This Community is Archived.

While this community is no longer active, we invite you to review and recommend past posts and resources. Membership for this community is closed, but we hope you'll join us in one of the many other communities on GHDonline.

Moderators of Technology for Patient Engagement and GHDonline staff

Idea Automated at-home rounding systems

By Mark Ott | 12 Dec, 2014

Idea submitted by: Mark Ott

A great model for patient engagement already exists and it’s widely used in hospitals across the country: inpatient nurse rounding. Rounding is proactive, personal, and efficient. Patients and families like it. It improves patient outcomes, safety, and satisfaction. And clinicians actually find it to be less work, not more.

If rounding works so well at managing populations of admitted patients, why shouldn’t it also be used to manage populations of at-home patients? Imagine post-discharge rounding, chronic condition rounding, or general wellness rounding.

Now take it one step further. Instead of clinicians doing this at-home rounding manually (via the phone or home visits), what if technology could round on patients automatically? It’d be proactive, personal, efficient, and it could operate at a scale so much greater than manual rounding ever could.

Clinicians manning the helm of such a system would only need to handle the issues, exceptions, and risks identified by the rounding system. And every interaction they have with patients who are being rounded on becomes so much more meaningful and focused because they can see exactly what issues a patient is having.

Automated at-home rounding can improve patient outcomes, safety, and satisfaction and at the same time reduce healthcare costs.


Joaquin Blaya, PhD Advisor Replied at 9:02 AM, 13 Dec 2014

I think this is a great idea and is inline with systems looking to have more contact with patients while being at home to be able to help them. For example, automated calls with nurse follow up ( which is now being implemented by several startups in the US. The key seems to be a combination of automating processes but having a human interaction whenever the patient says they need it, so that there is still the human touch and psychological and emotional support.
How would you see automating the rounding process?

Mark Ott Replied at 12:47 PM, 15 Dec 2014

Thanks for the thoughts, Joaquin. That's a good study to have for reference. I completely agree about the need to strike the right balance between automated system interactions and human interactions. A principle my team and I are using to help us find that balance is to let technology handle what humans aren't good at and let humans handle what technology isn't good at.

Automated rounding requires a system be capable of assessing patients based on a number of factors, such as primary diagnosis, comorbidities, demographics, medical event dates (like discharge date), etc. In the same way that inpatient rounding focuses on simple things (like the 4 P's), automated rounding can assess patients according to widely accepted, evidence-based best practices according to the patient's primary diagnosis.

The automated rounding system follows a predefined rounding plan per primary diagnosis that specifies how frequently rounding should happen (daily, twice per week, etc.) and for how long (30/60/90 days, continuously, etc.). For example, a post-discharge rounding plan for an HF patient may be 30 days long and the rounding frequency may be daily to start and then be less frequent towards the end of the 30 days, whereas a chronic condic condition rounding plan for a diabetic patient may be twice per week on a continuous basis.

Regarding the communication modality for rounding, people seem to prefer an asynchronous approach (via a secure website/app) compared to a synchronous approach (via automated phone calls). Furthermore clinicians following up should have the option to choose the modality best suited for the follow-up: written message (like an email), phone call, or even in-person appointment. There was a great article published recently about an asynchronous virtual visit system that's been developed at Mass General by Dr. Ronald Dixon (Article: and my point of view on it:

Kedar Mate Keynote Speaker Replied at 1:27 PM, 17 Dec 2014

Mark, this is a fascinating idea and in-line with some recent thinking we've been doing at IHI about "flipping" primary care (see articles in Lancet and Healthcare and I like the idea of letting technology help or assist with things that we are not so good at (automated hovering for example) and letting human work on the things that technology may be less effective at (relationship development etc).

While I appreciate the idea focuses on automating and what technology might enable, I did wonder how such an automated rounding service might integrate with other efforts to provide in-home service--for example through community health workers through programs like PACT in Boston or City Works in New York? Asynchronous may make sense for in-home contact, so how then can the synchronous interaction take advantage of the data that is produced by the automated at-home rounding system? Have you tested this concept with any specific populations? And finally, any sense of how we might start to think about paying for a service or system that looks like this?

Mark Ott Replied at 4:43 PM, 17 Dec 2014

Kedar, I love the idea of flipping the clinic! I'm a big fan of what RWJF is doing at Flipping the clinic aligns so nicely with the automated rounding model.

