Monday, December 31, 2012

Summarizing GeriTech in 2012

Today is not only the last day of 2012, but also marks the three month anniversary of the GeriTech blog. This is the 36th post, all written so far by your faithful practicing geriatrician (atypical practice, to be sure, but real practice nonetheless).

So it seems like a good time to look back at what's been posted, and try to summarize what this blog is about.

I'll admit that when I launched this blog, it wasn't with the intention of being the sole author. I was rather looking forward to hearing about what technologies other clinicians had found to be helpful in the care of older adults.

However, so far the clinicians have been rather quiet on the blog (although they tell me very interesting things about technology when I talk to them in person, and I hope to get some of this on the blog in time).

For now, GeriTech = Leslie Kernisan's professional interests, and hence reflects my personal priorities, interests, experiences, and discoveries about innovation and the future of healthcare.

Here's what this specifically means:

Geritech is about addressing the nation's most important healthcare question

No false modesty here! This blogger believes that her interests dovetail with the most important healthcare question facing the country, which is this:
How can we, as a society, provide effective compassionate healthcare to an aging population, at a financial cost that the nation can sustain?
(Do you disagree? Post a comment asap.)

Better primary care for older adults is the answer

The answer to the question above, of course, is that we must meaningfully improve primary care for older adults. This is a no-brainer conceptually. First of all, it's what patients say they want: to receive care that is comprehensive, coordinated, person-centered, prevention-oriented, as upstream as feasible, and as outpatient as possible. Second, it's what is cost-effective, compared to managing health problems later when they become health crises requiring hospital care.

(Again, if you disagree with the above, I'm eager to hear your case.)

Just about everything I'm professionally interested in tracks back to this essential issue: how to support and implement better primary care for older adults.

The problem is that it's very hard to figure out just how to improve primary care for older adults. Still, it must be somehow done.

GeriTech's key points and discoveries so far

All in the service of fostering better care for medically complex older adults, here's what I've found myself saying on GeriTech:

  • Most tech tools seem poorly suited to improving the health care of older adults. As far as I can tell, most tools are not designed for use by clinicians caring for complex older patients with multiple medical problems. This is a big problem, as we really need effective tech tools that do one or more of the following: 
    • 1) help clinicians like me do our work better, faster, or more thoroughly; 
    • 2) help patients and families do their healthcare work better, faster, or more thoroughly; 
    • 3) support the collaboration between clinicians, patients, and caregivers. 
I posted a list of likely obstacles to senior health tech innovations here.
  • Effective tools for older adults should be developed with the help of practicing generalist clinicians and geriatricians, i.e. generalist clinicians who have real-world experience managing whole older patients, not just specific diseases. (For instance, I might say that I don't particularly need continuous monitoring of blood electrolytes, but I do need help collecting data on symptoms.) We need the involvement of people who have experience with the space in which the rubber meets the road.
  • Patient engagement requires clinician engagement.  I believe that ideal health care is grounded in constructive collaborative relationships between patients, caregivers, and clinicians, in which clinicians serve as expert consultants in helping patients meet their healthcare goals. This means that solving for patient engagement means solving for clinician engagement.
  • We should rethink how we are trying to engage and motivate PCPs. Meaningfully improving healthcare for older adults in large part means helping PCPs change what they are doing. We will both need them to adopt new ways of practicing (including new technological tools), and we need many of them to change the way they engage and relate to patients, caregivers, and other clinicians. Also, most geriatric care will be delivered by PCPs (not nearly enough geriatricians to go around now, never mind in 20 years), so their working conditions should be of utmost interest to all those who want healthcare to get better. Right now I hear a lot of talk about incentivizing PCPs and tinkering with compensation. There is not nearly enough talk about nurturing clinicians' internal motivation, and relieving their feelings of burnout. As a doctor who left conventional primary care practice due to burnout, I'm concerned. Smartphones and tablets alone will not help us care for our parents and grandparents, even if they have the world's best computer algorithms behind them. We need clinicians to be ready and able to partner with us.

In a nutshell

This GeriTech blog is fundamentally about the pursuit of better healthcare for older adults, especially the frailer and more complex adults who have the most need (and who happen to be my patients). Improving primary care for this population is essential. We will need suitable tech tools and suitable systemic changes to achieve this.

Through this blog, I've been documenting the practical challenges that I find myself, my patients, and their families facing as we work together to improve their primary care. I hope that this information will help develop others develop better technological solutions, which can improve geriatric care.

Much of the ongoing conversation about improving healthcare doesn't have nearly enough focus on the particular needs of older adults and their families, or on equipping the average PCP to handle those needs. Hopefully this will change soon.

The most pressing healthcare problem our society needs to solve asap is how provide effective compassionate care to an aging population. This blog will continue to encourage clinicians, innovators, developers, and thought leaders to work on this problem, with a special focus on how technology can be part of the solution(s).

Friday, December 21, 2012

TEDMED's Chronic Diseases Videochat: Lots of Lifestyle, Minimum on Multimorbidity

Yesterday was the TEDMED Great Challenges virtual panel discussion on "Managing Chronic Diseases Better". I listened and participated by Twitter. (See it here.)

I came out of the talk thinking they should rename their Great Challenge:
How Coaching and Lifestyle Modification Can Prevent Diabetes, Obesity, and Cancer, and Can Also Help Manage Diabetes. 

Because those were the main topics discussed, and also seemed to be the primary domain expertise of the Challenge Team.

Now, these are indeed worthy topics of great population health importance.

However, such a focus marginalizes the millions of Americans who need much more than coaching and lifestyle modification to manage their chronic diseases. It also offers little help to those of us -- clinicians and caregivers -- in the trenches who are struggling to help these patients with their complex health needs.

Another disappointment is that the discussion didn't really address the challenges of managing multiple chronic diseases, also known as"multimorbidity". This is unfortunate, since managing chronic diseases becomes a different -- and harder -- ball game when people have several of them. (Here's one review of the evidence.)

This is especially true when people are older and frailer, or if they have cognitive impairment (which seriously impairs their ability to self-manage their chronic conditions).

