Now Reading: Frequency of Failure to Inform Patients of Clinically Significant Outpatient Test Results

Casalino LP, Dunham D, Chin MH, Bielang R, Kistner EO, Karrison TG, Ong MK, Sarkar U, McLaughlin MA, Meltzer DO. Frequency of Failure to Inform Patients of Clinically Significant Outpatient Test Results [Internet]. Arch Intern Med. 2009 Jun 22;169(12):1123-1129.[cited 2009 Jun 26 ]

This is a wonderful and well-timed study that has significant implications in the era of the Electronic Health Record and the Personal Health Record. As well done as it was, I would have loved a section of inquiry to be added about “the impact of patient and family access on test result notification.” Read on…

It’s impressive that in 2009, believe it or not, there really aren’t firmly established processes for handling information about test results. A lot of what is done today is bred from custom, such as the infamous “no news is good news,” which the authors found was the protocol in 8 out of 19 medical practices studied. Everyone who likes this approach to test result notification, please raise your hand….

With that background the study team started at a very low baseline, thinking about what kinds of test results patients should be informed about, in what period of time they should be informed about them, and then analyzed medical records (5305 in all) to see if theoretical best practices were carried out., and about 7.1% of the time, on average (up to 26.2% in one Academic Medical Center practice), information to patients was not furnished about their abnormal test results. We can imagine what that might mean in a practice whose policy is “no news is good news.”

The authors looked at the impact of having an electronic medical record and found that practices with a “full” electronic medical record were no more likely to have gaps than one without IF they had a good process for managing test results. So, process and workflow trumps technology in this case.


What’s missing in good process?

So, the number of abnormal test results in this study not communicated to patients is alarmingly high. At the same time, I immediately drifted to what’s missing in the process. The authors listed these steps as a good way to manage test results:

1. All results are routed to the responsible physician
2. The physician signs off on all results
3. The practice informs patients of all results, normal and abnormal, at least in general terms
4. The practice documents that the patient has been informed
5. Patients are told to call after a certain time interval if they have not been notified of their results.

Maybe this is good practice today, but what do our patients and families want in the era of the personal health record and full transparency (73 cents style)? How about this:

Good process for managing test results, patients and families at the center

1. All results are routed to the responsible physician and the patient and their proxies, if specified, at the same time
2. The physician and the patient and their proxies, if specified, sign off on all results (in a current PHR installation, this might mean verification that the patient has viewed the result…read on)
3. The practice informs patients and their proxies, if specified, of the meaning of all results, with specific recommendations to be made based on the information
4. The practice documents the shared decision made by the responsible physician and the patient based on the information obtained from results
5. Without 1-4 above, the practice reaches out to the patient via the most appropriate means (letter, telephone, secure e-mail) to achieve notification and shared decision-making.

If we think about it – in the era of the personal health record, do we really want to tell patients if they haven’t heard something within a certain time interval, they should call us?

Do we really want to continue a “no news is good news” policy, at the risk of “no news” meaning 7.1% of the time someone may be hurt in the process of care?

I think it’s important to remember that the ultimate reason a test of any kind is ordered in health care is for one reason – “to reduce uncertainty.”

It would be great in a future study to analyze the impact of patients having access to their test results in real-time or near-real time, to see what the rate of failure is, and also dig deeper, at the rate of understanding of what test results mean. This is the sweet spot for physicians and nurses, who excel at using test results to reduce uncertainty in the context of a patient’s overall health.

In terms of whether or not the new/improved “Good process” is more time intensive or not than the regular “Good Process,” I don’t think it is more time intensive. I think this is a great item for discussion in the comments. Let’s talk about the cost-benefit of doing things differently.

It’s worth noting that in the first quarter of 2009 alone, 5,078,442 test results were viewed by Kaiser Permanente patients and/o their proxy individuals on KP’s My Health Manager personal health record. In many of those instances, the test results were delivered to the patient at the same time as the physician. That’s a lot of experience both to tap into, and to understand that the old process is already changed forever for lots of Americans and the teams who care for them.

With thanks to the authors for a timely and useful investigation into an area of health care where we all want to improve.