Two goals have driven many of the
reforms introduced in academic hospital procedures in
recent years. One is to ensure continuity of care by
making sure that, whenever possible, the team that admits
a patient is the same one that follows up with inpatient
care. The other is the imperative to reduce junior doctors'
working hours.
The main reason for this second
change, of course, was to reduce the incidence of errors
caused by fatigue. But the caps on working hours, now
typically set at 80 hours across the US and Canada,
might be impinging on patient care in other ways.
That's one possible interpretation
of a new study of the effect of medical teams' workloads
on patient care in a large teaching hospital. The research,
published in the January 8 issue of the Archives
of Internal Medicine, suggests that teams harassed
by numbers of patients rather than numbers of hours
may be the ones making mistakes on admission.
ONLY
SO MANY HOURS
Dr Michael Ong, lead author of the study, argues for
"fairly strict limits on how many patients each team
can admit on a given day." Stretching time to fit everything
in may no longer be an option. In the past, he said
"they could stay later, try to get as much work done
as possible. But that's something that's not possible
with the 80-hour work week."
Following 5,742 adults admitted
to Moffitt-Long Hospital, a 525-bed tertiary care centre
in San Francisco, between 1998 and 2001, the researchers
found that high workloads on admission days translated
into longer stays, higher costs and maybe worse mortality.
Each additional admission by a
team of interns and house staffers on a patient's admission
day increased that patient's length of stay by an average
3.09% and total costs by an average 2.31%, and increased
risk of dying by 9%.
Each team also had a burden of
inpatient care too, yet strangely, extra workload in
terms of inpatients did not translate to compromised
care. In fact teams with more than 15 inpatients in
their care produced faster discharges and lower costs
without higher mortality.
Dr Ong believes this counterintuitive
finding can be attributed to several factors. First,
while teams can't control the number of patients they
face each admissions day, they can make more time available
for inpatients, for example by skipping conferences,
on other days. Also, teams with an obviously heavy inpatient
burden can attract outside help, like social workers,
and can be more focussed on discharge, which is no bad
thing, says Dr Ong, all else being equal.
EXPECT
THE UNEXPECTED
While the hospital adapted well to varying inpatient
burdens, it had no comparable mechanisms to adjust for
unexpected variations in admissions, notes Dr Ong. He
suggests that "training programs could restructure their
admission services so that, for example, there could
be more teams available on given days to match the potential
loads that they may see over the course of a year."
He stresses that his findings come
from just one teaching hospital and may not be universally
applicable. For example, Canadian hospitals have generally
adopted 80-hour working limits and a similar continuous
care model in which training teams have "admissions
days" when they take patients on and other days which
are devoted solely to inpatients. But the teams themselves
can be quite different. Queen's University's three teaching
hospitals, for example, have recently re-organized into
teams of three to four junior housestaff, with one or
two senior housestaff and an attending physician. An
optimal workload for such a group is reckoned to be
20-25 patients.
There were two kinds of teams at
Moffitt-Long Hospital. Each comprised an attending physician
and a resident physician, but some had two interns and
others only one. The three-person teams had a daily
admission cap of five patients, while the four-person
teams were capped at ten.
While the four-person teams can
clearly admit more patients per doctor under this system,
there may be a price to pay, says Dr Ong. "It appears
that if you're on a team that had one intern and one
resident, that the likelihood of your coming back to
the hospital is less. Our subanalyses suggested they
were getting a higher level of care. If it's just one
intern alone, the resident is more likely to step in
and help. On a two-intern team, it may be that the resident
doesn't step in, because there are more people around."
One widespread belief about teaching
hospital admissions appears not to be borne out by the
study: the July phenomenon. "This is the idea that patients
who come in during the summer months, when trainees
are new, may get worse care," says Dr Ong. "Other findings
have been equivocal about this, but we didn't see a
July phenomenon. We all know that if people recognize
a potentially unsafe situation, they'll work harder
to make sure nothing untoward happens. I suspect these
teams are working harder during those months."
|
