The population health benefits of EMR implementation done right

I’m back at it again, talking about my continued love/hate relationship with EMRs. From my conversations with doctors at different hospitals in our region, it seems that most docs appear to be falling into the “hate” column. Meanwhile, I’m still chugging along with the EHR that’s been installed in my office. And while it works just fine for the needs of a 3 physician single-specialty outpatient practice, it’s hardly the type of technology that, by itself, can change medical care for the better for a large number of people.

A recent study challenges that notion. In a study published in CHEST, researchers in England sought to determine if inhaled steroids are a risk factor for pneumonia among asthmatics. It has already been shown inhaled corticosteroids are associated with an increased risk of pneumonia among patients with COPD. To determine this they looked at a database of medical information known as The Health Information Network (THIN).

In the UK, EMRs have been in use for years, and general practitioners are encouraged (but not required) to participate in THIN. When a general practice elects to participate in THIN, software is installed in their EMR which runs in the background. The program collects data, while de-identifying it. The anonymized data is then uploaded to THIN, where approved researchers may have access to it.  There is no cost to the practices for participating, and in return for their participation practices not only receive in depth practice metrics, they also receive a percentage of any research revenue generated from the use of the THIN data. At the time that the study was conducted, the database contained data from 9.1 million patients.

But back to the question at hand. From a cohort of 359,172 people with asthma the researchers were able to identify 6,857 people with pneumonia, along with 36,312 control subjects.  They were thus able to find a positive correlation between inhaled steroids and pneumonia.

Even if EMRs aren’t necessarily connected in U.S. as of this writing, this study shows what can be accomplished when thousands of different data points from patients are taken from hundreds of different systems. The aggregate can make a potentially large and useful data set that can be used to help guide care both for individual patients and for large populations. While it seems that we’re still ages away from anything this integrated in the U.S., I’m still hopeful that someday, eventually we’ll get there.

Deep Ramachandran is a pulmonary and critical care physician, and social media co-editor for the journal CHEST. He blogs at CaduceusBlog and ACCP Thought Leaders, and can be reached on Twitter @Caduceusblogger

Comments are moderated before they are published. Please read the comment policy.

  • Dr. Drake Ramoray

    I don’t think anyone is debating that more precise coding with ICD-10 or even data dredging EMR’s is effective for public health research. The real issue is why is the most expensive to train, most highly qualified specialist in patient care, the doctor required to be the data clerk for these studies (even without considering the issue that not only does it not pay but permanently reduced productivity and time to care for patients)?

    • RES

      Note the difference between the UK and US. The EMR is not a billing tool in the UK, and so there is much less burdensome overhead/auditing/coding game gotchas. It’s not obvious to me that the UK physicians themselves deal with this level of detail. On the visits I have made only the night-call physician made notes himself when I took my father to him late one weekend night. However, he could access my father’s doctors records as though they were his own.
      Also note that what the researchers have access to is anonymised.
      The same technology was used in the UK aspirin study, and probably in the incentive pay for GPs to manage various expensive risk groups more intensively to reduce hospital utilisation.

Most Popular