Associate Brad Sokol sent me a link to a paper on Health Affairs, How Common Are Electronic Health Records In The United States? (abstract)(full pdf). It turns out there are many answers to this question. (The terms EHR and EMR are interchangeable in this post.)

Based on the surveys reviewed in this study, the EMR adoption rate for physician offices in the U.S. is a whopping 24 percent. The following table is excerpted from the study.

Range from medium- or high-quality surveys (%) Best estimates based on high-quality surveys (%)
EHRs in physician offices
Solo practitioners
Large physician offices
17 - 25
13 - 16
19 - 57
EHRs in hospitals
Not available
4 - 21

There are no publicly available studies on hospital EHR adoption. That's such an active commercial market; there are private market research firms who provide their studies to vendors for a fee. Perhaps a reader can share with us some hospital EMR adoption numbers. Large practices are defined in one study as having 20 or more physicians and 39 % adoption; another study defined large practices as 50 or more docs with an estimated 57% adoption.

Physician practice size is an obvious variable, with adoption going higher the larger the practice. The other big variable is the extent of adoption. According to the IOM, there are 8 key capabilities of an EHR. If adoption were limited to physicians who had implemented at least 4 of the key capabilities of an EHR, adoption rates would fall to only 9 percent. As with many new markets with extensive solutions, features are being implemented gradually.

Growing EMR adoption means that medical device connectivity is
becoming a bigger issue for medical device vendors who sell into the
physician practice market. When physicians adopt an EMR, they want the data produced by their devices to automatically flow into their electronic records. Manually entering data is inefficient and error prone. Just as in hospitals, data manually read or copied off medical devices can be entered into the wrong patient's record, the data can be transposed, or in some cases never entered at all. Common devices include spot vital signs monitors, SpO2 monitors, ECG, holter, ultrasound, spirometry, point of care diagnostics, and other categories of devices.

Before EMRs, physician decisions regarding the point of care revolved around medical devices - the diagnostic confidence of data, ease of use, and cost were chief considerations. Practices who have adopted (or realize they will eventually adopt) an EMR consider connectivity with their EMR a new key criteria for device selection. Physicians in the process of adopting an EMR frequently replace existing devices that cannot be integrated into their EMR. To device vendors it seems that EMRs have replaced devices as the new king of the hill in physican's minds - and they're right.

Pictured right is the Burdick Sensaire spirometer, which sports serial connectivity out of the docking station. (That's not an antenna on the right, it's a stylus for the device's touch screen.)