A Critical Review of a Flawed Recent EHR Quality Study

The Archives of Internal Medicine on 4/27/2009 published a study, "An Electronic Health Record—Based Intervention to Improve Tobacco Treatment in Primary Care," in which the investigators wanted to demonstrate an increase quality of medical care given to a group of patient smokers using certified electronic health records (c-EHRs). They did meet the end points of smoking status documentation and increased counseling assistance, but unfortunately the study had a big flaw, in that the main question is not whether c-EHR systems increase quality, but whether or not they are superior to paper records. This is a very important issue, since HITECH will in essence try to force physicians into not only purchasing expensive c-EHRs but to "significantly Use" them at an estimated cost of about $300,000.00 per physician over 5 years.

(ref:www.hcplive.com/mdnglive/articles/PC_Medicare_HIT_mandate)

This study is typical of other highly political and c-EHR biased publications. It was initially designed well enough to be backed by the National Institutes of Health and later to be published in the Archives of Internal Medicine, incorporating good elements of a scientific method like having a large enough patient group of smokers studied prospectively (132,630 of which 12,270 were smokers) which ended showing a significant statistical result, with P values < 0.001 in 2/3 parameters studied.

The study failed in 3 major areas:

1) The selection of the control group consisted only of historical controls of patients not targeted with any specific intervention. Instead, like the study group, they should have chosen a large prospective group of patients with similar characteristics and followed the quality measures using pen and paper methods, s.a. paper chart reminders, handouts, and reporting followed at the same time as the EMR treatment group.

2) It would be nice to study the effect of this increased documentation has on the overall quality of the patient visit. More electronic "paperwork" and less time given to the patient visit due to decreased reimbursements translates to decreased quality no matter how much HIT you throw at the problem. This extra reporting can eventually lead to worsening of outcomes as doctors spend more time in front of the computer rather than talking to the patient. In many situations, physician extenders will be used for not only the reporting aspect, but for the complete visit. Patients should have been interviewed, and charting reviews of other areas NOT studied, like time taken to perform a full physical examination should be reviewed. These decreased quality measures should also have been reported in this study.

3) They only proved that they were able to force providers to document more thoroughly with "electronic paperwork". They did not show any data about actual smoking cessation rates. You would normally wish for increased treatment concerning a single parameter to translate to decreased smoking and increased health, but this was not shown.

Their final statement thus has no real validity:

"Health care systems and clinicians may be able to provide more efficient, effective tobacco treatment by using health information technology to centralize their tobacco treatment efforts."

Other studies in the past that did compare EMR to paper records have shown no major difference in documentation with either modality (s.a ref: http://archinte.ama-assn.org/cgi/content/short/167/13/1400).

I am dismayed that this study was put together in this manner and that the Archives of Internal Medicine actually accepted it for publication. Its data will be used by politicians to further forward the HITECH political agenda even though the real issue of whether EMR is better than paper records has yet to be proven. It will end up wasting our limited health care dollars which could be better spent on improving health care for the poor and uninsured.

Al Borges MD

"HIT Realist"