RTCrawford’s Weblog

(Painting of Magnolias in a Bowl referenced later)

It is during the summer that I take a break from teaching and focus on consulting projects that keep me current in the field of management and prevent me from feeling sequestered from the competitive healthcare market. This posting is devoted to some common healthcare management themes identified during recent consulting projects and conference presentations.

What’s Broke, Who Did It, and How To Fix It.

Quality care is often boiled down to identifying what is broken, who is at fault, and what can be done about it.

There are a host of challenges confronting the leadership of any organization when attempting to get past the first and second stage, and, often as not, the third stage is never productively addressed.

What’s Broke.

Every participant in the healthcare chain is convinced they have an excellent grasp on what is broken, due to daily exposure to the breaks. The problem comes when trying to prioritize or triage the severity and calculate the impact on the larger organization. It is here (the triage) where strategic capital allocation decisions are made and where the scarcity of relevant data becomes abundantly evident. That which plagues my working life may not be more important than that which frustrates my colleague in a different department or service line, and neither of us may have the numbers necessary to quantify our discomfort or the ability to put that angst into terms found persuasive by the strategic leadership.

Often overlooked is the metric of staff turnover and the cost associated with that turnover. If the boss is frustrated, the staff may be equally frustrated (probably more so) and those in high-paying and high-demand fields are often willing to vote with their feet, under the “grass is greener” theory of job hoping. Figure out the yearly incomes of those who left during the past year (increasing it by a third to cover benefits) and multiply this by one-third (i.e., 0.33) to arrive at a very loose and hyper-conservative estimate of turnover costs (some use higher percentages).

To get at the frustration costs, do the same calculation for the marginal costs (difference) between, either, the low or average turnover organization and yours. If unable to find the data on average turnover in the market, compare your rate to other departments with similar complexity and staffing requirements (imaging would serve as reasonable surrogate for the lab, for example).

This is an imperfect measure, to be sure, but it gets at an often overlooked cost of inefficiency — failing to recognize that the workforce is a form of customer, as well. This estimate, however, is a conservative one, since it does not factor in all the ancillary expenses associated with advertising the new position, interviewing candidates, or frustration by the remaining staff as they work harder until the replacement is on-board and fully productive — to say nothing of the costs associated with increased customer dissatisfaction.

Who’s On First.

The “who is to blame” aspect of the effort is problematic. We tend to know that our staff (this department, clinic, ward, etc.) is working hard, so the blame must reside elsewhere. This informs the start of most CQI efforts, and it is often frustrated / exacerbated by the belief that we should avoid casting blame in an effort to cultivate a blameless culture. This tension between good behavior and secret suspicions, however, is one root cause behind the “Forming, Storming, Norming, and Performing” stages of team development, where it prolongs the Forming stage and shows up most prominently in the Storming stage. To get beyond this ruse of social interaction, where “blame” is the word never to be uttered but is on the tips of everyone’s tongue, I have taken to defining CQI as:

1. A data-driven,

2. Incremental improvement

3. Of process quality

4. To the benefit of the customer.

This definition is admittedly simple and ignores the focus on reducing variation — which is, instead, presented as a first-principles obligation of management. In fact, Deming argued that maximizing process efficiency IS the job description of frontline management, and the only way to succeed at it is to narrow variation, as originally defined by Walter Shewhart. If variation reduction is a management obligation and is taken as a given, this allows us to define CQI as having the four prongs listed above, where “data-driven” occupies first position.

Data is listed first because it is far easier for management to focus on the warm and fuzzy team-building and leadership aspects found in Deming’s 14 Points than on the number crunching emphasized by Shewhart, Juran, Ishakawa, and, more recently, Taguchi. More importantly, it is the quantitative that allows process improvement teams to focus on the problem at hand rather than the question of who is at fault. More tangibly, we achieve this change of focus more readily when creating process flow charts, Pareto charts, histograms, run charts, regression analysis, and process control charting — all of which possess a quantitative component if done properly. Get rid of the quantitative and the blame game becomes the only game in town and the only means of satisfaction — because the satisfaction of resolving the original problem will remain illusory.

How To Fix It.

The how to fix it issue is too often perceived as an exercise in reinventing the wheel. Wheels are tremendously hard to invent and do not become easier if reinventing them. In fact, I am painting a still life (see picture above) of a large magnolia blossum in a water-filled glass bowl (top down perspective), where the mouth of the bowl is more narrow than the base, and, while drawing the magnolia with all of its nooks, crannies, and folds is relatively easy, getting the circles of the bowl’s exterior and its opening is particularly difficult — despite having trained at the Art Students League of New York. With circles, as with wheels, the only thing worse than no wheel is one that is hopelessly warped.

