Uncategorized

I received a snail-mail letter from the Clinical Services Division of Silverscript the other day, and I didn’t know whether to laugh or sigh. So I did both.

Silverscript is a Medicare-approved Part D Sponsor. In other words, this company provides prescription drug plans to Medicare beneficiaries.

It also seems that they are attempting to provide clinical decision support to providers. In principle, this sounds like a reasonable idea. Older patients tend to take a lot of medications, and harmful interactions aren’t uncommon.

In practice however, I’ve yet to come across any similar missive from a Part D plan that is actually helpful.

Take this letter, for instance. It starts by informing me that “this confidential drug utilization review program provides educational information regarding your patient’s drug therapy. Our goal is to facilitate optimal, safe, effective, and high quality drug therapy at lower costs.”

Fair enough.

Next it says: “Reevaluating Anticholinergic Use in Benign Prostatic Hypertrophy (BPH)” and goes on to inform me that their records indicate my patient is receiving Spiriva and Tamsulosin. Their concern is that “anticholinergic agents in patients with BPH can counteract the effects of alpha-1-adreneric antagonists and alpha reductase inhibitors, leading to urinary obstruction and rentention.”

[Yes, they misspelled “adrenergic”.]

I’m then advised to consider slowly discontinuing the anticholinergic, changing the current medication to something less likely to precipitate urinary symptoms, or reducing the dose of the anticholinergic.

The letter concludes by acknowledging that “patient variables, unavailable to us, may make the current therapy appropriate for this individual,” and I’m asked to respond via fax or mail, by completing a form regarding how I plan to act on the information they’ve sent. I’m also advised that I can call a toll-free phone number if I want a copy of the patient’s medication profile.

Sigh.

The problems with this clinical decision support letter

 Two key problems stood out to me:

• The patient is no longer under my care. And hasn’t been for the past 9 months. This means that I’m inappropriately being shown someone’s protected health information. I assume it also means that the physician currently responsible the patient’s medications didn’t get this letter, but it’s possible that Silverscript has sent letters to several of us.

• There is no information on the likelihood of the potential interaction effect. This makes it hard to know how seriously to take this warning. Geriatricians do generally love to hate anticholinergics (they can cause delirium and worsen cognition), but we usually make an exception for the inhaled ones, since side-effects occur *much* less often than with oral anti-cholinergics.
• In checking UpToDate’s page on anticholinergics for COPD, I find a review of clinical research on urinary retention in BPH patients taking inhaled anticholinergics. In case you’re wondering, case-control studies have found some increased risk of retention, but in a randomized trial of 6000 COPD patients (many of whom had BPH), the increased risk of urinary retention was too small to be statistically significant.

In general, I’d say that the overall likelihood that a patient would benefit from a prescriber receiving the above letter is basically nil. In practical terms, the significance of this possible interaction pales in comparison to the many other interactions and side-effects that older patients are routinely subjected to.

Also, how often can such a letter be useful if it’s divorced from the context of the patient’s conditions? It so happens that this patient has quite advanced COPD; last spring, he was just starting to require daily oxygen in order to get through his day. And, he has BPH and urinary symptoms, so that’s another important quality of life issue.

Should we be bombarding providers with lots of little decision support tips?

Of course not. Providers, like all humans, can only pay attention to so many little decisions and distractions during the day. I know the prescription plan is trying to improve prescribing quality, but in truth, these kinds of letters are not helpful.

So I would certainly recommend that Part D plans, and the healthcare system overall, be thoughtful in deciding what they want to alert prescribers to, and how.

What would it take for these missives to be helpful? I’m not really sure, and when I take a quick look in the literature, I don’t easily find anything studying the effect of these post-prescribing recommendation letters.

In general, it would be good if we could somehow prioritize which interactions are most important to identify and intervene on. One approach would be to take action on those interactions which are above a certain likelihood of harm threshold, instead of sending warnings for any potential interaction.

Next, I’d suggest that anyone who wants to send practicing physicians “educational information” start by running their proposed materials past a few real practicing physicians. We can probably make some useful pragmatic suggestions, and help ensure that the educational materials are better received. At a minimum, we could provide a reality check.

Last by not least, I hope these Part D plans will get better at identifying a patient’s prescriber. I’m glad to receive this letter in that I can assume the patient is still alive, but I still think it’s a bit wrong that I’m receiving his health information so many months later.

Summing it up

Pharmacy benefit programs routinely send providers “helpful” letters pointing out that some aspect of the patient’s medication regimen should be reconsidered. I have yet to find any such letter useful.

