| |
Our
basic approach to quality improvement is to build knowledge
and insight through repeated trials of plausible improvements
by teams of workers. Each team tackles a focused set of
problems associated with their usual jobs. They make changes
to the system and test those changes, usually on a small
scale, in order to understand how the system responds prior
to implementing wide-spread or permanent changes. Learning
progresses as the team modifies previous changes in response
to outcomes, including increasing the scope of the improved
practice, or adding new changes.
Viewed
in the usual categories of study design, a quality improvement
project is prospective and interventional. Both researchers
and skilled quality improvement practitioners recognize
that such studies risk biases in cohort construction, imposition
of intervention, and measurement of outcomes that threaten
internal and external validity. Research methods that minimize
such risks, such as controls, blinded participants, and
randomization, are often not possible when large or complex
care systems are the targets of improvement. Thus the practice
of "pragmatic science" for the goal of quality improvement
is uniquely challenging and requires care in execution in
order to maximize the validity of knowledge obtained and
the chances of successful change.
Six
such challenges are described here. Click on a challenge
to learn more.
|