And as any primary care provider can tell you, there are *lots* of people out there with multiple chronic diseases. The UpToDate chapter on multimorbidity (which relies in part on this report) lists the following fun facts:
  • Estimated 1 in 4 Americans have at least two chronic conditions
  • Estimated 2 in 3 Medicare beneficiaries aged 65+ have two or more chronic conditions, 
  • Estimated 1 in 3 Medicare patients has 4+ chronic conditions 
The UpToDate authors go on to say that:
"Multimorbidity is associated with staggering healthcare utilization and costs. The two-thirds of Medicare beneficiaries with multimorbidity account for 96 percent of Medicare expenditures."

Doesn't this sound like a Great Challenge to you?

Well, it seems this is not the Great Challenge TEDMED has in mind. This Great Challenges team was very strong on prevention and lifestyle modification, had a strong patient advocate (a younger woman with Type I DM), but had precious little advice on managing scenarios like the one I describe in this post.

What to do when a person with diabetes, high blood pressure, arthritis, glaucoma, and eleven medications comes to the visit, where we discover uncontrolled blood sugar, too high blood pressure, falls, urinary frequency, anxiety, trouble managing medications, and social isolation?

This is managing chronic diseases as many of us experience it. We could use more innovative brainstorming sessions to work on solutions.

The TEDMED panel on the question I didn't ask

Here is a question that was mistakenly attributed to me -- in a twist of cosmic irony, TEDMED identified me as the author of a question that I would literally *never* ask --  followed by the initial answer provided by Dr. Micheal Roizen, Cleveland Clinic's Chief Wellness Officer:

Q: "What do I do with patients who are resistant to any kind of help with their chronic illnesses, and with behavior change related to that? How do we convince patients who are resistant to treating his or her chronic disease?"
A: "The only ways are either educating them about what is important, or incentivizing them strongly." (See the rest of the answer here; the details include coaching, environment modification -- get the Chips Ahoy out of the room -- and a nice $2000 bonus for those patients who succeed).

Dr. Roizen goes on to say that 63% of their patients with chronic disease have transformed. This is a nice result. But I doubt you'd make much headway using that approach with older patients suffering from multimorbidity.

Why? Let me start by answering the question that I didn't ask.

What, in fact, do clinicians like me do when we come across older patients who are "resistant to help with their chronic illnesses"? Here's what I do:
  1. Assess for cognitive impairment. Before I start "educating", I look into underlying reasons for why the person can't manage their health care, or otherwise isn't following through on the care plan. In older adults, cognitive impairment is often discovered, once one looks. (Early dementia and medication side-effects are the most common underlying problems that I find in the outpatient setting.)
  1. Try to figure out how the patient and family see the illnesses, in order to understand what's interfering with their ability to address the illnesses. Common things that turn up include:
    1. Financial problems getting the medications
    2. Overwhelmed by too many medications, by too much complex medical care in general, or by caregiving demands
    3. Skepticism about allopathic medicine or other doubts about our conventional proposed approach
    4. Misunderstanding how serious the illness is, or how treatable it might be
    5. Substance (ab)use
    6. Low health literacy
You'll notice that many of the problems above are far more common in people with multiple chronic illnesses.

Now, I'm not against lifestyle changes and behavior modification. My goal, and I know this is the goal of many PCPs and geriatricians, is to figure out a mutually agreeable, and feasible, plan to help the patient with his or her health. Often this includes coaching on lifestyle (if I can help them get it) or environmental modification (if feasible). 

But a lot of it is figuring how to help patients follow-through on conventional medical management. Like picking one or two generic medications to focus on (a good opportunity to talk about what's likely to bring the patient the most bang for their buck). Or picking a symptom to focus on managing. Or sitting together to review what the specialist said, and putting it in light of the patient's overall health condition (and other chronic diseases).

The TEDMED panel on geriatrics and chronic disease

Another illuminating question and answer: here is the geriatrics-related question I had submitted via Twitter:
My Q: "So much chronic disease occurs in geriatric patients. Why aren't there more resources targeted especially towards the elderly, or those with dementia?"
The question was directed to Dr. Dileep Bal, a public health officer from Hawaii, and you can view Dr. Bal's answering the question here.

He gives a long detailed answer focusing on -- surprise surprise -- prevention and wellness. He says the "focus needs to be in keeping them well, especially for elderly population. Fifty percent of our health costs are related to people in their last year of life. So I think both from a financial, and from a lifestyle point of view, specially for the elderly, you need to focus on how do you keep them healthy."

He also says "Keep the well elderly well at home, before they show up in my clinic." He goes on to describe a program of preventive care including senior centers, exercise programs, and systemic dietary interventions (like limiting soda and fast food availability). He mentions people in their 90s participating in exercise programs, and the need to modify societal cues (McDonalds is mentioned).

Another physician on the panel, Dr. Deneen Vojta (whose bio highlights extensive experience in diabetes prevention and management) offered a different perspective. She noted that older women commonly do not list exercise and healthy eating as priorities, but rather care about their finances, friends and family, and staying in their home.

True that! Then Dr. Vojta goes on to describe how lifestyle changes should be framed as ways to achieve those above priorities. She doesn't address how patients suffering from very symptomatic chronic conditions, such as heart failure and COPD, can be supported in making these lifestyle changes (hint: for many, it requires medication optimization so they can be more active).

Sickcare versus health education and coaching

One part of the talk that did resonate with me was when Amy Tenderich pointed out that as our healthcare system is really a "sickcare" system, it's problematic to ask it to be responsible for prevention. She suggested that we might consider adding another arm to the system, which would focus on health education and coaching.

I like this idea. However, for those patients who are older and sicker, education and coaching needs to really integrate into their "sickcare," both by helping patients navigate the sickcare system, and by taking their various diseases into account when providing health education and coaching.

For instance, I've seen many older diabetic patients develop mild dementia, and struggle with their diabetes care. They need help figuring out simpler and safer strategies for their diabetes. However, many diabetes educators don't seem prepared to problem-solve around mild dementia. (Or perhaps they just don't notice my writing "suspect mild dementia" in the referral? Would earlier definite diagnosis via brain scan help?)

Is TEDMED's panel missing a key point of view?

I found myself wishing TEDMED had included an expert able to really discuss managing multiple chronic diseases in primary care, such as Ed Wagner, who pioneered the Chronic Care Model. (This commentary by Wagner on chronic care management addresses multimorbidity and person-centered care, and is a really fantastic read for those who have journal access.)

