How the Newly Vigorous Patient Safety Regulatory Environment May Be Harming the Cause of Safety

In responding to dysfunctional systems, America instinctively turns to “more regulation” (Exhibit A: today’s Wall Street). But regulation can, and often does, go too far, and – in patient safety – I believe that it now has.

Note that this comes from someone who believes that healthcare was under-regulated until recently, not a popular viewpoint (just more mavericky behavior, I guess). But you must admit that it was rather odd that until 5 years ago, I was more likely to have my order read back when I called my Chinese takeout restaurant than when I called my hospital ward with a complex medication order. (Parenthetically, the reason for this disconnect is that my takeout restaurant has a more powerful business case to avoid screw-ups – they lose a customer – than does my hospital). So now there is a Joint Commission requirement to perform “read-backs” of important verbal communications. And it’s hard to question the need for regulation when a prescription that said, “Inject 10U Insulin,” could be followed without question, despite the fact that such orders have been mistaken for “100 Insulin” thousands of times, leading to scores of patient deaths.

I reviewed the first 5 years of the patient safety field in 2004. In an article in Health Affairs, I opined that increased regulatory/accreditation pressure had been the most potent force for change in the first years of the safety revolution:

Because physicians remain highly individualistic (which causes them to resist regulatory solutions and standardization), and hospitals continue to lack a robust set of incentives to drive patient safety, regulatory solutions have arguably been the most important early step, particularly when it comes to procedural safety (creating safe systems, standardization, and redundancies) in hospitals.

At the time, I was referring largely to the actions of The Joint Commission (formerly JCAHO). (I understand that TJC is an accreditor, not a regulator, but this is a meaningless difference when a rebuke from “The Joint” can lead to a freezing of Medicare payments.) Part of my positive assessment was based on changes that TJC had enacted over the past decade. Starting in 2003, the Chicago-based organization began to issue a series of commonsensical National Patient Safety Goals, including the abolition of high-risk abbreviations, and requirements for “sign your site” and the surgical “Time Out.”

Moreover, TJC has completely re-tooled its hospital surveys. Previously, hospitals were given several years notice of an upcoming TJC visit, and the inspection consisted mostly of reviews of policies and procedures, gleaned from flipping through thousands of pages in dusty binders. A few years ago, the Joint Commission changed to unannounced visits (for our last inspection at UCSF, we had a one hour notice, which is as it should be), and the new Tracer Methodology brought the inspectors out to the floor, talking to patients and providers to get a sense of what is really happening on the ground.

Don’t get me wrong. It isn’t that I love Joint Commission visits, or that I don’t recognize that the process can be arbitrary and the inspectors a bit – how shall we say – musty. But I did, and do, believe that the overall pressure from TJC has been an immensely positive force for patient safety.

In the same article in which I praised TJC and the stepped-up regulatory environment, I predicted that problems were on the horizon:

Unfortunately, the history of regulation is beset with examples of overreaching and unintended consequences, both of which can ultimately hamper flexibility and innovation. Moreover, it is hard to regulate the creation of a safety culture and the implementation of information technology.

On this one, I’m afraid I was right, but the problem hasn’t really come from TJC. Sure, TJC has gotten into the weeds requiring complex changes like Medication Reconciliation before anybody really knew how to do it, but overall their requirements have remained measured and sensible. In fact, for every doctor or hospital exec moaning about TJC overreach, there are important stakeholders criticizing the organization for not being aggressive enough. This, of course, was predictable – if TJC was doing its job, how could there still be terrible medical errors? The result: intense pressure on TJC to become even tougher, and the recent endorsement of a private accreditation agency (DNV Healthcare, a U.S. subsidiary of a Norwegian company) to provide competition.


All of this represented a natural ebb and flow… but then states jumped into the patient safety regulatory pool. Beginning in around 2003, many states began to increase their safety enforcement profiles, using the Centers for Medicare & Medicaid Services (CMS) “Conditions of Participation” as their touchstone. Armed with immense regulatory powers, including the ability to fine hospitals (potential hospital fines in California were recently bumped from $25,000 to up to $100,000) and suspend their operations, state health inspections throughout the country have become far more aggressive and unforgiving than TJC visits.

So what’s the problem, you might ask. It’s not exactly like we’re cured the problem of medical errors, right?  Well, there is a problem: because many of the new state processes are so draconian – a single transgression can mean catastrophe for a hospital – this Regulation 2.0 environment is sapping much of the joy and enthusiasm from the safety field, replacing any effort to create a safe, collaborative culture with one that is far more frantic, with much more finger-pointing and nastiness, than the one many of us had been trying to build.

Just as importantly, rather than promoting a holistic approach to safety improvement (with a broad portfolio of investments in training, staffing, simulation, new processes and procedures, IT, and more), a state inspection leads to a laser-like focus on the problem at hand. The environment begins to resemble airport security, where even a single failure is catastrophic. This is fine for the TSA (a failure in airport security can lead to hundreds of deaths) but a little skewed if the transgressions are procedural snafus that might (repeat: might) result in harm to a single patient.

