they traditional or pharmacoeconomic end points.
The handling and analysis of pharmacoeconomic
data should follow good clinical practices (GCP)
guidelines. Data collection instruments need to be
selected, or created and incorporated into case
report forms, just as for any other end point. Data
analysis plans should be created prospectively. The
statistical analysis plan should be prospective, and
should help put the pharmacoeconomic measures
in the context of other properties of the test med-
ication (Table 23.3). Are they included to test a
hypothesis or is this a hypothesis-generating study
for the pharmacoeconomic measures? Is thegoal to
evaluate patients, discriminate between patients or
predict how patients might act? The type of data
collected should drive the level of analysis (con-
tinuous vs. categorical data). If there is an investi-
gators’ meeting for the study, the pharmac-
oeconomic components should be presented at
the meeting so the investigators and/or the study
coordinators fully understand their role in data
collection. As the study is ongoing, appropriate
levels of monitoring should be conducted. Queries
that arise during the study and reconciliation of the
data afterward should be handled in the same man-
ner in which clinical queries and data reconcilia-
tion are handled.
23.7 Reporting and publications
Most companies have some form of standard oper-
ating procedure by which they generate clinical
study reports. Pharmacoeconomic data should be
handled and reported in the same way. In some
cases it may be appropriate to issue the pharma-
coeconomic component of a study as an appendix
to a larger clinical report. This will depend on the
level of pharmacoeconomic involvement in the
study and how closely related the end points may
be to the pathological measures. If there were just a
few pharmacoeconomic measures that were being
tested, an appendix to a clinical report might be
appropriate. In contrast, for example where recov-
ery from anesthesia is measured by ‘street fitness’
(the humanistic outcome) and neurological mea-
sures of balance and coordination (the physiologi-
cal end point), then it could be cogent to report
these two types of data together, and to examine
how well they correlate; this would not be suited
for an appendix for the humanistic data.
External reports are most likely going to be
manuscripts submitted to peer-reviewed journals.
Placement of pharmacoeconomic articles in non-
specialty journals is important but difficult. Some
editors do not understand the intrinsic properties of
pharmacoeconomic data, and some reviewers will
blindly apply statistical constraints that are inap-
propriate or not valid to humanistic outcomes (e.g.
power calculations to measures of the adverse
effects of drugs on QOL measures).
The basic principles of scientific writing and
reporting apply to pharmacoeconomic research,
and little need be said here. The structure of the
paper is the same (Introduction, Methods, Results,
Discussion, etc.). It is important to be consistent
and appropriate in the use of terminology (e.g.
‘costs’ is not synonymous with ‘charges’, and
cost-effectiveness is not a cost–benefit analysis;
Sanchez and Lee, 1994). New mediums such as
the Internet offer new possibilities for publication,
dissemination and debate (Medical Outcomes
Trust, 2001 (www.outcomes-trust.org); American
College of Clinical Pharmacy, 1996).
It must be said that how such information gets
disseminated is controversial in the United States.
A good recent example is an investigation of ato-
vaquone versus i.v. pentamidine in the treatment of
mild-to-moderatePneumocystis cariniipneumo-
nia. This report included a decision tree to estimate
the costs and cost-effectiveness of atovaquone
versus pentamidine for cotrimoxazole-intolerant
patients (Zarkinet al., 1996). Clinical outcomes
were based on data from a previous phase III RCT,
which compared the two medications. Economic
outcomes were based on treatment algorithms
derived from discharge data, published reports
and clinical judgments by the co-authors. The clin-
ical data were from a randomized, double-blind
study. A sensitivity analysis was conducted. The
major conclusion of the study was that there were
significant cost savings to be had from treating
Pneumocystis cariniipneumonia on an outpatient
basis. An FDA representative, during a platform
presentation of this paper, even indicated that these
data could be used in promotion.23.7 REPORTING AND PUBLICATIONS 299