This problem of reinventing the wheel in healthcare is akin to the physician who opposes clinical pathways on the grounds that every patient is different (due to unique diagnoses, comorbid conditions, lifestyle constraints, etc.). This insistence on reinventing anything of complexity implies that our working lives are like a season of “House” episodes, where every patient or process inefficiency is unprecedented, intractable, and the challenge only capitulates to the force of human brilliance in last act (following several near-death resuscitations and just before the final credits roll). The reality, according to studies of physicians and their practices, suggests the practitioner uses a set of mental check lists — created in medical school and refined through experience. This is far different from the “House” model, but the “House” model is present in more alarming form when it is realized that physician A’s model is likely to be far different than physician B’s model based on when the two went through medical school, where in the country they practice, and with which approach each feels most comfortable. Present with the same diagnosis to two different physicians and the recommendations are likely to be inconsistent. What is the most common treatment recommendation of academic attendings at Comprehensive Cancer Centers? Watchful waiting with diagnostic follow-up to confirm progression and avoid false positives. What was the most common counter argument among physicians around the US urged to join multidisciplinary cancer teams? Lost revenues to more appropriate practitioners in other specialties.

As for reinventing the wheel, Dr. James Reason, MD, spent a career identifying the common heuristics of system and process failures and, more importantly, identifying what interventions work for each type of failure. In short, Dr. Reason invented the very wheel non-reasoning administrators insist on reinventing.

This is important for all of those Type A managers who oppose creation of process improvement teams (because they are costly, time consuming, and not guaranteed to foster success) — as though Mr. or Ms. Type A. Manager exercising independent judgment represents a superior and more assured vehicle to success. Learn and teach the heuristics of failure along with the solutions and your process improvement teams become more efficient and effective — and, more importantly, they will be needed only when the problem is, in fact, intractable and worthy of a “House” episode.

Why Can’t Healthcare Be Like Federal Express?

One conference speaker made reference to the efficiency of Federal Express when advocating adoption of bar codes on nearly everything related to healthcare — from medications to specimens to patients to, presumably, staff name cards, allowing automated identification of who prescribed and who delivered the treatment. My conference-presentation colleague was right to lament our failure to use modern technology or to learn from the successes in other industries. In fact, the heuristics of failure and the identified solutions of James Reason were adopted by the airlines, and air safety has increased dramatically, while medical error continues to injure and kill at least 44,000 Americans annually (some contend the death toll, alone, is as high as 199,000).

Bar codes, however, are the modern equivalent of ancient technology. I know this because Hugh Deaner and I wrote a White Paper on the use of bar codes in a high volume hematology oncology clinic nearly a decade ago. Looking forward and at lower cost and higher efficiency, we may want to consider RFID tags for tracking medications and specimens, or using those funky little biometric finger print readers (now found at all international airports and on the sides of modern cash registers) to replace use of patient social security numbers, medical ID numbers that vary from one institutional provider to the next, or the impersonal practice of asking the patient to identify themselves before commencing treatment (each of which was recommended and rejected at a recent conference but all of which seem ineffective with Alzheimer patients, head-trauma veterans, post-operative Neuro-ICU patients, etc.).

In short, healthcare is 10 years behind other industries, and adopting that which other industries adopted a decade earlier will not move us to the head of the line but, rather, sustain our laggard posture.

The Color of Healthcare.

Related to the previous item, a member of the audience at the same conference described the benefits of changing color codes on an assortment of items (medications with similar sounding names, patients with similar names or demographic attributes, etc.), and the dramatic improvement this produced on error avoidance. While these changes are consistent with Dr. Reason’s recommendations, the practice was arrived at locally and cost nothing to implement.

Why is this related to bar codes, RFID, and biometric scanners? Technology is expensive and it represents a change in systems. CQI, on the other hand, focuses on processes. Bar codes, RFID, biometric scanners, using Palm Pilots or IPODs to house clinical pathways, etc., represent high-dollar system choices for the senior-ranking allocators of capital budgets. Changing color codes represents a process change, and, equally important, it notes that process improvement need not represent a capital budget item to produce dramatic improvements. In fact, I’ve lead more than a hundred process improvement initiatives and my students perform their own when taking my class, and none has cost more than $1,000. The expense in CQI is in the training, not the deployment.

The Check Is In The … .

One high-volume laboratory center described with pride their effective quality assurance system of checks and rechecks. Checks and rechecks are a cost of quality, borne of an imperfect system and its processes. Philip Crosby’s research indicated that between 20 percent and 40 percent of an organization’s cost structure is attributable to such quality assurance systems or the cost of rework — doing right the second time that which would have cost half if performed correctly the first time. There are, in fact, other costs associated with reworking poor quality — including that portion of marketing and advertising designed to replace disgruntled customers with the unsuspecting new arrival.

When it Absolutely, Positively Must Be Checked … .

This problem with Quality Assurance (QA), however, is old news to those trained in quality improvement, but the description of the QA system prompted another attendee to lament the absence of sufficient staff to perform the required independent, second check. That second check, by protocol, was to be performed by another, fully-trained specialist at a facility where work flow and staffing supported just one specialist.

