Monday morning quarterback: The big 3

Each Monday, I'm going to comment about 3 events that are likely to impact the art and science of patient safety, plus preview what I'm tracking this week.

Here's what caught my eye last week:

1. The Joint Commission (TJC) released information about the 2010 National Patient Safety Goals (NPSG). The number of NPSGs dropped from 20 to 11, with the phased-out goals migrating to regular JC standards chapters. There was widespread perception that the 2010 NPSGs reduce requirements, which on the surface appears to be true. The number of NPSGs is down, and no new goals have been added.

You'll find a variety of views about what these changes mean. Heather Comack (@PSeditor) wrote Joint Commission's 2010 Patient Safety Goals Reduce Requirements, a piece in which the changes were welcomed while Mark Graban (@leanblog) shared concerns in A "Step Backward" in Patient Safety. (The comments to Mark's LeanBlog post are worth a read, too.)

The take-away for people interested in risk-reduction?

• NPSGs call attention to points in care where patients are frequently harmed. It's a good thing when standards are revised to clarify their intent or to make compliance with the intent easier to achieve (provided that "easier" doesn't make the process less effective). And at some point, expectations set forth in NPSGs have to become part of the way front line clinicians "do business" for safer care to become the norm.
• The science of what makes patients safer hasn't changed nor is it influenced by where standards are housed or which items are called out on a master list:

Processes used to deliver care should make it hard to do the wrong thing (employ barriers).

Clinicians should expect and advocate for double-checks (redundancies) at high-stakes junctures.

And recovery opportunities--that is, the chance to identify errors that have been inadvertently set in motion before patient harm occurs--must abound.

• As Larry the Cable Guy says, "That's good stuff. I don't care who you are."

2. Two highly respected, in-the-trenches physician leaders in patient safety wrote a Sounding Board piece in the New England Journal of Medicine. In it, Robert Wachter and Peter Pronovost call for greater accountability for compliance with basic patient safety practices (such as hand-hygiene). Both Wachter and Pronovost are well-versed in how system design impacts human performance. They're suggesting that when systems have been engineered in a way that makes it possible for good people to do the right thing, it's time to move beyond coaching and counseling when they don't. Safety leaders are suggesting that it's appropriate to impose sanctions and penalties on healthcare workers who exhibit patterns of non-compliance with safety-sensitive expectations. Johns Hopkins explains the thinking here, while Wachter digs deeper on his personal blog addressing errant physician behavior specifically.

The take-away?

• Cognitive science upholds the notion that negative reinforcement can extinguish undesirable behavior. Every behavioral choice that has the potential to harm a patient does not stem from a system deficit. But ensuring that sanctions meted out for risky or reckless behavioral choices are fair and based on behavior, not the status of individuals (or their financial contribution to an organization), will require the wisdom of Solomon to implement.
• If you are a front line clinician, patient, or patient advocate, this is the time to ask, "How is this organization equipped to evaluate and deal with performance deficits? Is this process transparent and fair?" When it's perceived that people are punished for simple human error or that less powerful people receive harsher punishment than more powerful people, voluntary error reporting is driven underground and the opportunity to learn from mistakes and near misses vanishes.
• I think Just Culture, an evaluation method guided by well-thought out and reliable algorithms developed by David Marx (a systems engineer and lawyer), offers the most robust process for dealing with real-world overlaps involving system design, flaws, and behavioral choices. Learning to use Just Culture algorithms requires discipline, but the methodology is sound, aligns with desired outcomes, and stands up to high-profile scrutiny.

3. Finally, Michelle Fabio wrote a guest post at Code Blog about a Kentucky case that appears to be good news for patient safety. You can read more about the student's blog, what she shared, how the university where she was enrolled responded, and how a U.S. District Court judge ruled at What Can Nursing Students Blog About.

I see this ruling as a good thing for patient safety because the case helps establish boundaries between information students may share and what patients have a right to be held in confidence. It's a delicate balance. The "day in the life of" stories that clinicians publish on personal blogs provide a window into healthcare culture, customs, and norms that are not otherwise accessible.

The take-away?

• Transparency is good thing. (But, as this case illustrates, transparency may let us peer into really messed up houses. It may be helpful to remember there are other ways to deal with what's inside than to tape up the windows.)

And what I'm watching this week?

Really looking for tweets out of the Health 2.0 conference being held in San Francisco. The place appears to be packed with patient advocates and start-up IT folks, a hopeful sign that innovative solutions are incubating. (If you're on Twitter, follow #hcsfbay for a real-time preview.)

Safe Practices to Safe Patients

Understanding the "science behind the compliance" of patient safety is important for a number of reasons. Here are the ones I find the most compelling:

• When process changes are made, professionals who are impacted deserve to know why a change is occurring as well as what they are expected to do. (People don't roll out of bed to go to work and do things they perceive as useless, stupid, or burdensome simply because they're told "Joint Commission requires x, y, or z.....").
• Healthcare is distinguished from other high-stakes endeavors (like aviation and nuclear power) by the number and complexity of distinct high-stakes processes that individual clinicians engage in each day and by the number of distinct high-stakes endeavors an individual patient is exposed to at each encounter. ("Just Do It!" doesn't "do it" for this crowd.)
• The complexity of human beings is such that deviating from a standard approach to accommodate a physiologic, social, or cultural variation is often necessary. (This doesn't mean that standard operating procedures never work or shouldn't be used to form the scaffolding of patient care.) Since every eventuality cannot be tightly defined and scripted, teaching frontline clinicians the principles that inform safest practices allows them to recognize risk points when they must deviate for cause. Safety-knowledgeable clinicians are able to make adaptations that uphold core principles of safest system design. (This is another incarnation of the "teach them to fish" parable.)

