Sunday, March 18, 2012

Progress in managing the national injectable drug shortage

Yay for Saskatoon Health Region (SHR) Pharmacists! As I mentioned last time, SHR (along with the rest of the country) is dealing with a shortage of injectable medications.  Pharmacists have the task of managing our region's supply.  Right on schedule (March 13), they created a table of suggested alternative medications to address the shortage of injectable medications.  (Even though I advocated for this in my last post, don't give me any credit, because they were already in the process of creating the table before the idea occurred to me.)

The table has been circulated, posted on the SHR network, and has already been updated.  They also have a plan to track use of injectable medications that are in very short supply and have conversations with physicians using those medications to make sure that alternative administration routes are used as soon as appropriate.

This situation made me think of the Model for Improvement, a widely-used technique for process improvement.

The Model for Improvement has us ask ourselves 3 fundamental questions whenever we want to improve something:

What are we trying to accomplish?

How will we know a change is an improvement?

What changes can we make that will result in an improvement?
 In the case of the med shortage, the answer to the first question is obvious: We're trying to conserve the supply of injectable medications.  At least, that's how it looks from an operational point of view.  Our real aim is may be something like:

We will ensure that patients receive appropriate care and access to necessary medication.  We will do this through careful management of the injectable drug supply and by use of alternative medications and administration routes.

SHR Pharmacists have already given answers to the third question: What changes can we make that will result in an improvement?  They have notified all clinical staff of the shortage and have asked that we consider changes to practice that will conserve medications that may be in short supply.  Further, they've helped us achieve that goal with the alternative medication table mentioned above.  They will have individual conversations with practitioners, if there seem to be opportunities to appropriately change clinical practices.

It's the second Model for Improvement question I'd like to consider further: How will we know a change is an improvement?  This is about measurement.  Or, more accurately, measurement to enable change.

Pharmacists will be tracking use and supplies of injectable medication.  But, as they are not the ones using the medication, that information will have limited use if it's only in their hands.  Clinicians understand the importance of this initiative to appropriately use medication to preserve supply.  We can do a better job if we get feedback on our performance.

Show us whether we (individually, by department, or hospital ward) have made helpful changes to our practice.  Compare rates of use of injectable to non-injectable meds before and after this change initiative.   Have the reminders/exhortations from Pharmacy made a difference?  If so, plot these rates over time so we know whether helpful changes are sustained.  If not, then we need a different strategy.

Take advantage of positive deviation in practice.  Find out which departments/wards are making the changes more successfully and explore how they've achieved that.

Show clinicians that there actually is room for improvement.  I suspect that many docs had the same thought as me when I received the notification about changing prescribing behaviour: I'm already optimally using these medications.  This change doesn't apply to me!

What is the gap between current and desired prescribing behaviour?  If the gap is wide, then let's work to narrow it.  If not, then efforts to change clinician behaviour are a waste of time and we should pursue other change ideas.

Specific feedback will keep clinicians' attention on this important issue.  Without it, emailed updates will become background noise.  

Sunday, March 11, 2012

Clinicians need help in order to conserve drugs in this time of shortage

Eliminate slogans, exhortations, and targets for the work force asking for zero defects and new levels of productivity
- W. Edwards Deming (10th of Deming's 14 principles of management)

I love it when 2 ideas smush together in a chocolate-and-peanut-butter way.

It happened for me this weekend.

On Friday, I had a great time at BCPSQC's Quality Forum.  In addition to attending some stimulating breakout sessions, I met Dr. Keith White, BCPSQC's Clinical Lead for Medication Reconciliation.  As he explained his work, he told me about the challenges that family physicians have in ensuring antibiotic use for patients with respiratory infections.

Most patients seeing their family physician about flu and cold symptoms have a virus, and antibiotics are not useful (and may be harmful).  Despite public education campaigns, some patients may still have an expectation that they will be prescribed antibiotics.   Public health organizations have exhorted physicians not to prescribe antibiotics in these situations, so as to reduce the risk of side effects and the development of antibiotic-resistant bacteria.

Dr. White explained to me that it's not sufficient to encourage this behaviour in family physicians; we need to provide the tools to help them.  He suggests using "scripts", or rehearsed responses, that address common concerns patients raise, such as "The last time I felt this way, the doctor gave me antibiotics", or "What's the big deal? It's just an antibiotic."

When prepared with responses to "frequently asked questions", physicians can confidently address common concerns and misconceptions that patients may have about antibiotic use in viral infections.  But, we can't expect that they will magically have these answers at their fingertips.  An initiative to curb inappropriate antibiotic use should include providing this resource to physicians, and showing them how to use it.