In the case I describe above, the letter highlighted a potential interaction without providing me any guidance regarding how likely this interaction was. It turns out that the risk of harm was quite small, vastly outweighed by the benefits the patient is getting from the two medications. It also turns out that this patient hasn’t been under my care for 9 months, yet I’m still receiving letters containing his protected health information.

Receiving many of these letters makes it difficult for practicing doctors to spot problems they *should* take action on. It would be nice to know if there is any evidence supporting the use of these kinds of letters from prescription drug plans. In the meantime, I suggest such letters include more information on the likelihood of harm, and that pharmacy plans get feedback from practicing doctors before sending them out en masse.

[This post was first published on The Health Care Blog on 3/1/13. The original title was “How to keep track of multiple health issues” but THCB’s retitle is so much better that I’m using it here too. Many interesting comments on the post are here.]

One of the many challenges I face in my clinical work is keeping
track of a patient’s multiple health issues, and staying on top of the
plan for each issue.



As you might imagine, if I’m having trouble with this, then the patients and families probably are as well.

After all, I don’t just mean keeping up with the multiple recommendations that we clinicians easily generate during an encounter with an older patient.

I
mean ensuring that we all keep up with *everything* on the medical
problem list, so that symptoms are adequately managed, chronic diseases
get followed up on correctly, appropriate preventive care is provided,
and we close the loop on previous concerns raised.

This, I have found, is not so easy to do. In fact, I would say that
the current norm is for health issues to frequently fall between the
cracks, with only a small minority of PCPs able to consistently keep up
with all health issues affecting a medically complex adult.

What kinds of things fall through the cracks? Here’s a list off the top
of my head (for the primary care of adults with multiple chronic
conditions):

• Pain
• Insomnia
• Incontinence
• Cognitive impairment
• Depression and/or anxiety
• GERD
• Falls
• Advance care planning
• Moderate anemia
• Chronic kidney disease
• COPD (esp Stage I and II)
• Difficulty managing medications

Why
do these fall through the cracks? I suppose it’s a combination of lots
of little things. To begin with, many of these are problems that don’t
lead to easy concrete steps a PCP can quickly implement within a short
follow-up visit. Within a busy clinic day, it’s almost impossible to not
find oneself gravitating towards the path of least resistance and least
cognitive effort.

Then there’s the too many people involved issue. Many patients,
especially in Medicare, see several specialists. It becomes easy for
clinicians to assume that another clinician will address an issue.

Then
there’s the way most primary care visits are structured. The biggest
problem is that they are short (10-15 min), which makes it hard to
address more than 1-3 issues. Patients often have their own acute
concerns, which can make it hard to follow up on chronic conditions. And
many doctors spend only minimal time reviewing the chart before the
visit, or even after the visit.

And finally, we have our clinical charting habits, and our charting systems.

Charting and comprehensive problem list management

Historically,
most charting systems haven’t prioritized keeping excellent track of
all problems affecting a patient. In the old paper chart system, many
providers kept a problem list within the chart, but there was no
mechanism for ensuring regular checks on every item there. Furthermore, a
problem mentioned by a patient, even if documented in a progress note,
might easily never make it onto the problem list.

In truth, the old system stunk from a
lets-be-sure-to-not-lose-track-of-issues perspective. We essentially
left it to individual clinicians to figure out what needed to be
followed up on, and they cobbled together various error-prone
strategies, like leaving little to-do lists scribbled in the margins of
their notes, briefly looking over the assessment & plan from the
last 3 visits, or trying to skim the whole problem list at the annual
physical.

Needless
to say, patients and families were pretty uninvolved in this process,
unless they were in the minority who take careful notes and try to keep
their own problem list. This meant that not only were patients often
poorly informed as to their health issues, but that we were missing an
opportunity to let them help us make sure we didn’t forget about any
important health issues.

And now we are all transitioning to EMRs. This makes things generally
a little better, but not a lot, because again, the focus in designing
EMRs has been to help doctors document for billing and for in-the-moment
clinical care, rather than making it easy to keep up with a longer
comprehensive list of ongoing problems.

For
example, although I really like the EMR system I currently use (MD-HQ),
when I start a clinical note, I’m faced with a blank text box. Sure, I
can easily look at the list of past medical problems (and insert it into
the note), or the recently used ICD-9 codes, but that’s not the same as
seeing what problems I need to follow-up on, either because I addressed
them recently, or because it’s been a while since they were addressed.