They could've also considered someone particularly focused on the unique needs of older adults. I might nominate someone like Cynthia Boyd, a geriatrician who has published fantastic articles on multimorbidity and on Guided Care, a program that help older adults manage and coordinate the care of their many chronic conditions. (Her 2005 JAMA article on what happens when you try to apply clinical practice guidelines to a typical patient with multiple conditions is a classic. She also co-authored this very good 2012 JAMA commentary on designing healthcare for multimorbidity.)

Last but not least, although I'm disappointed in the way that the TEDMED talk skewed towards prevention and lifestyle, I can't say that I'm surprised. Much of what I've come across these past few months, as I've been learning about healthcare innovation, is skewed towards younger, educated people who either want to prevent disease, or are heavily invested in the management of one particular disease.

This despite the fact that the experiences of older adults drive most healthcare spending, not to mention the impacts on these patients and their families.

In a nutshell

Older adults and those with multiple chronic illnesses are two very large, important, and challenging populations to care for. Improving chronic disease management for these groups is essential, both for the sake of the millions of patients and families affected, and because this group drives the bulk of national healthcare spending.

TEDMED's team for "Managing Chronic Diseases Better" seems to have special depth and expertise in the prevention of chronic disease. Their recent videochat largely focused on healthier lifestyles, coaching, and prevention, and had very little on the crucial challenges associated with managing -- not just preventing -- multiple chronic illnesses. They also had little to say about the ways that chronic disease management often should change to meet the unique needs of older adults and their caregivers (such as adaptations when patients develop cognitive impairment). Conditions such as heart failure and COPD weren't discussed.

I'll end by quoting the intro to the 2012 JAMA commentary cited above:
"The most common chronic condition experienced by adults is multimorbidity, the coexistence of multiple chronic diseases or conditions."
If TEDMED wants its Great Challenge to be relevant to really making management of chronic diseases better, I hope they will find a way to address older adults and multimorbidity in future events.

For more of my take on TEDMED's Great Challenges so far, see this post about the Great Challenges overall, and this post about last month's videochat on caregiving.

Wednesday, December 19, 2012

Improving health information exchange from hospitals to outpatient

In my last post, I described why it took me almost two hours to sort through the recent hospital records of an elderly patient. The records had arrived in my digital fax queue as a PDF with 159 pages of images of text.

It seems a bit nutty, in this digital age, that so much of medicine is still being done through paper and faxes, but there it is. Most hospitals have EMR systems, but if a clinician who's not within the hospital system needs medical information, that information usually gets printed and faxed. (At my previous job, I used to get CDs with PDFs from Kaiser. Better in some ways, but still relatively slow to work through.)

What might work better? The most popular plan I hear to solve the problem of independent clinicians accessing hospital records is to create Health Information Exchanges (HIEs).

I really don't know what to think of HIEs, especially since I haven't yet had the chance to use one.  They seem to be slow to set up -- the Bay Area has been working on one for a while -- and it's unclear how well they will smooth the workflow of independent outpatient doctors.

Health Information Exchange features we need

Let's assume the HIEs will eventually arrive. How should they function, to allow outpatient clinicians to get needed information in an easy and low-hassle manner? The best-case scenario I can envision with a HIE would be a online system that I could easily log into, and that would allow me to do the following:

  • Allow me to search through records to find specific items related to my patient.  I can't emphasize search enough. Browsing seems to be the main EMR paradigm. It's slow and a pain and increasingly divorced from our experiences with consumer software, where using the search function rules (and delivers). As clinicians, we shouldn't be trying to figure out which section of some other medical center's EMR contains pulmonary function tests; we should just be able to search for it.
  • Make it easy to copy certain records to my own EMR. The ideal would be to easily tag items that you want to copy, and then have them transfer to your own EMR with useful titles included. (Of course, if the data I get is searchable and my own EMR has a good search function, the titles become less necessary.) An alternative would be for providers to be able to use something similar to the EverNote web clipper tool, where you highlight what you want and it gets easily transferred.
  • Push or pull certain commonly requested groups of records on request. Just as most labs have created groups of commonly ordered labs (CBC, comprehensive metabolic panel, etc), you could create groups of commonly requested records for certain purposes. For example, when I need to review records from a hospitalization, I want to see the Admission H&P, the discharge summary, all radiology (except chest xrays; just one of those please) and studies, and the last available lab values. (I don't need every CBC drawn in the hospital, just the last one.) Ideally providers could customize the groupings that they wanted. The consumer version of this is creating a search/query, and easily being able to save/reuse it.
  • Lets the patient directly give me permission to access his/her records. Whether or not the patient gets direct access to all his or her hospital records, wouldn't it be great if the patient could directly and easily give permission for certain clinicians to gain access? Would speed things up immensely, and make it easier for patients to make sure their care is coordinated. 

What's the best path forward for information sharing and care coordination?

I admit I'll be surprised -- stunned, really -- if the HIE, whenever it finally arrives to the hospitals near me, has many of the above features. An HIE is after all enterprise software, purchased not by the daily users, but by administrators or better yet, groups of stakeholders.

Are there alternatives? Some EMR companies, like Practice Fusion, are spearheading a move towards peer-to-peer sharing of medical information. This is a promising idea, but doesn't help when one is trying to extract information from a hospital.

Another option: that health information sharing among clinicians will be driven by the patient's control over his or her medical records. Certainly could happen, since connectivity with patients may happen a lot faster than PCP connectivity with hospitals.

Personally I'd welcome this, as I'd like to see patients get to gatekeep most of their medical data, including laboratory data. But I don't know how close we are to patients getting access to their raw hospital data (and am skeptical that the Blue Button output would cut it, when it comes to clinicians getting the needed info to coordinate care). OpenNotes is a promising start but was just for the outpatient setting -- and the patients were mainly in their mid 50s with relatively few medical encounters.

Will patients soon be able to download meaningful hospital data into their personal health records (PHRs) and share with clinicians of their choosing? Only if they demand it.

In a nutshell:

Health information exchanges (HIEs) are projected to eventually allow outside PCPs to access hospital records. Features I hope they'll include are capacity to easily search and copy the information. Bonus if patients end up able to directly give other clinicians permission to access their medical information.

For more information: in doing a little brief research related to this post, I came across a few good resources for those who want to learn more about HIEs. The National eHealth Collaborative has a report on "Secrets of HIE Success," which offers insights into how some HIEs have come into being.

But far more interesting and informative to me is Robert Rowley's series of blog posts on HIPAA and HIE: Part 1, part 2, part 3, and part 4. Love it when articulate primary care docs write about healthcare.

