the devil you know

I've got a patient right now with a big white spot on his lung, coughing
up blood, and losing weight. He's been gradually getting worse over the
past few weeks. We are having fairly regular doctors' rounds every
Friday where we discuss cases that we're not sure what to do; one of the
things that we find is that we often make decisions not on necessarily
what we think is going on, but what we think we can treat. So the man
with the spot on his Xray; could be cancer, could be TB, but we can
treat the latter and not the former, so we give him a diagnosis of TB
and treat him as such. Same with the woman with the swollen lymph nodes
in her leg. Or the man with blood in his CSF -- could be a bleed, or
could be encephalitis, but we can only treat the second one.
This obviously leads to overuse of some resources on patients that don't
have the diseases that we are treating them for, and skews our success
rates for treatment of certain diseases. It also makes explaining
prognosis to patients and families difficult (e.g. "I'm going to start
you on some medicine, but I don't really think it will help, since I
don't think you actually have TB"). I struggle with the right way to
approach these patients, however, because I think that false hope and
wastage of resources aren't really good things, but missed diagnosis of
treatable disease is also a bad thing. So we usually err on the side of
treating them.