Make it easier for them to do the right thing.

That was the chocolate.  Here's the peanut butter:

Drug shortages have been in the news, and on the minds of health professionals across Canada.  Because of production problems at a manufacturing plant that supplies most of our country's injectable drugs, physicians are being asked to conserve supplies and consider using alternative medications via other administration routes.

Late last week, our health region circulated a memo asking nurses and physicians to "immediately change to oral equivalent medication to conserve the injectable drug supplies...".  Medications affected include common pain-killers such as morphine, as well as other commonly used drugs such as heparin.

All clinicians will recognize the importance of making these changes, where appropriate and safe for our patients.  We realize that we need to conserve injectable drugs so that they are available for situations where there is not good substitute.  But, good intentions may not be enough.  Perhaps federal, provincial and regional administrations should take Dr. White's advice and make it easier for clinicians to make these changes.

Specialists become comfortable and familiar with a fairly small palette of medications.  We know the dosage, side effects and indications.  We may occasionally use an alternate form (say, if our patient is allergic to the more commonly used drug), but don't necessarily have the same level of comfort.  It would be very helpful to receive specific suggestions on which drugs could be appropriately substituted, along with equivalent dosages (especially for narcotics!) and any special considerations or side-effects.

We already have some confusion in our urology department about whether we can continue to use injectable heparin in postoperative patients (to prevent harmful blood clot formation).  Surely, we can't be the only surgical department in the country with this issue!

Don't expect that busy clinicians are going to research each medication change.  We may just default to status quo, and continue using the same medications - via the same administration routes - that we're familiar with.

But, administrators can help us.

I suggest supplying each clinician with specific recommendations/options for substituting drugs and administration routes.  Give us the bare-bones information, but also a way to drill down if we want to explore in more depth.  A website would be good; an iPhone app would be better. (Seriously, this is an important problem; throw some resources at it.)  Ideally, we'd have national experts create specific guidelines that can either be sent directly to clinicians, or sent to health authorities to distribute after they review them for local relevance.

This needs to happen by the start of business on Tuesday, March 13.  

Yes, I realize I'm dreaming to think a national initiative could be mounted in 24 hours.  So, instead, let's see it happen in each health region.  This is an urgent situation.  Patient care is and will be affected.  The shortage may continue indefinitely.  The sooner we start making appropriate choices for drug use, the more medication we'll be able to conserve for patients who really need it.

It's a fairly short list of medications.  I imagine a hospital pharmacist could come up with a recommendation summary in one working day, and circulate it to clinicians by email.  Post it in all care areas where these drugs are commonly used.  Put a copy of it on the front of all inpatient charts.

Don't worry about making it perfect the first time around.  As long as it gives safe information, the first version doesn't have to be comprehensive.  Ask for feedback and we'll let you know what other information we need.

Make it easy for us to do the right thing.

Then, give us some feedback.  Report each ward's medication use before and after this intervention.  (N.B. Use charts, not data tables.  We like pictures!) We'll be curious to know how we're doing.  As this is an unusual situation, I doubt that there are published benchmarks for the conversion of injectable to oral administration routes.  But we can do internal comparisons (for example, between surgical units), to get a rough idea whether or not we're being diligent in making the requested changes.  "High-performing" units will be a source of ideas for their colleagues. But, this will work only if we know how everyone is performing.

Most importantly, make it clear to all clinicians that appropriate medication is never to be rationed to patients.  If the alternate drug or administration route is not suitable or effective, then our patients must receive the standard drug according to established practice.

Our health region memo goes on to say "If this voluntary conservation method does not maintain minimum supplies, stricter measures will be employed."  That sounds sensible and prudent to me, but let's make sure it doesn't come to that.

Don't just tell us what we should do.  Show us how to do it, and help us along the way.

Make it easy for us to do the right thing.

Wednesday, March 7, 2012

Variety is the spice of confusion and waste

Two months ago, a patient’s wife told me off before his surgery.

In the pre-op area, I was reviewing the planned surgery with them.  I mentioned that he would likely stay 1 or 2 days in the hospital.  Her forehead furrowed.

“Well, which is it,” she demanded, waving an information pamphlet in front of me.  “The nurse told us 2 or 3 days in the hospital.  This pamphlet says 4 or 5 days.  Now you say 1 or 2 days!”

