EHR Data Will Help Advance Comparative Effectiveness Research

Comparative Effectiveness Research (CER) has been an ongoing effort to identify best-practices for health care, based on clinical evidence. The AHRQ defines comparative effectiveness as research “designed to inform health-care decisions by providing evidence on the effectiveness, benefits, and harms of different treatment options.”

However, the vision for truly getting good CER data, digesting the findings, and turning those findings into usable policy, is hampered by limited data sources. Mostly, such data comes from clinical research studies – which take time, funding, and long time intervals to complete – or from other data sources such as claims-based data from payers. Given these limitations, the acceptance of CER data has met with some resistance from clinicians and other audiences – some of it is simple resistance to change, and some of it is skepticism of the data in the first place.

Confounding the issue of what to make of CER data is the fact of different professional organizations taking different, even conflicting, stands on recommendations for treatment – such as when to begin and how often to do screening mammography in women. Clinicians on the front lines are confused by this, which casts a cloud on the utility of CER data research in the first place.

How EHRs may change things
Unlike the tradition of CER research in the past, which relied on carefully-designed studies that involved relatively small numbers, modern EHRs have the potential for generating vast amounts of clinical information. The information is observational, though prospective studies can also be done – identifying candidate patients via EHR data, and then tracking their response to study-based interventions using the EHR.

The federal government recognizes this potential, and has created an approach that attempts to gather clinical data from locally-installed, stand-alone EHRs – the Query Health project. It is unclear how successful this approach will be, and one can imagine that (at least at first) the main participants will be large institutions (academic health centers, large hospital systems, and integrated delivery networks) which use large-scale EHRs. Smaller practices may not be very aware of this initiative, and may not have the ability make a connection and actively upload their data.

However, web-based EHRs (such as Practice Fusion) may achieve a similar goal more quickly. By combining EHR data of practices everywhere – particularly smaller, independent community practices not in hospital-driven settings – clinical data is already being aggregated.

The use of such EHR data for clinical effectiveness research has a staggering potential, when one stops to think about it. The data sets are very large, and the validity of the data is much more real-time than billing-based data. Both retrospective observational studies, as well as prospective studies can be carried out on scales not previously imagined.

Further, by implementing widely used web-based EHRs (and their patient-facing PHR portals), clinical research can be broadened as well. Historically, clinical studies were focused around academic medical centers, which conducted the studies, and tended to involve patients easily reached by the academic centers. With widespread web-based EHR use, the possibility of enrolling patients in widespread geographies – sending them materials by mail, having phone outreach, utilizing resources local to the patients – becomes possible. Clinical research companies following this novel model are starting to emerge.

Conclusions
Comparative Effectiveness Research is set to undergo a dramatic increase in the depth and breadth of data available for improving our understanding of disease, as well as the best and most effective ways of treating those diseases. EHR data, if collected and centralized, can be leveraged in ways that will foster health care improvement more quickly than we have seen in the past.

The federal government is looking at ways of querying data from separate stand-alone systems, which will likely draw data from large hospital and academic settings, and this will be a great step forward. Web-based EHRs already have centralized such data, and may yield insights even more quickly, which will help move the field of evidence-based medicine forward like never before seen.