Let me briefly describe my own hospital’s situation (if he’s still reading, it is at this point that my medical center’s CEO is starting to cringe). Over the past few years, we had a couple of serious medication errors that we dutifully reported to the state, even as we began to aggressively attack their root causes. Fine so far. But the state inspections that followed – accompanied by a series of required changes that were extraordinarily expensive to implement, and often not evidence-based – have consumed a huge proportion of the medical center’s financial, organizational, and psychic bandwidth over the past 18 months.

As one tangible example, inspectors found that some medication orders didn’t precisely hew to the correct form (a typical example: ordering a medicine giving both a dose and a time range, such as Morphine sulfate, 2-4 mg every 4-6 hours prn pain). Under the microscope of a state inspection, such orders, if found too often, could have led to the closure of our hospital. Since our computerized order entry system is still a year or so away, our only viable response was to hire literally scores of new pharmacists and “medication safety nurses” whose main job was to catch these “errors” before meds were dispensed. The cost of these stop-gap and patchwork solutions: tens of millions of dollars.

Did all of these expenditures improve safety? Probably a little. Were they worth the cost in scarce safety money? No way. Could we have spent the same money and created more safety benefits? Easily. Did we have any choice but to do what we did to get out of the doghouse? None whatsoever.

I have spoken to several other hospital leaders in California (and elsewhere), and they all harbor the same concerns. But few raise these concerns publicly because – and here’s another problem with regulation – nobody wants to anger the regulators lest that triggers another visit. This is understandable, but it means that we’re accepting a status quo that may harm patient safety. That seems just plain wrong.

I had the privilege of being the keynote speaker earlier this month at a national meeting of these same state inspectors (the “Association of Health Facility Survey Agencies”), and I told them just what I’m telling you. I found them to be highly respectful and quite open to these critiques. These are good people being asked to do a tough and important job, and they recognize that overly draconian tactics and inflexibility might well cause harm. Kathleen Billingsley, who runs the surveying program for the California Department of Health Services, vowed to try to create more of a “partnership” with providers. I found her and the other surveyors sincere, and I hope they follow through on that promise.

In the meantime, the state activity is having an unanticipated (or maybe it was anticipated, in which case it was very clever) consequence. I often get asked whether there is a business case for investing in safety and quality. Of course, the facile and PC answer is “yes,” because we want it to be so. And most folks think that, even if there isn’t really a business case today, there will be one eventually.

But my own view is that there really hasn’t been a business case for safety, at least in terms of short-term payoff. By this, I mean that until recently, a hospital CEO or board was not making a wrongheaded business decision by choosing not to invest in teamwork training, or simulation, or bar coding, or CPOE. A problematic ethical decision, to be sure, but not necessarily a flawed business decision.

However, the fact that state inspectors can swoop in and require you to spend several million dollars in clean up and prevention costs does shift the ROI. I know that my hospital now regrets that it didn’t move more quickly on CPOE – though the decision to implement our Electronic Medical Record (rather than CPOE) first was completely rational several years ago, we hadn’t (and couldn’t have) factored in the cost of compliance in today’s hyper-regulatory environment. Today, we might well have made a different choice. Would that have been a good thing for patients? I’m not sure.

I hope that regulators will read this and consider the possibility that their work and style may need to be modified to further the goals of patient safety. I also hope that the regulated will recognize that the inspectors are well-meaning professionals trying to perform a difficult and crucial task – and that some of the problems they’re identifying do need fixing, and have for years.

With regard to America’s financial meltdown, we’ll all soon discover that – while there are problems with under-regulation – over-regulation creates important hazards of its own. Here’s to hoping that our patient safety regulatory framework quickly assumes a thoughtful place between the poles of under- and over-regulation – a place that best protects our patients and promotes the wisest use our scarce safety resources.

6 Responses to “How the Newly Vigorous Patient Safety Regulatory Environment May Be Harming the Cause of Safety”

  1. menoalittle October 24, 2008 at 5:06 am #

    Bob,

    Since no one else jumped in on this, I will be first, I think. Your presentation succinctly covers a complex and pervasive subject, providing safe and efficacious care to patients. Now that the magnitude of the problem is known, entrepreneurial and regulatory folks have entered into the mix, inserting costly products and edicts of unknown safety and efficacy, rife with adverse unintended consequences. The greatest strides in patient care may come from within the people…with hospitals being better off investing in their human resources to do it right the first time, in contradistinction to industry sponsored CPOE projects (laden with not surprisingly dangerous adverse unintended consequences) that have been promulgated by lawmakers affected by HIT industry lobbying. Your C-suite folks should not bemoan their clairvoyant decision about delaying their purchase of CPOE contrivances.