To your questions about how automated rounding might integrate with existing home visits or other synchronous interactions, automated rounding has two potential impacts: 1) It could possibly reduce the need for some home visits because the community health worker can see their patients' current health status in the rounding system. He or she could then choose to forgo a home visit for a patient if the patient didn't need it. Incidentally, this also frees up the community health worker to focus on patients who have more pressing needs. 2) It eliminates the "cold start" at the beginning of the visit, which improves the quality of the limited time the health worker has to spend with the patient. They're able to focus on things like addressing needs, problem solving, and deciding on next steps.

As for piloting the concept, yes. We have piloted this concept with a number of different populations such as post-discharge patients, diabetic patients, and ESRD patients – less than a 1,000 patients in total. The data have been very encouraging. Rates of patient participation have been great: 60% on the lower end and up to 80+% on the high end. And these populations are generally older and include a mix of socioeconomic statuses. We even have some patients who have been engaging consistently twice a week for 6 months!

Who pays for a system like this? Any provider organization making a move to value-based care. The incentives in value-based reimbursements, risk contracting, bundled payments, etc. make a system like this very desirable for provider organizations. Another recent development is the news that CMS is going to start paying providers $42/patient/month to manage patients with chronic conditions (Source:

Even though patient engagement is the right thing to do, providers won't embrace it without the right incentives in place, which looks to be finally happening.

Lisa Shufro Advisor Replied at 4:20 PM, 2 Jan 2015

Hi Mark -

I think you've hit one very important nail on the head, which is to focus on patient/clinician interaction outside the clinic. In addition to providers, I think you have a case for large employers as well. If their premiums go down because of better use (influenced by timing and quality of connection), they have a high motivation to encourage a service like this.

You may have already encountered these orgs before, but I will give you a few folks to look at in case they are not on your radar:

1) Iora Health: whose patient records include a "worry score" that triggers when the clinical team should reach out. In their daily huddles, the team decides which member should reach out (usually the health coach), and to which member of the caregiving community (sometimes the team decides to call the spouse, for example). Although this worry score is an algorithm, a clinician has the ability to review and override the assigned score.

2), which dips into data available for passive collection from your smartphone, which I think can help trigger that balance of automated/human initiated contact.

3) Eliza corporation, although this is a largely automated outbound calling system with speech recognition, I know that the tech team there thinks a great deal about how/when/tone to reach patients and behavioral support for chronic conditions. Might be worth talking to folks on their tech team.

4), which I like because it allows for a dashboard function across a client's multiple devices and enables the "clinician" (in this case, a PT or personal trainer) to customize an approach that is likely to keep you engaged. It strikes me that there could be some similarities between their thinking and the role of a community health worker/clinician reviewing all the data we can get about a patient's activities, not just self-reported and related to one condition/incident of care.

Hope this helps!


Mark Ott Replied at 3:38 PM, 7 Jan 2015

Thanks for your thoughts, Lisa! I see your point about employer interest in this type of technology. Makes a ton of sense. I'm also familiar with Iora, Ginger, and Eliza. NextStep is new to me. Iora's worry score sounds pretty interesting and definitely aligns with our system's focus on gauging a patient's emotional burden of disease. After all, stress and depression are huge leading indicators of an adverse health event. Also, love what Ginger is up to. Hope to hear more from them this year.

One of our important early insights we've had while working on this concept is that patients want better relationships with their providers, most especially when an adverse health event disrupts their way of life. Most people don't want fancy health apps.
Most people don't want to be their own recordkeeper. Most people don't want to quantify themselves (at least manually!). They want to heal from a disruptive health event and get back to normal life. Or they want help adjusting to a new normal if the health event makes a previous way of life not possible.

A provider using this automated rounding system recently shared a letter they received from a patient of theirs who also used the system. It was great. The patient talked about how much they appreciated the provider reaching out and engaging them at home. The patient talked about their plans for continuing the good progress they had made. The best part about the letter is that even though our automated rounding system is what facilitated the provider-patient interactions, not once was the system itself mentioned. This is exactly the role we intend for our technology to play. Technology for technology's sake is not be the goal.

When applied correctly, technology should fade into the background. When that happens, providers can focus on what actually matters: improving patient care, improving the health of populations, and reducing the per capita cost of health care.

This Community is Archived.

While this community is no longer active, we invite you to review and recommend past posts and resources. Membership for this community is closed, but we hope you'll join us in one of the many other communities on GHDonline.

Moderators of Technology for Patient Engagement and GHDonline staff