Clinicians and others, have you had good experiences with HIEs yet?

Monday, December 17, 2012

Flawed tech & processes hinder care coordination; why reviewing a hospitalization took me two hours

Last week, I received a digital fax of 159 pages from a local hospital. A 159 page PDF. Text not searchable since it's a PDF of an image.

This is what happens when you are the primary care provider (PCP) -- or the geriatrician asked to assist the PCP -- and you try to figure out the whole story, in order to understand the medical situation and properly coordinate care.

About these records: they are all related to the patient's recent hospitalization. The patient's DPOA requested my help recently, since the patient has been home for a month, is still quite confused and weak, and is having difficulty making the trip to see her usual primary care doctor.

Time required to work my way through them, properly enter information into my EMR, get my own questions about the patient answered, and reach the point at which I could summarize the most relevant parts to another doctor?

Almost two hours.

It's an instructive example to consider, as we think about how to move towards a system in which medically complex patients can get the right kind of primary care, and follow-up care. Patients like these are the ones who are supposed to benefit from ACOs, and perhaps even Medicare's new payment of post-hospital care coordination.

So let's talk about what that care coordination can be like for PCPs, especially post hospitalization.

The patient herself presumably qualifies as a high utilizer, but probably not an exceptional one. She's elderly, has some chronic problems, was hospitalized for a serious illness, and her hospitalization included about two weeks in the ICU. She was discharged to a rehabilitation facility, she had to go back to the hospital for a week (not clear that her readmission was preventable), she went back to a rehabilitation facility for several weeks, and now she's been home for a few weeks.

Neither her PCP nor I are affiliated with the hospital in question. This means that we can't directly access the hospital's records online. This too, is not an exceptional situation.

The workflow and work of reviewing an extensive hospitalization record

You may be wondering why the heck it took me almost two hours. I'm going to tell you.

Here are the steps I just went through:

  • Single 159 page fax arrives digitally in my EMR fax inbox. I was able to open the PDF and view it within the EMR. Practical problems:
    • Can't search the file for specific text, to find answers to my specific questions (such as, did she get a head CT or MRI?).
    • Everything mixed together: admission H&Ps, discharge summaries, labs, multiple radiology studies, procedure notes, consent forms. It's all basically in whatever order the hospital's medical records staff printed it, and then faxed it. It's not in an order conducive to my putting the story together in my own head, or in my EMR.
    • My EMR offers a text box on the side to summarize key findings within PDFs. Nice when the document holds just one item. Not so useful when the document is a lengthy amalgamation of lots of different items.
  • 159 pages need to be sorted, organized, and triaged. Not only do I need to read this information and capture what I need right now, but I need to be able to search through it later on, as additional information queries arise. This means:
    • Deleting or setting aside pages that I don't need now and don't think I'll need later (e.g. the umpteen daily chest xrays from the ICU, or the consent forms)
    • Finding a way to save the xrays to my xray documents section, the labs to my lab documents sections, etc.
  • Adobe used to extract pages and create smaller documents. I ended up breaking the hospital's fax into 24 documents of use to me, plus one that contained the leftovers (including the umpteen ICU chest xrays). I gave the documents titles that explained what they contained.
  • Medical documents reviewed and filed in proper places within EHR. For many, but not all, I used my EMR's text box to summarize the key points. 
  • Key information about hospitalization summarized in chart note. By summarizing the main points I'd learned by reviewing the records from this hospitalization, I have something that I can send to other clinicians, or even to the patient's DPOA.
  • I decided to look up a more unusual medical problem. The patient saw several subspecialists while in the hospital. After reviewing the hospital record, I decided the patient's current difficulties might mainly be related to one particular specialty problem. I look up the problem on; after all, this isn't a condition that I co-manage often, and I need to figure out how this problem might be affecting her overall trajectory and prognosis. 
    • What about a specialist? Well, I don't know that her outpatient specialist has reviewed all these records, since she hasn't seen him since being discharged. (He's not part of this hospital system either.) Plus, she's feeling too unwell to leave the house often for follow-up appointments. But I have called him and hope to connect with him soon.

In short, I was hindered by:
  • getting the information in a non-searchable format and difficult to reorganize format
  • having to identify each item and title it myself
  • having to spend time properly placing the information in my own EMR

See why care coordination between PCPs and hospitals is often problematic?

Now, it's true that I work alone in my micropractice, so some of the work I've done could've been shared with someone else. But there are downsides to sharing this with a non-physician teammate:
  • Teammate has to be able to: accurately evaluate value of each page, title each sub-document properly, and file properly in EHR
  • Harder for non-physician to determine whether additional info needed from hospital
  • Physician needs to wait for teammate to process hospital records --> delay
  • If teammate has incorrectly organized hospital info --> further delay

I have to say that I found dealing with post-hospitalization records a real killer at my last job. We often didn't have the records at the time of the follow-up visit, and when we did have them, the staff had difficulty sorting through a large set such as this one (as did I, riffling through 50+ pages right before walking in to the room to see the patient). I also had much less time to review records, since the practice involved busy sessions seeing patients with relatively little time for care coordination and deeper thinking about complex medical situations.

Lest you think this took almost 2 hours because the patient is relatively new to me: it's certainly possible that this added a little time. However, at my last job, we often had new or relatively new patients come in post-hospitalization. These were often people who either had not been getting primary care prior to the hospital, or who had decided they were dissatisfied with their prior provider. So a system meant to support PCPs in providing good post-hospital care should not presume that PCPs always are very familiar with their patients.

Could better technology and upcoming innovations help? I think so, and will share thoughts on this subject in the next post.

Clinicians, any comments or ideas regarding your experience reviewing records after lengthy hospitalizations? 

[Note: Many thanks to the patient's DPOA, who gave me permission to share this example on GeriTech.]


In a nutshell

Reviewing the records from a longer hospitalization, and integrating this information into the longer-term medical plan, can be very labor-intensive and time-consuming. When PCPs and other outpatient clinicians aren't part of a hospital system, receiving hospital records means getting reams of information by fax. This information is usually not searchable on computer, and requires time to sort, summarize, re-organize, and file into one's own EMR. For a complex hospitalization, the time required can be considerable.

We need technologies and healthcare systems that support PCPs -- and other outpatient physicians -- in doing this important work.

Thursday, December 13, 2012

PCP behavior change: crucial for tech adoption and improving healthcare

These days it seems that everyone wants primary care providers (PCPs) to change, and to improve.