Meanwhile,
the patients and caregivers are also lacking a way to keep up on the
problem list, since this isn’t currently shared through the patient
portal.

Can’t we do better? Really, I end up thinking that what I need is a
health issue management system, accessible to me and the patient, that
can help us keep track of all the medical problems and their status.

 
Primary care as ongoing problem list management

What would this look like? Well, I’ll start by describing what I do
now, and then I’ll offer some thoughts on what I think would help me.

What I do now:

• During a visit, I almost always organize my clinical thinking around “problems.” These end up listed in my assessment and plan at the bottom of the note, where I usually include a comment as to the status, the differential (if applicable), and my own plan for managing and following.

• I provide written recommendations, listed by problem. However, as I pointed out in my blog post on multiple recommendations, the patient’s version doesn’t usually list every problem that I documented in my note. (I find that people get overwhelmed by a list that is too long, plus the more I write in layman’s terms, the more time it takes me.)

• I review the past problem list in each visit. At a follow-up visit, I look at the last visit’s problem list, in order to follow-up. This means I often copy and paste the assessment and plan, and then edit the details.
• But I’ve run into little operational problems with this. For example, now that I provide a fair amount of care by phone and email, many encounters generate a shorter less structured note, so I find myself scrolling back through my notes, looking for one that has a longer problem list.

• I try to keep track of problems not recently addressed. It’s not always possible to address everything listed in the previous visit (people come in with new concerns, among other issues.) So sometimes I have a “Not addressed today,” section at the bottom of my list, but I’m not as diligent as I’d like to be, especially when things are busy. 
• Besides, it’s hard to not notice that this isn’t exactly standard of care among other docs; the other day, I actually reviewed a PCP’s note that just said “Findings stable. No change in plan.” Sigh.

Some
problems I list are actual ICD-9 diagnoses, but not all. For instance,
in my multimorbid adult patient population, a problem like shortness of
breath could be due to COPD, CHF, anemia, or all the above. I also deal
with a lot of geriatric problems like falls, functional decline,
cognitive impairment, etc, which are usually multi-factorial in nature.

Ideas for facilitating collaborative and comprehensive problem list management

Here’s what I think would be helpful to me:

• If each patient’s chart included a list of problems. And
  this should always be viewable by the patient. Patients could be able
  to add problems from their end, but then we will also need a method to
  reconcile and periodically try to streamline the list, or we’d likely
  end up with some redundant problems.

• If I could use the problems as tags for other data in EMR. When
  we get a diagnostic study back, we should be able to tag it with one or
  more relevant problems. We should also be able to easily tag
  medications and other aspects of the treatment plan with a given
  problem. Ideally the system would intelligently propose problem tags as
  you work (a brain MRI in a patient with a cognitive impairment problem
  should probably be tagged “cognitive impairment”; PFTs are likely COPD,
  etc)

• The ability to link a follow-up activity to a problem. Let’s
  say we decide to follow-up on depression symptoms in 8 weeks. As I list
  the depression item in my assessment and plan, it would be nice to be
  able to tag it with some kind of prompt for future action. That way, if I
  start a SOAP note in 3 months, the overdue problem should pop up.
  Better yet, in seven weeks the system should remind me (and the
  patient!) that this problem is coming due, and encourage me to
  electronically send a symptom questionnaire to the patient, so that we
  can get the right data gathered prior to the visit.

• If it were easy to view associated history, studies, and future events for a given problem.
  I often want to look back and see how an issue has evolved over time.
  It should be possible to see all the related SOAP notes, studies,
  hospitalizations, recommendations made to patient, pending follow-up
  actions, etc, when we look up a problem.

• If SOAP notes prepopulated with recent problems, or overdue problems.
  In this computerized age, why do I have to scroll back through my notes
  to figure out what I was last working on when thinking about this
  patient? Would be nice if EMRs could help with this.

• If the patient and I could negotiate which problems we’ll address prior to the visit.
  Rather than discuss this right during the visit, we should be able to
  set a little agenda before hand. This would allow us both to prepare a
  little better. We could also better anticipate how much time to allot
  for the visit.

• If the system were smart enough to propose data relevant to a given problem.
  If it’s time to review anemia, show me the CBC trend and anemia-tagged
  meds. If the issue is CHF, show me the most recent echo summary. We
  clinicians should be able to tweak these to our preferences, but having
  to create each grouping of relevant data from scratch would be a little
  too bad.

