May 2017

Sun Mon Tue Wed Thu Fri Sat
  1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30 31      

« NHS: Doctors have no confidence in NHS database, says BMA News | Main | Standardization in a Multi-Physician Clinic - Office Layout »


Jim Busser

The principle inadequacies that I have seen have been the failures in such systems to anticipate and/or provide support for:

1. The care relationships... whereas most systems will support one doctor within a group to be designated the patient's customary or most-responsible doctor, these systems nearly invariably designate the individual to be "the family doctor" and therefore make no provision to be able to store and identify a primary care physician that exists outside of the practice... I would think that even in an EMR that was supporting GPs, it would make sense to be able to preserve that a given patient has a second or other primary care physician outside the group... think "university students" or patients who work at a distance from home.

2. Directionality... referrals in some EMRs are always assumed to be from a GP to a specialist. There is often a poor concept of being "referred to". GPs can serve as consultants if they have areas of specialty, especially in smaller urbs, and it isn't all that unusual for specialists (especially general internists, but subspecialists as well) to other specialists. Failure to support both ends of this directionality is IMO egregious. I have found maddening the number of systems that do not well-support that a patient has been referred by, and/or to, several individuals involved in care and, likewise, do not support flagging which of the relationships are active, versus not to be reattempted for this patient! :-)

Apart from the above, which need homes in the structural foundations (the data schema), I think the backend (storage) could potentially be the same whether the practice be primary care or specialty.

The other key differences will be in the extent (the range and nature) of the kinds of content with which any particular specialists will want to see or work with. This will affect the views (juxtapositions of various content of interest) and the workflows, and can require moderate (but not total) re-working of the user interface (UI).

Billing is, in cases, very different. But here again, it's mostly an issue of the UI over and above the subset of information that needs to be dealt with.

Michael Varelas

Accessibility of components of my EMR by other health care providers is an area that may as well be different between specialist and GP. Even within my own practice, there are records (e.g. consultation reports) that should be available with certain access, as well as other documents (e.g. reports of tests performed) that may have a wider access. I report on studies performed in the hospital that the hospital wants available within its own network, and ease of this kind of connectivity would be very helpful.

Jim Busser

Michael's comment opens a complex area. I wonder if it is better dissected-out.

Where he discusses reports that he generates within the hospital, which the hospital would like within its own network, it sounds like he is discussing the hospital's version of an EMR, more-conventionally referred-to, by corporate IT people, as an "Health Info System" (HIS).

Hospitals and health authorities are struggling with whether and how to make parts of their info variably available.

The issues with which they must contend include:
1) what if the doctors involved in care don't have hospital privileges? what about their medical office assistants?
2) how, in the absence of a patient-provider registry, does one determine who should get access, and to which parts?
3) to what extent should this be a priority, given the competing wholescale failure to develop a transmission model in which the hospital could "send" results to those doctors who could have received them
4) if the hospitals are to support "lookup on demand", should they do so on an individual basis, until some repository system can be made available for consenting patients?

I see this issue of hospital and other publicly-managed facility data as needing to align with, but maybe best kept distinct from, the focus of Canadian EMR blog, which I suspect is electronic care systems that cater to individual *groups* of practicing physicians - but maybe I am wrong here :-)

On the question of sharing across *non-hospital* EMRs, I think we want to recognize that specialists and GPs who care for the same patients have a need for each other's information. I'll therefore *re-emphasize* the need to keep such (bi) directionality near the centre of design.

Unless and until some suitable "store and forward" system is able to be put into place — such as was piloted in the BC eMS project ... see — I think the only practical solution for sharing electronic info anytime soon is encrypted email, or other secure information transfer, with the information packaged in a simple document form. I suspect that to entertain managed access of groups and individuals *directly* into each other's EMRs could be to entertain a nightmare.

Martin Pusic

I think that the idea of an EMR for specialists and one for family doctors is not necessarily the way to think about it. Those might be "perspectives" on a core of patient data just as there might be separate perspectives for nurses, business personnel, the health authority, college etc. Design of the EMR should take into account the possibility of these different perspectives.