I write a quarterly column about patient safety for Medscape, a robust continuing education platform. The most recent column, From Safe Practices to Safe Patients: The Evolution of a Revolution provides an overview of where the discipline of patient safety has come from and previews what's next.

It's an easy-read, no matter where you are along the journey.

I'm @SafetyNurse. Fly me?

Over the past decade, healthcare has borrowed a number of engineering strategies widely used in commercial aviation to improve safety. Barriers, redundancies, and opportunities to uncover errors inadvertently set in motion are increasingly used to help prevent patients from "going off the runaway."

In the aftermath of the US Airways flight that safely landed a planeload of people in the Hudson River earlier this year, a pilot's words about the investment the flight deck has in safety stuck with me. "We're first up and last down on every flight." His words resonate because they so clearly align with a jet's nose-up take-off and wheels-down landing, an image indelibly etched in my mind after flying hundreds of thousands of miles.

We don't have reinforcers like that built into healthcare. But there are signs that the stakes are going up for leadership engagement in patient safety. Last week, The Joint Commission issued a sentinel event alert describing safety-sensitive beliefs and actions required of leaders--many that challenge healthcare's historic "Just Do It" approach to safety.

The Joint Commission's bulletin stresses the importance of matching what leaders say is valued with what's visible to front line professionals during routine and high-stakes junctures of care, particularly in the aftermath of high-profile error. Once again, we'll be borrowing from aviation.

So here's a nod to modeling transparency, an essential element of a culture of safety. I'm @SafetyNurse. Fly me!

Disruptive behavior: Why this dog don't hunt

A hound dog lays in the yard and an old man in overalls sits on the porch. "Excuse me, sir, but does your dog bite?" a jogger asks. The old man looks over his newspaper and replies, "Nope." As soon as the jogger enters the yard, the dog begins snarling and growling, and then attacks the jogger's legs. As the jogger flails around in the yard, he yells, "I thought you said your dog didn't bite!" The old man mutters, "Ain't my dog."

That's disruptive behavior.

Since January 1, 2009 , The Joint Commission has begun addressing hostile behaviors using two new performance expectations:

• The hospital/organization has a code of conduct that defines acceptable and disruptive and inappropriate behaviors.


• Leaders create and implement a process for managing disruptive and inappropriate behaviors.

Disruptive behaviors include the easy-to-recognize ("you-better-duck") kinds as well as passive-aggressive varieties that tend to dog morale, undercutting intention and outcomes. A short, but salient, discussion of disruptive behaviors can be found in "Behaviors that undermine a culture of safety," a Joint Commission Sentinel Event Alert published last summer. The document (and reference list) is chock-full of negative organizational outcomes that can be predicted to arise from disruptive behavior. Irrespective of your stake in the healthcare system, you'll recognize them: medical error; patient dissatisfaction; diminished recruitment and retention; litigation; and increased costs. The document also acknowledges that disruptive behavior is endemic in healthcare settings.

The relationship between disruptive behaviors and highly reliable performance is interesting to me, as are most variables that inform patient safety. I think there are valid reasons, rooted in what's good, right, and just, for people to treat one another with civility and respect. But what "respect," "civility," and "disruptive" look like to large populations of diverse people is subject to cultural nuance, and has always struck me as being exceedingly difficult to operationalize. (I was once a participant in a focus group where someone defined a "nice family meal" as one in which no one stabbed another diner.)

So my engineer's mind wanted to know this about disruptive behavior: "Does managing it lead to better performance?"

Significant barriers prevent this question from being studied in a linear fashion, but work done in commercial aviation over the past 30 years suggests that it does. The analyses of a series of fatal airline crashes in the 1970's led aviation safety experts to understand that human error is causally linked to failures of interpersonal communications, decision making, and leadership. To manage risks associated with human performance, a series of high level communication-oriented behaviors and performance activities, termed Crew Resource Management (CRM), evolved.

Normalizing behaviors within crews has improved the safety of air travel. Malcolm Gladwell tells riveting stories about why and how CRM works in his book Outliers. And a classic scholarly (but easy-to-read) 1999 analysis The Evolution of Crew Resource Management by Helmreich, Merritt, and Wilheim says more and is available online.

Crew resource management is emerging in healthcare with early AHRQ-funded projects now reporting and organizational templates, like TeamSTEPPS now available. The Joint Commission's acknowledgement that disruptive behavior creates breakdowns in the teamwork necessary to deliver safe patient care is an important foundational step in nurturing cultures in which safety is paramount.