Of
course, since most of my patients also see other providers, I’d need a
way to integrate what those docs do into the data for each problem, and
this would be tough if they’d labeled problems differently. But that’s
part of a PCP’s (or geriatrician consultant’s) job: to keep track of
what the specialists said and did and integrate that into the
comprehensive care of the patient. Even just being able to tag an
incoming faxed consult report with a problem would help a little.

Summing it up

Keeping
track of an older patient’s lengthy problem list is difficult, and it’s
easy for issues to fall through the cracks and get forgotten.

We need EMRs to support clinicians in partnering with patients and
families to keep better track of ongoing problems. I would love to see
EMRs move towards really facilitating the organization of clinical data
by problems, and then supporting patients and clinicians in properly
following those problems.

Patients
and families should be able to access their problem list and see the
plan for each. This would help them understand their health (because
just looking at today’s clinical notes probably won’t cut it), plan for
visits, and ensure that all their health issues are followed up on (i.e.
engagement!).

[Consider checking out the comments over at The Health Care Blog; comments here are also very welcome!]

• Avoid medications that make cognition worse.  In particular, avoid sedatives and
  anticholinergics. Older adults and people with cognitive impairment are
  especially prone to get worse when taking these medications.
  Unfortunately, although many docs are quick to prescribe Aricept or
  Namenda, far fewer will help a patient taper off Ativan, or Ambien. (I
  come across far too many patients on dementia medications AND on
  sedatives.) This is a huge problem. Not only have these drugs have been on
  the Beer’s list for decades, but now evidence is accumulating that
  people taking these drugs develop dementia at higher rates (see here and here).
  The trouble, however, is that physicians find it much easier in a busy
  clinic visit to prescribe a sedative than to troubleshoot someone’s
  anxiety, or insomnia. Plus many patients are loath to give up sedatives.
  Then again, most have not been told that their sedatives are almost
  certainly making them mentally worse than they’d be otherwise.

• Avoid delirium.
  Delirium has been linked to further cognitive deterioration. This means that if you’re worried about possible
  Alzheimer’s, you and your family should learn about delirium ASAP: what it
  is, how to recognize it, how it should be worked up, and how to avoid it.
  Medications are often a contributor to delirium, so there’s another reason
  to review the Beer’s list and look out for sedatives and anticholinergics.
  For those who need to be hospitalized, proven strategies exist to minimize
  the risk of developing delirium, such as the Hospital Elder Life Program. At the caregiving website Caring.com there is also a Delirium and Dementia Solution Center which has some good info (disclosure: I wrote most of
  it), and should be helpful even to those who don’t have a definitive
  dementia diagnosis.

• Minimize stress in the affected person. Minimizing stress is important in order to help the
  person function in the best mental state possible. This means working on
  things like a regular routine, good sleep habits, and offering assistance
  with any anxiety related to the ongoing workup, or other life stressors.
  It may also mean reducing the person’s load at home or in the workplace if
  he or she is still working, simplifying the medical plan, and encouraging
  a family to offer a little extra support. Regular exercise is also good,
  both for stress reduction and because it seems to improve cognitive function.

•  Help the caregivers cope. Helping a
  person’s caregiver or family is key to minimizing stress on a cognitively
  affected person, plus it’s good for the caregiver’s health as well. I find
  that when dementia is a possibility, caregivers need help with the
  following:
• Basic management techniques to help people with
  cognitive impairments.
  This means things like not expecting a loved one to be as rational or
  independent as before, and just helping a loved one find stability and
  feasible solutions.
• Constructively working with their fears, frustrations,
  and sorrows. Stress-management techniques
  and cognitive-behavioral approaches can be great, when people are able to
  access them. I also encourage caregivers to access support groups.
• Help coping with uncertainty. First there’s the uncertainty of “is this
  Alzheimer’s or another dementia?” Later there will be other
  uncertainties (how fast will he change? is she safe at home? should we
  pursue this knee surgery? etc.), so it’s good when caregivers can learn
  to be more comfortable with uncertainty, while still working to get good
  care for their loved one.
• Learning to engage and develop individualized
  solutions. Caregivers often look to us
  clinicians to fix a problem, but in many cases, there is no standardized
  pill or test to prescribe. Instead, what’s often needed is a little
  behavioral trial-and-error, to figure out what works for this particular
  person. For example, one older impaired man’s family kept coming to me,
  upset that the patient was not taking his diabetes medication. (He was
  one of those “it might be very early Alzheimer’s; too soon to tell
  for sure” patients.) I kept explaining to them that they needed to
  find some way to help remind him, or encourage them, and that I couldn’t
  tell them exactly how to do it.