i'm l o'clock or like to tell you about work the colleagues and i have

done to identify when large medical data bases are suitable places to conduct comparative effectiveness

research

the introduce some graphics

a user capital sigma to designate a provider it's as a nation side because doctors

add up data and come to decisions

a question mark a device a patient

in american cancer patient comes to something that he wants fix

a doctor brings experience in training

the doctor elicits information

the doctor uses all the available information

the country treatment decisions

adaptation can elsewhere a different doctor mitre brought different training and experience to bear because

to a different decision

another doctor might have made yet another decision

and so on

let's imagine for simplicity that all transcribers choose one of two treatments

the same patient stands an equal chance of getting treatment a or treatment be

depending only on which doctor you happens to visit treating community is in the corpus

our analysis screens population medical data for a coke voice

we propose that where we find it we found the suitable place for doing comparative

affection of this research

these graphs illustrate the technique with community acquired pneumonia

the x-axis or patients according to their probability of getting the treatment shown in that

this is called a preference score

the x-axis to pick seventy patients there are at each preference score level

the two lines each crap represent the patient preference scores for the two three bits

exam

the left and graph comparison use only fluxes in is it true motion at different

preference scores

at the midpoint of preference we have patients room that reading communities in perfect a

corpus

how to either side

we can skip channels about the vehicle points

so it should what lots of getting one treatment or the other

among patients receive closer mizuno marks of lexus and fewer fell into this range of

wealth of the corpus

if not patient characteristics predict treatment

then the to treatment groups will be alike

here's a situation that leave a fluxes and is it from us in patients

the two groups are very similar

okay treatment failure

the patients receive leave of losses and had fewer failures

they did better

this is finding deserve ad hoc research we believe that that's

our tools identified one treatment decision for which at least one prescribe or community seems

to be of divided opinion

i difference of twenty percent improvement failure rates would be important of true

the tools shown that so this population is suitable for comparative effectiveness research in that

many patient characteristics are balanced and so we're not confounded

but patients are more the totality of the recorded characteristics

two patients to appear to be similar with respect to the recorded characteristics differ with

respect to their unrecorded characteristics

proper research requires that all important covariance the measured accounted for

keepers of large medical databases can easily apply our tool to identify situations of the

parrot empirical a corpus

we can now when compared of effectiveness research might be successful

and when we should point

thank you for your attention