X hits on this document

73 views

0 shares

0 downloads

0 comments

19 / 39

are multiple observations for a regimen, either because a manufacturer conducted separate trials of the same regimen, or because a regimen may have been the treatment group in one clinical trial and the control group in a subsequent trial. In these cases we calculate the mean attributes across the separate observations. Where necessary, we supplement the package insert information with abstracts presented at oncology conferences and journal articles.

The attribute information is summarized in Table 1, organized according to the year when each regimen was introduced. We record three measures of a regimen’s efficacy: the median number of months patients survive after initiating therapy; the percentage of patients who experience a complete or partial reduction

in the size of their tumor (i.e., the response rate); and

months (across serious state.16

patients For all

in the trial) before their cancer three of these measures, higher

the mean advanced values are

number of to a more associated

with superior health outcomes. We also record whether a capecitabine tablet, which should make the administration

regimen contains the of the regimen more

convenient for a patient, a second-line treatment.

and whether the regimen is approved (and was tested) as Efficacy measures for second-line regimens will generally

be worse than those for first-line regimens because the patients’ to be more advanced at the beginning of the clinical trial and the was not completely successful.

cancer is likely first treatment

We also collected data on the percentage of patients in phase 3 trials who experienced either a grade 3 or a grade 4 side effect for six separate conditions: abdominal pain, diarrhea, nausea, vomiting, neutropenia, and dehydration. Al- though many more side effects are recorded for most regimens, these six were consistently recorded across the 12 regimens in the sample. Side effects are clas- sified on a standard one to four scale, with four being the most severe. Higher values for the side effect attributes should be associated with worse health out-

16Cancers are classified into four stages, with higher numbers indicating that the cancer has metastasized beyond its initial location.

19

Document info
Document views73
Page views73
Page last viewedFri Oct 28 22:30:41 UTC 2016
Pages39
Paragraphs850
Words10609

Comments