Taking on disruptive behavior as the "safety-negative" bully it is, codifies something healthcare providers intuitively know but may have forgotten: If you stab people at the dinner table, they're unlikely to leap to their feet, performing the Heimlich maneuver with efficiency and aplomb when you choke.

The average Walmart shopper understands risk reduction!

A few weeks ago, a pediatric nurse shared an intervention that lessened the likelihood of an IV medication error. Her story celebrated nursing advocacy, the ability of professionals on the front line to recognize risk, intervene proactively, and make things safer for patients.

Last week, I went back to this pediatric infusion case, reviewing what a system analysis of the medication use process and seminal medication safety research tell us about risk. The take-away lesson? Errors that originate upstream are more likely to be discovered and corrected before the error reaches a patient while errors that originate close to the point of administration are less likely to be detected.


Here's another image prepared using the same seminal medication error data, illustrating more take-away lessons: errors that originate downstream are not only more likely to reach a patient, they are more likely to cause harm when they do.



Source: Leape L; Bates D; Cullen D; Cooper J; Demonaco H; Gallivan T; et al.Systems analysis of adverse drug events. ADE Prevention Study Group JAMA. 1995;274(1):35-43.

Now, let's look at the details provided in the original post:

• the patient was 3 years old
• the care setting was one that routinely cared for pediatric patients
• the patient was receiving an IV medication
• the medication infusing was not commonly used
• the professional staff were not using a drug administration protocol familiar to them
• no written guidelines for how to prepare the infusion or administer this particular drug were available
• the medication infusion required titration (the dosing unit provided in the post: mg/kg/hr)

On-duty personnel at the end of the night shift were observed struggling to perform the calculations needed to titrate the prescribed dose (0.84 mg/kg/hr) while adjusting doses in mL/hr increments. Calculations of this nature are possible, but they are complex, error-prone, and more likely to be botched when performed by fatigued workers. It's also high-stakes work that's occurring while the infusion is attached to the patient, about as far downstream as you can go in the medication use process.

A routine medication, dose, and rate check performed by the oncoming nursing team revealed that the medication was being administered as prescribed. A change in the medication concentration advocated by the oncoming nurse simplified the rate and dose relationship, removing the need to perform complex mathematical calculation to titrate the dose at the bedside. Risk was reduced.

This represents very good work by the team at the bedside, who used tools available to them to reduce error potential. Other positive call-outs include: the setting was one that routinely cared for pediatrics; independent double-checks are part of the clinical culture; staffing was such that a seasoned nurse had time to consider risk-reduction strategies and advocate for change; and the professional culture is described as one that values inter-disciplinary communication and respect.

When I first read this case, though, I was struck by the image of front line clinicians trying to avert disaster, much the same as an airline crew in flight might have to work to solve an emergent in-flight problem. Where was the ground crew, I wondered? How much of the time-robbing, disaster-avoidance described by the nurse at the bedside could have been averted by better "pre-flight" processes? Do solutions implemented by bedside clinicians, using a relatively closed set of variables, yield the strongest possible risk reduction, or does the process become just a little "less risky"?

Administering intravenous medications to a pediatric patient is a high-stakes activity, but it is not a rare one, at least in this setting. (Spoiler alert: You're not going to find a link to a mega-document with a full-blown Failure Modes and Effects Analysis about pediatric medication infusions!)

Instead, let me leave you with a few high-level risk reduction strategies to consider, many that you'll recognize if you practice in a setting where The Joint Commission (TJC) standards frame clinical care. I'm not claiming expertise in TJC standards interpretation nor am I offering advice about what any particular organization should do to minimize risks associated with pediatric medication administration. (You can, however, find some here.)

But I like to think that anything that can be understood by the average Walmart shopper is worth sharing, so I'm closing with a few observations about how this Walmart shopper sees risk-reduction.

Risk-reducing activities are often reflected in TJC standards, and the "science behind the compliance" is often based on failure mode and effects analysis. These are things the "ground crew" should be thinking about to ensure the people on the flight deck have what they need to get the job done:

• Establish standard concentrations for all IV medications (even the ones not often used). When IV medications are added to an organization’s formulary, they are subject to specific processes (usually under the auspices of a Pharmacy & Therapeutics Committee). These should ensure that standard drug concentrations are defined and incorporated into the tools used by professionals to prescribe, dispense, and administer the medication.


• IV drug infusions are ideally prepared in a pharmacy. When operational barriers to pharmacy preparation occur--in care settings without 24 hour pharmacy or in regions where unit-based drug preparation is the standard of care--staff members who admix medication should have access to guidelines specifying the standard concentration along with detailed admixing instructions. Clinicians who administer infusions should have easy-access to sanctioned dose conversion charts.


• “Smart” infusion pumps, with drug libraries programmed to reflect standard concentrations, make weight-based dosing even more simple. Dose-checking programming modes remove the need for manual calculations while immediately alerting clinicians--just prior to administration, the last possible discovery point--if an inappropriate dose has been inadvertently programmed.

There are other high-end strategies for reducing pediatric drug errors, some in development and some already being used in clinical settings. I encourage you to share yours. As for me, it's Friday, and I'm off to Walmart!