5. Limitations
While we attempted to make considerable efforts to
design a sound research, the study had several limitations.
Although a high inter-rater reliability was established for
the measurement of bleeding and hematoma, data for
these two variables were obtained by observation and
palpation techniques, where the presence of bias is
probable. Second, as we used a combination of methods
for positioning patients in the experimental group, it was
not completely clear that which intervention (HOB
elevation, changing position, or early ambulation) produced
more effect on the outcome variable. However, we at
least can attribute the variations in the outcome variables
to the combination of interventions. Moreover, the
assessor was not blinded to the study.