien, Eichenwald and Weinstein are to be
commended for addressing the difficult issue of pre-discharge, hospital
monitoring of preterm infants. Not surprisingly, they found that when
babies are kept in the hospital for increasing periods free of symptoms,
the "cost" of each quality-adjusted life year (QALY) preserved goes up, as
fewer QALYs are added to the benefits column while the costs increase.
This can be attributed to either a decreasing likelihood of an apparrent
life threatening event (ALTE) happening after an increasing period without
an ALTE or to an increased maturity at the time of discharge. They
conclude that "monitoring lower gestation infants for a longer period and
higher gestation infants for a shorter period may result in better
outcomes and decreased costs compared with the status quo."
First, the authors do not make clear what is the "status quo." I
interpolate that it is 7 symptom free days of monitoring for all infants
born before 35 weeks of gestation who have reached thermal and nutritional
stability. This may not be true for more mature infants. Second, and more
importantly, while the results may be helpful to policy makers and
insurance companies, they are really not useful at the bedside. Should a
regulatory agency some day have the fortitude to say that only treatment
that yields QALYs at less than "X" amount of dollars will be funded, then
the results will be useful at the bedside. I am concerned that insurance
companies will use the data to justify the speedy discharge of infants
from the hospital without concern for the social and legal fallout.
What the practitioner needs are the data for risk of ALTE that the
authors used to calculate lost QALYs. We can discuss with parents the
likelihood of a significant event happening after each day of symptom free
observation. It would be nice to know if the likelihood becomes asymptotic
to the baseline after either a certain number of days or post menstrual
age (PMA = gestational age plus postnatal age.) At that point,
hospitalization is of no further benefit. Using PMA may be a better way of
looking at safe discharge across wide ranges of gestational age. We could
then help parents weigh the benefits of continued hospital stay versus
discharge home.
I doubt that many parents can grasp the significance of cost of a
QALY. In fact, if I were a parent, I would ask that my baby stay in the
hospital long enough to be sure he will be safe at home, or provide a way
to protect him at home. The day or PMA at which the risk at home equals
the risk in the hospital could/should be the day of discharge. I would
consider the cost irrelevant if there were a safe way to attempt protect
my child. (Hence the (? excess) use of home monitors.)
I agree with the authors that equipoise does exist about the duration
of in-hospital monitoring. In such a situation, the safest (for the
practioner) approach is a clinical trial with fully informed, consenting
parents. My dilema in designing such a study would be our inability to
predict ALTE's at home, and even to define a "significant" event in the
hospital or at home. It may in fact turn out that in-hospital monitoring
does not actually improve QALYs. Furthermore, while home care is certainly
better in the long run than hospital care for a "well" child, I do not
know the benefits of discharge that is one week earlier than otherwise
planned. Thus a risk-benefit analysis for parents (as is required for all
research consent forms) is currently difficult to do.
Kenneth Harkavy, MD
eLetters: ![[Read eLetters]](/icons/misc/sectionDOWN.gif){kind=icon}