If you believe in better healthcare through use of new technologies, then you want PCPs to adopt new technologies --both within their offices (EMRs and care coordination platforms) and by engaging with technology in the patient's environment (smartphone apps, aging-in-place technology, etc).

In other words, you want PCPs to do things like prescribe apps (see here for Susanna Fox's comments on clinicians and stagnant health app adoption), and integrate "observations of daily living" into their clinical work. Or maybe even practice like Eric Topol (see here for why I'd have trouble doing it).

If you believe in more patient-centered and individualized care, then you want PCPs to spend more time developing meaningful collaborations with patients and families. You may also want PCPs to start relying on more accurate individually-generated medical data to make clinical recommendations.

In other words, you want PCPs to consider a patient's genomic information or personal biometric data set when recommending treatments, and you want them to engage in shared decision-making.

And if you believe in patient and caregiver engagement, then you want PCPs to support and respond to that engagement.

In other words, you'll want PCPs to encourage, collaborate with, and coach patients and caregivers on successfully managing their health needs, and you'll want PCPs to be more available to respond to patients' concerns.

Sounds good to me. I believe in all three of these ideas. The second and third are core components of the geriatric approach (albeit historically done in a much less tech and data intensive fashion), and I think properly leveraging technology will be essential to managing the considerable needs of an aging population in a time of limited resources.

Furthermore, we expect all the above to lead to the holy grail of healthcare improvement: better care at a lower cost. This is plausible: health services research shows that a better primary care infrastructure generally corresponds to better population health outcomes, and more cost-effective care.

Plus, people seem to like having good primary care. Almost everyone prefers to have their health problems treated in the outpatient setting, and would like medical intervention earlier, in order to avoid hospitalizations.

In other words, PCP behavior change seems to be the great hope for improving US healthcare. (It's certainly my own great hope for improving outpatient geriatric care, since most of that care will be delivered by the nation's non-geriatrician PCPs.)

Which is why I think all of us advocating for healthcare change, healthcare improvement, and healthcare tech adoption should be spending lots of time talking about how to motivate and enable PCPs to make these changes.

And if you believe in the motivation science presented in Daniel Pink's "Drive" - and so far I do -- then we should definitely emphasize harnessing PCP's intrinsic motivation to be better PCPs. (See here for why I think this is important to the healthcare of seniors and frail elders.)

So, I'm currently considering attention to PCP intrinsic motivation as I come across various stakeholders discussing desired changes to the healthcare system.

For instance, in a recent blog post, Caroline Popper describes how she's helping HHS figure out how to "move reimbursements from 'fee for services' to 'fees for performance.'" She asks “how do you measure performance? How do you pay for it?”

Well, those are literally billion dollar questions. If she thinks PCP behavior change is part of the answer, then I hope she and her team will consider the way monetary incentives tamper with intrinsic motivation in professionals. (At the very least we might want to have ACOs be very careful about the way they try to apply incentives to PCP behaviors.)

In another recent post, a mHIMSS editor comments on the need for physician champions to help take mHealth to the next level.

I am all for it, but historically exhorting PCPs to do more or do better hasn't been enough (although it probably helps to hear it from another PCP). If the mHealth community wants PCPs to engage, can they help create PCP working conditions that nurture internal drive, rather than attempt to change PCPs with monetary incentives?

What about the rest of you? What do you think are viable ways to motivate PCPs to engage and adapt to change?

In a nutshell

The overall improvement of healthcare, especially for seniors, hinges on developing a better system of primary care. This will require PCPs to make substantial behavior changes, especially if adopting new technologies and new ways of practice are required.

The social science described in Daniel Pink's "Drive" describes the perils of relying on external rewards and punishments, when trying to motivate people to do their best work. In particular, such strategies can seriously erode intrinsic motivation, which is often key to performing creative work or empathic work.

The healthcare changes we all want will require PCPs to step up and do their best creative and emphathic work. If we want PCPs to engage and be open to using new technologies, how can we help harness their intrinsic motivation?

Ideas sorely needed.

Tuesday, December 11, 2012

Technology predicted to replace 80% of what doctors do: challenges to overcome for senior health innovations

If you, like me, believe that technological innovations are key to providing better care to seniors at a reasonable cost, then Vinod Kholsa's recent article on technology and doctoring is a must read.

Kholsa predicts that "computers will replace 80% of what doctors do and amplify their capabilities."

I myself am all for getting technology to currently do -- or help me do -- many of the tasks that are involved in my line of doctoring: comprehensive high-quality primary care to seniors with multiple chronic medical problems, as well as to frailer elders with geriatric problems such as cognitive impairment, falls, and overall declines in function and independence.

I especially appreciate Kholsa's emphasis on the very sensible ways technology can augment clinician's capabilities (like capturing and processing data), and allow us to focus on what we might do best (the human relationships parts of medicine). I really can't wait for the day when I can focus on relationships and helping families navigate their healthcare challenges, rather than getting bogged down in data chasing (how often is she agitated at night? when did she poop? what labs has she had? what meds is she taking?) and communication issues (what did the oncologist tell you? what medications have others prescribed for you?).

So my question is, when and how do we make this happen for the healthcare of older adults?

The challenge of getting technology to replace what doctors for seniors do

The thing is, it's much easier to get technology to do 80% of what my doctor does, compared to getting technology to do 80% of what I do as a doctor.

I'm a healthy 36 year old woman. My doctor mainly needs to help me with prevention, healthy lifestyle choices, and family planning.

Whereas when I provide medical care to my patients, I do much much more. That's because my patients have multiple chronic problems, many of them at advanced stages, plus often cognitive impairment and physical limitations.

It should go without saying that it's by providing tech assistance in the care of the more medically complex patients that we collectively stand to gain the greatest benefits, both in terms of improving the quality of care for individual patients (and families!) as well as getting better value for the money society spends on healthcare.

There are certainly some very complicated pediatric patients, and younger adult patients. However overall, the bulk of illness (and healthcare spending) is concentrated in older adults.

Everyone agrees on the need to treat chronic illnesses upstream, and in the outpatient setting. But unfortunately, as best I can tell, most healthcare tech innovations are NOT geared towards facilitating high-quality outpatient care of seniors with multiple problems.

Why? I've been trying to figure this out, and here's what I've come up with so far.

