Study Limitations
While this meta-analysis provides robust evidence for the utility of AUC
QI interventions and incorporates studies with a diverse group of
providers, we believe that there is still not enough information to
assess whether the results can be generalized to physicians across
specialties and whether there are inter-specialty differences in the
effectiveness of the interventions. Majority of the studies utilizing
educational and feedback tools were targeted at three particular
specialties: cardiology, internal medicine, and primary care physicians.
However, in true clinical practice echocardiograms are ordered by
physicians across all clinical specialties as well as by mid-level
providers. Additionally, given that the majority of included studies
using educational tools targeted a limited group of physicians, it is
unclear how feasible it would be to implement similar QI interventions
in larger, multi-provider settings.
One aspect that limits generalizability of this meta-analysis is the
heterogeneity seen across studies. This is likely a result of
differences in the inherent design and implementation of QI protocols
across difference studies, thus the heterogeneity was expected. We
addressed this via subgroup analysis by type of QI intervention and by
duration of follow-up. This was effective in reducing heterogeneity in
the decision-support tool subgroup. We further addressed this using a
random effect model for analysis, which accounts for inherent
differences in methodology and design across included studies.
Lastly, all studies included in this meta-analysis were performed at
academic medical centers, making it difficult to generalize the results
to the community setting where a significant portion of echocardiograms
are ordered and performed.
The application of AUC-based QI interventions appears to be effective in
reducing rarely appropriate echocardiography testing. This meta-analysis
has shown that educational tools and decision support tools are
effective QI interventions regardless of whether a feedback tool is
incorporated. Further studies are needed to elucidate the effectiveness
of different formats of feedback tools as well as the generalizability
of AUC-based QI interventions in the community setting and across
medical specialties.