Tom Kinahan

I am a specialist and helped design an EMR designed for specialists. In designing the solution we recognized that there were key differences between specialists and generalists;

1. On the "health journey" of a given patient, specialists' encounters represented "waypoints", where a data snapshot is required, as compared to a longitudinal record more typically required by generalists. Dr. Varelas points out that specialists often need to view a subset of the generalists' EMR, not the whole database, the view being limited as appropriate by the generalist

2. Typically generalists EMRs are more data storage intensive and less complex decision intensive, where specialists are the exact opposite. However, as Dr. Pusic and Dr. Busser point out, we are more alike than different, so we really don't divide up this conveniently; more than that, EMRs are multifunctional and may be used in different ways by different physicians. In general, EMRs that support a given workflow will be the most attractive to an individual doctor; we have done well with an "open architecture" of sorts which allows customizing of the EMR to the physicians' needs, with a relatively large degree of specific input by the physician; the "COTS" (commercial off-the-shelf) EMRs may be best applied to the generalist if it reflects an average or typical model of generalist practice, but what is that? What does that look like? I think the number of solutions would equal the number of adopted physicians if we accept that model.

So how do we resolve this? What I think is most important is that EMR reflect the workflow of most of us, but be customizable for all of us, and be complete, i.e. include all of the features that help to run all of the clinical and administrative functions of your office, including interoperability; I don't believe it will be a technical impossibilty to access each others' EMR if managed well, even if EMRs are from different vendors, so long as communications standards and permissions are well managed.

Most importantly, EMRs must provide the promise of clinical utility at an affordable price (not just cost, but both related to implementation and any ongoing issues related directly to using an EMR), or the average physician, specialist and generalist alike, will continue to decline to implement (current EMR adoption Canada-wide still sits at a measly 26%, increasing at a rate of about 2% per year).

Ali Husain

The problem with conventional EMRs is that they are too complex and 'sophisticated' for most surgeons needs. From a surgical point of view, we don't order or follow up anything like the number of blood tests as our medical colleagues. We also don't write as many prescriptions. I have tested 4 or 5 EMRs from various companies and while they are very well made, they actually slow you down not make you more efficient. At least for a surgical practice. I have put together my own little DIY EMR using MS Outlook for scheduling and MS OneNote 2007 for charting. It works extrememly well and is very efficient. It is also very low cost. $500 as opposed to several thousand.

Don Smallman

I took the plunge last year when expanding my solo office to go electronic. As an ophthalmologist I was concerned that the EMR would not meet my needs, as I was unable to find an economical ophthalmology-specific EMR. By developing my own templates it has gone well bu has been labour intensive. I agree with Ali's comments that most surgeons don't need/want the complexity of the GP-directed EMRs but I have found the workflow aspects of generating and tracking consults, letters, referrals, messaging etc invaluable. And the competition among the GP-derived EMR companies has driven the costs of the programs down considerably since the last time I looked into it 4 years ago.

James Holmlund

I agree with Dr. Busser that EMRs generally fail to reflect salient care relationships and directionality of referral at the level of the data model.

I think Dr. Kinahan is correct in identifying some key differences in perspective that specialists and generalists bring to 'the clinical data,' although I would invite him to put the matter a little differently. Consider, for example, the data that is relevant to a psychiatric 'one-shot' consultation vs. the data that becomes relevant in on-going psychiatric care. (Substitute 'pediatric' or 'geriatric' for other instances of what I think is needed.)

Dr. Husain's point emphasizes key differences in work-flow between generalists and some specialties--and also helpfully suggests that the best EMR solution would be more like a kit, perhaps based on the entire MS Office suite)that would allow the user to customize interface, while 'automatically' handling scheduling, billing, filing, and archiving tasks.

It still seems to me that the industry is busily trying to 'build a better mousetrap' rather than trying to create something new.

It is one thing simply to convert paper charts to electronic charts--a case can be made on the basis of storage and staff costs. But at the end of the day, it's still a chart, and bells and whistles like electronic prescribing and electronic labs don't really add that much to patient outcomes or physician understanding (unless the original paper chart was impenetrably disorganized in the first place).

What will increase adoption is the development of a must-have information system that will show me what effect my interventions are really having--i.e., will generate practice-based feedback and 'research'. It is at this level that the difference between generalist and specialist EMRs stands out--the questions are different.

The comments to this entry are closed.