Why it's hard to design useful healthcare tech innovations for the primary care of older adults with multiple chronic problems

  • Medicine emphasizes a disease-based focus, so technology has done the same. Even though primary care doctors must often deal with multi-morbidity (patients having several chronic illnesses), healthcare is still mainly organized and specialized along disease-based lines. Many healthcare tech innovations seem to be following suit, possibly because the developers tend to team up with a specialist in the clinical area. Also, a number of tech innovations are spearheaded by a young tech-savvy person with a given disease. (Will we have to wait until the tech developers hit their 60s and are suffering from multi-morbidity before we get tools designed for people with multiple illnesses? I hope not!)
  • A narrower focus is easier to design solutions for for than a broader one. We already have this problem in research: easier to design and conduct a good study when you focus narrowly on a certain population and exclude the messy complex people with additional health diagnoses (or trouble taking their meds). Unfortunately, in the real world of clinical care for older adults, many patients have messy and complex health needs. And/or don't take their meds. This has made it hard to apply research findings to them, and is going to make it hard to apply many tech solutions.
  • Older people are perceived as less inclined to use technology. There is definitely something to this. I'm not even very old and I see a big difference in how people ten years my junior are integrating technology into their lives. So this adds to the design challenge for the innovators: designing solutions for seniors means figuring out how to meet them where they are technologically, and how to make things extremely user-friendly. Added twist: the way older people use technology is rapidly evolving, and partly depends on what kind of technology is available.
  • Technology for the health of older adults requires more interfacing with clinicians. In other words, if you design a tool meant to help an older adult manage a medical problem, you need to design something that works for the patient, AND the clinician. Two users is harder than one. Even if it's a nominal clinical interface (like all the web portals for each app; please note that we doctors will probably not be willing to log into more than 1-2), it's still more work than designing some wellness app for consumers to use on their own.
  • Technology for the health of older adults needs to be accessible to those with limitations, and accessible to caregivers. Specifically, many older adults have physical limitations (vision, hearing, finger dexterity) as well as cognitive limitations (dementia) which could affect their ability to use a technological health tool. Plus older adults with limitations are often being assisted by family or paid caregivers, so tech tools need to accomodate that as well. Add another two ticks to the list of design challenges.
  • The business case is trickier for the Medicare population. Who will pay for the use of the technology? Will it be the insurer? The patient/consumer? The family? The ACO? These questions seem to be especially uncertain when it comes to the Medicare population. As the perceived business case is very important to the innovators and entrepreneurs, this may be why they aren't focusing as much on developing solutions for older adults.
  • Healthcare for seniors is perceived as less consumer-driven than healthcare for younger adults. This is probably a combination of insurance issues and cultural issues. More and more younger adults are either uninsured or under-insured for primary care; this means there is more of an opportunity to directly offer them technological solutions for their health needs. Whereas older adults obviously have Medicare. On the cultural side, older people are more likely to accept the old-fashioned model of medicine in which the doctor takes care of things, and you try to do what the doctor tells you. (I know this because my patients often ask me to tell them what to do.) As Kholsa points out, consumer-driven healthcare is a powerful partner for healthcare tech innovation. But since consumer demands tilts towards the young, innovative solutions are tilting towards them too.
  • Regulatory issues are trickier, the more medical a technology solution is. How these mobile health and other new technologies will be regulated by the FDA is up in the air. In the meantime, personalized medical information = protected health information, which means it's subject to HIPAA. This presumably complicates things if you are trying to design an innovative solution meant to help older patients manage their health. For instance, I recently blogged about patients needing help implementing the multiple recommendations we clinicians usually have for them. Would a solution need to be HIPAA compliant? Probably.
  • Healthcare emphasizes hospitals and transitions more than regular outpatient care. Hospitals are where most of the healthcare dollars are spent, have more data on what's going on, and are also more visible to most academics and other experts in healthcare. Guess where ACOs are going to be focusing their efforts (and sponsoring technology to support this)? It's not on the average outpatient senior. It'll be on the "high-utilizers," i.e. the ones with frequent hospitals and transitions. In principle everyone wants to help people in the outpatient setting before they become high utilizers, but in practice the attention goes to where the money and clout is.

These are the main factors I've identified so far. Does anyone have additional ones to add to the list, or comments on these?

And how to work around these factors? Given the above factors, it seems fairly daunting for a tech start-up to create innovations for the primary care of seniors, unless foundations were to step in and provide a much needed boost. (Is there an RWJF Pioneer-like program for outpatient health innovations, for seniors with multiple medical problems?)

Now I don't want to say that no one has thought about meeting the outpatient healthcare needs of seniors. But I do see the innovations tilting towards the younger and less medically complex. Which is understandable, but regrettable from a senior/geriatric public health perspective. (See last week's post for my plea that mHealth help me help my patients, who need more than wellness and help making better lifestyle choices.)


In a nutshell:

Healthcare technology innovations are disproportionately oriented towards the needs of younger, more tech-adept individuals. Older adults have more complex healthcare needs. They (and us as a society) stand to gain the most from technology improving healthcare, and amplifying the capabilities of clinicians to provide care. But their healthcare needs are harder for the innovators to meet. So, if we want technology to help us with the healthcare of seniors, we will have our work cut out for us.

My latest list of why it's hard to provide good tech tools to my patients is above. I'd love some help revising and refining it.

As always, comments and suggestions as to how to harness healthcare tech innovation in the service of better outpatient care for seniors will be much appreciated.

Wednesday, December 5, 2012

mHealth: We need more than prevention, information, and lifestyle

With smartphones and wireless technologies becoming ubiquitous, mHealth (mobile health; also known as "connected health") is one of the hottest areas for healthcare tech innovation.

What will this mean for primary care? Well, if you're a clinician and want to know what many thought leaders in mHealth think is important (and think of us), I highly recommend you read this essay by Robert McCray, the President and CEO of the Wireless-Life Sciences Alliance.

Here's a summary of the key points as I understand them:
  • The problems:
    • An epidemic of obesity and chronic diseases, in large part due to people's behavior choices and lifestyles.
    • Individuals erroneously assuming that technology will save them from the consequences of their behavior.
    • People have ceded responsibility for fixing their health to institutions paid by third parties.
    • Physicians have focused on "transaction-based medicine" and have not focused enough on improving the health of patients through modifying harmful lifestyle choices.
    • Physicians have focused more on defending their turf, autonomy, and income, than on the health of patients.
    • Neither patient nor provider has responsibility for health outcomes.
  • The solutions:
    • People must "embrace the responsibility for their health and demand the tools to discharge it," rather than assume someone else will take care of the problem.
    • Citizens and consumers must set expectations for the institutions that deliver healthcare services.
    • Physicians should "'prescribe' healthier living habits by directly addressing the harmful lifestyle choices that their patients present."
    • mHealth tools will provide access to "all the knowledge that is needed to reduce the need for healthcare and to select the best healthcare approach when it's necessary."