I stood by my estimate, and asked to see the pamphlet.  While the information about the surgery and postoperative care was accurate, the hospital stay quoted reflected a practice that was about 5 years out of date.  I didn’t realize that this information was still being given out, and understood why she would be confused.

After a 2 day (phew!) hospital stay, he went home.  I called him the next week to discuss his pathology report.  After we had finished, I asked to speak with his wife.  I apologized again for the confusion about the length of stay, and told her that our nurse educator had made the necessary changes in the pamphlet.  I was interested to find out why it had bothered her so much.

“I wasn’t upset about how long he was going to stay in the hospital,” she replied.  “I was upset because I thought, if you people at the hospital weren’t talking to each other about a simple matter like how long someone stays in after surgery, maybe you wouldn’t talk to each other about how to look after my husband.  I didn’t know if I could trust you!”

From a minor (I thought…) and easily explained discrepancy, she had concluded that we were not a cohesive system, and was worried that her husband’s health might be in jeopardy because of it.  This small oversight had shaken her confidence in our ability to provide safe care.

I wish I could find fault with her reasoning.


Variation in clinical practice has been on my mind since a discovery several years ago.  My interest was reawakened recently due to our practice’s latest improvement initiative.

About 3 years ago, as part of our Advanced Access/Clinical Practice Redesign work, we found a wide variation in urologist practice regarding patient recall.  This 2008 post explains that, but in a nutshell, we found that the rate of patient recalls (internal demand) varied from almost zero up to 25%.  We realized that some recall visits are helpful, but that they also use capacity that could shorten wait times for new patient consultation.  For that reason, we worked toward reducing unessential recall.

We discovered, through group discussions, that most of the variation between our practices was “just because”.  That is, we recalled patients with certain conditions and at certain intervals because that’s the way we were trained and the way we had always done it.  We had never discussed “appropriate” recall as a group.  Once we recognized the variation, we developed methods (such as follow-up algorithms) to facilitate follow-up by the patient’s primary care provider.  Our recall rates dropped significantly over the 6 months after that initial discovery.

In that case, clinical variation was a marker for waste in our office system.

Our practice’s latest improvement initiative is around improving care for patients with bladder cancer.  Before Christmas, in order to explore what our current system looked like, I asked one of our office managers what she thought we could improve.  She had no hesitation in her reply.

“Get your act together with BCG,” she suggested.

BCG is commonly used chemotherapy treatment to prevent bladder cancer from returning.  Patients receive 6 treatments, started after their bladder surgery.  After that, they have regular endoscopic bladder examinations to detect any tumor recurrence.

That sounded pretty straightforward to me. I didn’t see much variation there.

She corrected my thinking.

“You all have a different way of ordering the treatment.”

She went on to explain that each of the 8 of us used a slightly different BCG protocol.  Each protocol was medically reasonable, but there was variation in the interval between surgery and treatment, between each of the treatments, and between the treatment and follow-up examinations.  Our staff had to keep track of each urologist’s unique habits (which occasionally changed from patient to patient!).

Staff had to be careful when explaining the treatment and follow-up schedule to patients as, if they gave the protocol belonging to the wrong urologist, the patient would be confused.  (And, have their confidence shaken.)

Also, our staff felt that a common protocol would improve patient safety. (Yes, we really do have amazing staff who think this way!)   Keeping track of multiple protocols increases the chance of confusion and the chance that we’ll miss scheduling important follow-up examinations.

Significant clinical variation had once again gone unrecognized.  (Well, unrecognized by me because I don’t see any variation in my own process.  And if my own process does vary, I convince myself that there’s a darn good reason for it.)

We physicians zealously guard our professional autonomy. We may see efforts to reduce clinical variation as a threat to that autonomy.  I agree that some variation is important to preserve, that is, variation related to each patient’s unique disease process, experience, needs and wishes.  But variation related to poor coordination of our healthcare system leads to waste, affects patient safety and erodes trust.

I think that much variation in practice is present not because practitioners are exerting their autonomy, but rather because we simply haven’t yet identified the variation and appreciated its impact on our patients.  If our healthcare system had methods to identify clinical variation, and encouraged clinicians to be curious as to its cause, clinicians would develop their own solutions to reduce it, and thereby improve our patients’ care.

Anonymous lights a fire under me

Anonymous posted this comment on my last entry:

You know, just because you do a poll on what people want, doesn't mean you should just quite writing for a month when the majority say shake it up, with a bit of both.:o)
Looking forward to another post!
  Thanks, Anon., for checking in, encouraging me and (gently) taking me to task!