Healthcare Innovation and The 5 Whys

How to build healthcare solutions that are truly Innovative

 

The healthcare industry is making tremendous strides in implementing new technology into new and innovative solutions. However, healthcare innovators should be cautious about applying traditional and potentially non-innovative thinking into the solutions that they are proposing. This is especially true in the area of voice technology, natural language processing, and machine learning in healthcare.

Health innovators should ask The 5 Whys, to develop truly innovative solutions.

For example, one of the most talked about applications for voice technology in healthcare has been the transcription of doctor’s notes into information that can be entered into the Electronic Health Records (EHR). According to the Annals of Internal Medicine, for every hour that a physician spends with a patient, they must spend two hours on related administration.

“Among practitioners, everyone talks about ‘pajama time’—spending a couple of hours every night to complete their administrative duties.” says Toby Cosgrove, M.D. former Cleveland Clinic CEO, who has joined Google Cloud’s healthcare and life sciences team, as an executive adviser.

Applying the 5 Whys to healthcare innovation we can ask:

1.       Why do physician have to transcribe their notes every night?

Doing this work at night “eliminates the need for a physician to sit at a keyboard for an entire (doctor) visit, something that’s unpopular with patients.” Cosgrove said.

This has led innovators down the path to allow “doctors (to) speak to the EHR. This would eliminate the need for a physician to sit at a keyboard for an entire visit.” Cosgrove said.

This sounds like a very innovative and valuable solution. However, we should not stop there, we need to dig a little deeper to get to the root cause of the problem.

2.       Why is there is so much information to transcribe.

Traditional innovative analysis has focused on the complexity of the codes associated with entering patient’s medical information and billing codes into the EHR. The challenges of knowing and understanding the information that needs to be entered in a format that a EHR can understand, is a specialty in itself. This analysis is true, however there is more to the story.

Digging a little deeper we quickly discover there is no, or very little communication with a patient between doctor’s visits. Doctors are trying to learn and understanding everything that has happened to the patient in the week, month, 3 months, 6 months or even the past year since they last spoke to the patient. Ok so,

3.       Why don’t patients and doctors communicate more often?

With a full patient load doctors are unable to see patients more often. Doctor spends about 15 – 20 minutes per patient and in this short period of time they are trying to gather as much information as possible from a patient. There are not enough hours in the day and not enough doctors to serve the patient population. Traditional thinking has focused on doctor patient ratios, how to serve rural populations, and virtual doctor visits, however none of them has been able to make a significant improvement in the quality of care or reduction in the amount of time needed to transcribe notes.

4.       Why not add more staff?

The answer is simple cost. Additional staff costs money. Additional Doctors cost money, additional Physician Assistants cost money, additional billing specialists cost money. The cost of care is already high, and this approach would only make it more expensive. This deep dive into the problem appears not to be getting us any closer to a better solution. What else can we look at.

The big challenge with today’s approach is that information is communicated by the patient, transcribed by a member of the healthcare community, a doctor, and then entered into an EHR. So far, all the proposed solutions focus on transcribing the doctors notes into the EHR. Which leads us to the 5th why.

5.       Why not use technology to transcribe information directly from patients?

Earlier this week Amazon announced the release of Amazon Comprehend Medical.

Amazon Comprehend Medical, (is) a new HIPAA-eligible machine learning service that allows developers to process unstructured text and identify information such as patient diagnosis, treatments, dosages, symptoms and signs, and more.[1] This could be the first step to using voice technology, natural language processing, and machine learning to communicate directly with patients, on a more regular basis, possibly using virtual medical voice assistants.

Applying HIPAA compliant solutions directly to patients would reduce the amount of information, needed to be transcribed into the EHR. Information could be entered directly from the patient’s mouth into the EHR, in a pending physician review queue, eliminating the step required to transcribe doctors’ notes.

This approach could use machine learning to proactively alert doctors of potentially serious medical conditions prior to in person visits, helping reduce emergency room visits or emergency appointments. Also, this direct to patient approach could allow doctors to review notes already entered into the EHR at a patients regularly scheduled appointment, rather than focusing on collecting information from the patient. Of course, any additional information needed to be added by the physician could use the physician to EHR voice transcription as well. However, the goal of an innovative patient facing solution would be, to reduce the amount of information needed to be collected from the patient, therefore reducing the amount of information and time the doctor needs to transcribe their notes, and allows them to focus on analyzing patient reported data and discussing appropriate next steps for their patient.

To learn more about developing innovative solutions at your organization contact us at info@vohesu.com.


[1] Amazon Comprehend Medical https://aws.amazon.com/comprehend/medical/