In summary, a major problem is that people are making crummy lifestyle choices, and physicians aren't doing enough to help them with this. People need to take responsibility and armed with the tools of mHealth, need to start ensuring that they get help making healthier living choices.

Well, maybe this could work if we're talking about a population of younger educated Americans. But what about the millions of older people now struggling with multiple chronic health problems? Don't they need much more than healthier living habits and prevention? And how much responsibility can we expect them to take on?

When consumers need more than lifestyle changes

McCray makes some good points in his manifesto. However, his proposed solutions aren't enough to handle the most pressing health policy priority of the next 30 years: effectively managing the primary care of the Medicare population.

In particular, how well would these ideas work for the many older adults currently suffering from multiple chronic illnesses?

Consider the patient scenario I described in my recent post on managing multiple recommendations: an older adult diagnosed with diabetes, high blood pressure, arthritis, and glaucoma, and experiencing falls, urinary frequency, anxiety, social isolation, and difficulty managing medications.

Let's say this person is a 70 year old overweight woman, and let's call her Janet Doe.

How well do McCray's analysis and proposed solutions fit with what I know of patients like Janet Doe?

Well, it's true that poor eating and exercise habits, and obesity, are strongly correlated with developing and exacerbating diabetes, high blood pressure, arthritis, urinary complaints, and cognitive impairment (which could be why Janet's struggling with her meds).

So maybe she made poor lifestyle choices. Or maybe she made the best choices she could in her life circumstance. Who knows? Public health professionals know that people's health is often strongly affected by factors that individuals have difficulty controlling (like fast food advertising, access to safe parks, etc).

Otherwise, I wouldn't be surprised if she didn't get the best comprehensive primary care (so hard to provide good primary care under conventional practice circumstances!) or good behavioral interventions. And yes, she probably was assuming that the system would take best care of her.

Let's move on and now talk shop about mHealth helping her out. I have four particular comments in response to McCray's article:

  • Many patients need more from clinicians than the prescription of healthier living habits. A patient like Janet Doe needs high-quality outpatient medical care to manage her many medical problems. This should be grounded in a mutually respectful collaborative relationship with a clinician. Obviously we have a long way to go before clinicians routinely offer patients the engagement and patient-centeredness that everyone deserves. We do need patients and families to constructively demand more from their doctors and their healthcare. But let's be clear about the outcome we need. It's not doctors addressing harmful lifestyle choices. It's doctors applying the full range of their medical expertise, in order to help patients and families achieve the best health outcomes possible given the medical circumstances and the patient's preferences.
  • Many patients don't want to be in charge or responsible. Hard for us educated control freaks to always appreciate, but in my experience true. I find patients and families especially prone to become overwhelmed once there are more than 1-2 chronic diseases to deal with. And of course, less educated patients generally struggle more than educated ones.
  • Many Medicare patients will develop cognitive impairment. Presumably in McCray's vision, responsibility then devolves to the spouse or next of kin, as it does now for financial affairs. But these family caregivers are already struggling to manage medical responsibilities. We should only lay more responsibility upon them if we really are able to offer tools and resources that will make this added work manageable. 
  • Medically complex patients absolutely need a physician's expertise to synthesize the ongoing care of multiple chronic conditions.  You can call it a quarterback, a navigator, an expert outpatient consultant (my current practice), or just a plain old-fashioned good PCP. Whatever you call it, it takes a live person with physician-level expertise. Patients need and want to talk to someone who can help them sort through the complexity and choose among the options. New technologies can offer much needed support to everyone involved. But if we want to improve the care of older people, we need to support that patient-clinician partnership, rather than imply that it can be replaced.

What we really need from mHealth, to care for the health of aging adults

My short(ish) answer to this question is that we need tools that 1) allow clinicians to do what they need to do, faster and more efficiently; 2) allow patients and families to do all the extensive self-management that they have to take on, and 3) facilitate effective collaboration, communication, and shared decision-making between clinicians, patients, and care circle.

Here's a wish list with some specifics that could help me with my daily clinical practice:
  • Tools to manage the multiple recommendations we generate when we see patients like Janet Doe.
  • Tools to help clinicians and families manage medications, especially when multiple providers are involved.
  • Tools to help patients and families gather the daily data we all need to manage their chronic problems, i.e. tools that collect the symptom information as easily as possible, and then make this data digestible and actionable for patient, caregiver, and clinician.
  • Tools to help multiple involved parties communicate. Older patients have family caregivers, private caregivers, home health agencies, hospitals, facility staff, and other clinical specialists involved. Need help keeping everyone in the loop and coordinated.
  • Decision-support tools. Do you want us to discuss the risks and benefits of a certain medication or procedure? Make that data easier for the clinician to access quickly. Make it easy for families to find suitable decision guides for common medical decisions.
  • Better personal health records. Patients should be able to easily access their medical information. They need meaningful summaries. They need access to their labs and other diagnostic data. They need to be able to easily share this with the clinicians of their choosing, so that they can help coordinate their care, get the right care in an emergency, or easily ask for a second opinion.

I could go on, but I'll stop there for now.  Clinicians, what's would be on your own mHealth wish list?

To Mr. McCray and his colleagues, I applaud their much needed efforts to change healthcare, and I do think mHealth technologies have wonderful potential to improve the health of the elders and families I work with. The questions to keep discussing, of course, are how to apply these technologies, how to redefine the roles of patients and families, and how to enable clinicians to do their best work in partnership with patients.

In a nutshell

An approach relying on consumer-directed health and mHealth-powered lifestyle management could work for the educated and relatively healthy American. But many seniors with multiple medical problems will not be able to take on primary responsibility for their health outcomes.

Most older adults need more than prevention and healthier living habits. When people have multiple chronic illnesses, they have substantial ongoing primary care medical needs. These older adults will need physicians to help them synthesize the care of multiple conditions, and to navigate complex medical decisions. mHealth can't replace this partnership, but can certainly support it, by creating tools that facilitate effective collaboration and communication between clinician, patient, and caregivers.