The solution is to craft multi-disciplinary teams, where cross training follows from the day-to-day work. While this may seem reasonable for processes requiring little training or expertise, it works in more complex settings, as well. At UNC Hospitals, where I was once the business manager for clinical oncology, multidisciplinary tumor board conferences meet daily to craft tailored treatment plans. With each case, the patient was / is “presented” at conference — starting with pathology, followed by imaging, surgery, RadOnc, HemOnc, Psych, etc. –, and each uses the opportunity to educate the others on the considerations leading to their conclusions — including projected radiology and pathology images.

Dear reader, I know what you are thinking … “Just what I need, another excuse for my colleague to lecture me about what I found boring in Med School … If I had a foot fetish, I would have become a podiatrist. If I wanted an intimate relationship with my malpractice insurance broker, I’d have become an OB/GYN or an orthopedic surgeon … etc.” This fear, however, is misplaced.

Over time, the tumor-board-conference learning has proven so robust that the presentations are an abbreviated form of short-hand (faster and more efficient than if relevant practitioners were coordinating by beeper or phone tag or, for that matter, even if simultaneously bumping into each other in the hall for a quick discussion). The conference discussions have the increased efficiency of a cross-trained Indy pit crew — dispatching more than 40 patients in a 90-minute conference, combining lunch and saving lives through a pleasant exercise in multi-tasking.

More importantly, I have seen a HemOnc identify a faint and nearly indistinguishable smudge on a CT scan that was missed by the radiologist (converting a stage I non-small cell lung cancer patient to stage IV patient and changing the recommended course of treatment completely) and, believe it or not, a surgeon successfully challenge a pathologist on whether a tissue sample represented a primary or secondary malignancy … based on the pathologist’s “lecture” for a different patient weeks earlier.

In the “Fifth Discipline,” Peter Senge describes the Learning Organization as where a critical mass of functional expertise works to achieve a common end and benefits, beyond all expectation, from the collective intelligence sitting at the table. Historically, clinical healthcare has been different, where practitioner independence is cultivated as the fail-safe to error (that is why treatments follow from physician “orders.”) It is that fail-safe, however, that led to “Crossing the Quality Chasm,” “To Err is Human,” the five-year follow-up reports in JAMA and NEJM, and, more recently, the Rand studies and the recurrent Dartmouth findings (all indicating significant short-comings in US healthcare quality and all disputed by a stalwart set of practitioners denying the common thrust of those findings). If one physician represents the best and brightest our society can produce (an assertion I believe with heart, soul, and experience), a room full of such rare talent holds special promise. And for the short-staffed hospital or lab or imaging department seeking an internal second opinion but lacking the specialists, multidisciplinary teams may represent a next-best alternative … and, perhaps, a superior approach, because multidisciplinary review is less likely to promote group think.

One Last Item.

Finally, I keep running across the problem of non-tactile staff (nurses and physicians, especially).

No, I’m not talking about providers who fail to touch patients or those lacking in bedside manner. Instead, when I ask for a show of hands indicating who walks down the hall using the hand rails, sends a spouse or friend to the information desk to ask a question (any question), or, in the wards, orders and eats a delivered meal from the same cart as used for patients, the poverty of raised hands is disconcerting.

This is especially true of the measures of housekeeping quality, given the nosocomial threat.

There are two uses for a hallway hand rail — to steady physically uncertain patients and as the repository of choice for used chewing gum.

A hand raked across a common-area wall will, either, come away with dust clinging to the fingers or it will not, but the result is a test of what is known as “high dusting” — one of the first things an inattentive or overworked housekeeper will drop from their to-do list of cleaning obligations.

The corners and edges of linoleum floors are most difficult to keep clean and represent another measure of housekeeping management (since management schedules floor services), and, day-to-day, attention to detail in this area can be assessed by looking behind the opened hallway doors for “dust bunnies.”

Eating a meal from the patient delivery cart will make it abundantly clear if the food is good, warm, and worth the cost charged to the patient and payor.

When it comes to the information desk, the “mystery” shopper is a mainstay of modern retail, and, while it would be an exercise in poor judgment to adopt mystery shoppers for brain surgery, the practice is too rare in healthcare’s non-invasive settings — especially, where the service rendered is more like a hotel.

And, most importantly, every member of the organization has a vested interest in the quality of the whole — your incomes, promotions, reputations, and salary will increasingly rely on it, as Boomers retire, margins narrow further, and the division between employer and employee becomes indistinguishable.

Indistinguishable? Absolutely.

In the near future (next five years), employees will, out of self-interest, become figurative business partners or move to where their partnership is valued and career stability can be monitored and tangibly enhanced. Simply stated, quality staff in specialty areas will manage their careers more closely, as constrained margins promote industry consolidation, and, valuing stability and career advancement, they will be quick to leave employers whose financial and operational stability is suspect or where career advancement seems uncertain and mercurial.

But that is topic for a different posting.