Monday, December 3, 2012

When multiple problems lead to multiple recommendations; how to help patients succeed?

How to help patients and caregivers keep track of and follow through on the many things we ask them to do?

I found myself thinking about this yet again last week, after the group of UCSF students I was teaching brought up the following concern:

"We're worried about the patient getting overwhelmed by our making too many recommendations."

Good point, learners.

I had to admit to them that I probably don't think of this nearly as often as I should. But it's such a good thing to consider. Because the truth is, I think it's often harder than we realize for patients and caregivers to keep up with the plan for the many problems on an older adult's list. (It can also be tricky for a clinician to not lose track of everything going on -- a topic for future posts.)

In geriatrics, of course, we usually have a lot of problems to address: six or more in a visit isn't uncommon in my practice. That's because frail older patients have many ongoing chronic conditions, and often raise additional complaints at each visit.

But even younger Medicare patients often present with multiple problems.

Here's a common scenario: If you have diabetes, high blood pressure, arthritis, glaucoma, are on eleven medications, and in the visit we uncover uncontrolled blood sugar, too high blood pressure, falls, urinary frequency, anxiety, trouble managing medications, and social isolation, then we are going to have a lot of shared decisions to make. Which will lead to a long list of recommendations.

Furthermore, I consider difficulty managing meds to be a sign of possible cognitive impairment or possible financial difficulty; if either or both are present, then plans for every other medical problem will need modifying.

(Also note that this isn't a very medically complex patient in geriatrics; many present with this stock set of problems plus congestive heart failure, COPD, kidney disease, and complaining of pain and shortness of breath too. Oh and, rash and toe fungus. )

So here we are now with our mutually-agreed upon problem list and care plan. Now what?

The challenge for patient and caregivers: so much to do

My own experience has been that most patients and families have trouble keeping up with more than 1-2 recommendations. The trouble is, for adults with multiple medical problems, a single visit can generate multiple recommendations. And of course, the more action a recommendation requires, the less likely it will happen.

The simplest recommendations ask patients to do something simple, just once. Like "Stop medication X."

But most recommendations are more complicated: "Increase your metformin to 500mg three times daily" requires a small ongoing change in what a person does every day.

Then there's "Talk to your daughter about helping you figure out a way to not forget to take your medications." (This is the kind of amorphous directive that the Heath brothers warn about in Switch, one of my fav books on change management.)

Or, "Check your blood sugar every morning before eating. If your blood sugar is over 150 for three days in a row, increase your Lantus by 2 units."

Or "Start bladder training. You can review the attached handout for instructions."

Let's reconsider the example of the patient above, with her out of control blood sugar and blood pressure, plus falls, anxiety, urinary frequency, difficulty managing medications, and social isolation. Here are the problems we routinely face in helping such patients:
  • Do we list recommendations for all the problems we uncovered in the visit, or just for a few?
  • For every problem we make recommendations for, how to help patients and caregivers follow through?
  • If we focus one just one or two problems, how do we ensure we don't lose track of the other issues? (Patients seem to hate coming back every week for another visit, but that's often what we suggest, assuming we have appointment slots available.)
  • How do we clinicians update the plan for a problem, based on feedback we get from patient and family? (Many problems require a trial of an intervention before we can decide how to proceed long-term, or may require diagnostic results before we move on.)

What kind of technology can help us?

Technology is already helping, in that EHR systems are now creating clinical summaries for patients (it's a criteria for Stage 1 Meaningful use).

This is a step in the right direction, but it's not enough.

What patients and caregivers (and the clinicians helping them) really need is something more on the order of a sophisticated task management system. Or possibly even project management tools.

Specifically, we need something that:
  • Can keep track of multiple problems and recommendations
  • Helps patient and clinicians identify which recommendations will be the focus now, versus later
  • Allows us to update each other on the status of problems, and the plan for each problem
  • Allows us to break down the plan for a given problem into manageable and actionable chunks
  • Allows us to share information as needed with others involved in addressing a problem, i.e. other clinicians (such as nurses for problem-specific coaching), family caregivers, etc.
A bonus in my view would be technology that allows the clinician to enter several recommendations all at once (when the clinician is thinking comprehensively about the patient), but allows patient and caregiver to work through them over time.

So for instance, let's imagine that for the patient above I have recommendations for the following problems:
  • diabetes management
  • blood pressure management
  • urinary urgency
  • falls
  • anxiety
  • medication management
  • socializing
Let's also assume that the recommendations for these problems are NOT interdependent (often they are, but we'll go for a simpler scenario).

I've actually discussed and provided management recommendations to patients and families on this many problems within a single 30 minute visit (it's a bit of a hustle, but possible).  But asking the patient to take action for each problem would probably be overwhelming.

However, imagine a platform in which I propose recommendations for all problems and prioritize them. On the patient's side of the technology, he or she gets to work through a problem for a period of time, and then later is given recommendations for the next problem. (These could be staggered based on time, such as weekly, or based on completion of the previous problem's plan.)

Such technology could allow us as clinicians to be more comprehensive, with less risk of overwhelming patients and families. This would enable clinicians to make best use of their time, and help patients be more successful in taking the many steps needed to improve their health.

So is this technology out there? I'm not sure. Several companies are working on consumer products meant to help caregivers manage multiple tasks, but it's not yet clear to me how easy it will be to integrate clinical recommendations into the programs.

On the provider side, patient portals are becoming de rigueur, but as far as I know, they have not yet evolved to support this kind of sophisticated care plan management for patients.

But hopefully that will change. In the meantime, if anyone has come across technology that helps clinicians help patients manage multiple problems, I'd love to hear about it.

In a nutshell

It's very common for older adults to present to a primary care clinician with multiple problems, which can generate multiple recommendations for patients and caregivers to implement at home. But patients often have difficulty managing more than 1-2 recommendations, so a comprehensive approach is often too overwhelming.

We need technology to help patients follow through when there is a lot to do. Specifically, we need help making multiple recommendations more manageable for patients and families. Something along the lines of sophisticated task management (or perhaps project management?) would help.

The ideal technology would facilitate and reinforce effective collaboration between clinicians, patients, and caregivers, and would allow all to work through a longer problem list together over time.

If you can suggest any such technologies that might work for older adults with multiple problems, definitely let me know.