    Even EMRs are being depreciated (probably appropriately) as noted in Dr. L. Gordon Moore’s presentation to a meeting of the AAFP…”when you put an EMR into a primary care practice, your life is hell for the next year…THE ONLY ONES MAKING MONEY FROM MOST OF THESE PRODUCTS ARE THE VENDORS SELLING THEM.”

    And speaking of computerization and patient safety, the following is reported from the UK by 4NI.co.uk News:

    “22 September 2008

    Cancer Patient Dies After Medical Trial

    A cancer patient has died after being given an overdose during a drugs trial at the university College London Hospital (UCLH), an inquest has heard.
    Gary Foster, 27, was repeatedly given twice the amount of chemotherapy drugs he should have been prescribed, the hospital said.
    Mr Foster, a graphic designer from Waltham Abbey, Essex, was due to get married this month.
    The University College London Hospitals NHS Trust offered the 27-year-old – who had advanced testicular cancer – a place on a drugs trial, testing different combinations of drugs.
    A second man who also took part in the trial was affected by the mistake, but survived.

    It is understood UCLH’s computer system was to blame for an error in the set-up of the trial, leading to the man’s death.

    A statement released from the hospital said: “He was treated intensively and despite an initial improvement in his condition, died as a result of drug toxicity.

    “The measures introduced to address the lessons the trust has learnt include: a second separate check by senior pharmacy staff in collaboration with the supervising consultants of all protocols on the trust’s electronic chemotherapy prescribing system and additional checks by nurses of the chemotherapy toxicity for every patient before repeated doses are given.”

    A coroner ruled Mr Foster had died of lung damage caused by the overdose.”

    Patient safety is a human culture which can neither be legislated nor dominated by illusions associated with artificial intelligence of geek devised contrivances.

    Best Regards,

    Menoalittle

  2. chris johnson October 24, 2008 at 3:40 pm #

    Bob:

    Nice article, one which agrees with my experience. In the old days joint commission visits were all about reviewing folders in a conference room. And, of course, the guy assigned to judge the PICU on the last ordeal I had to go through with them was a retired child psychiatrist and totally clue-free about what we did there. The visits were all about process and not actual events.

    One other thing. I’m probably showing my ignorance here, but what’s wrong with the order “morphine 2-4 mg q 4-6 hours prn pain”? (I just wrote one like that an hour ago and I want to stay out of trouble.) Aren’t the nurses allowed to use their discretion in both dose and time interval as long as they keep within the ordered boundaries? We have CPOE and the software we have even allows it.

  3. Bob Wachter October 24, 2008 at 4:47 pm #

    Chris — we have been told by our inspectors and accreditation consultants that this sort of dual range order is prohibited. A quick review of the nursing and accreditation literature demonstrated that this turns out to be a controversial area. The best article that I found illustrating the controversy highlights this standard from TJC: “The litmus test on survey would be if two nurses would interpret the range order for a patient in the same way.”

    In response, the authors of the article note,

    we have observed that hospitals that allow range orders and do not meet this criterion fear being cited for noncompliance by accreditation surveyors who are unfamiliar with the intricacies of pain management. Establishing criteria that will satisfy accrediting agencies and maintain individualized pain management has become so confusing and challenging that some hospitals are eliminating range orders altogether. This is not the best solution for patients; for example, such a policy can delay the treatment of pain when nurses must wait for new prescriptions.

    Here we see the conundrum: while there have undoubtedly been some errors related to ambiguous range orders, a blanket proscription must cause significant harm through under-treatment. Hopefully, we’ll ultimately get the balance right.

  4. chris johnson October 24, 2008 at 8:10 pm #

    Thanks, Bob. It seems to me that more and more we’re going to “pathways” (in this case, some kind of ‘pain management pathway’) to get around this issue.

  5. DZA October 24, 2008 at 11:21 pm #

    in regards to individualized analgesic orders, my group uses a CPOE that does not permit dose ranges but does permit hold parameters and added instructions. my workaround, which i would argue is safe by it’s very nature, is akin to “4mg morphine sulfate IVP Q 3hrs prn pain… may give less”. this presents a top limit set by me but allows bedside nursing discretion. it does permit an “underdosing” error, but this type of error is unlikely to do serious harm (ie is safe), and at institutions with progressive pain management standards and reliable and consistent bedside nursing (happily ours is such a one) gives range with safety. how do you think an order constructed this way would be viewed by safety regulators?

    BTW-JCH was just here. the only clue i needed was the sudden disappearance of all the crap in the hallway. incidentally, much of that crap was used linens and assorted trash from semiprivate rooms wheeled BACK INTO the rooms as the auditors went floor to floor. i postulate that drastic and acute changes in institutional behaviors such as this could actually increase iatrogenic infections and other adverse outcomes. it certainly generated patient complaints. hmmm, it is a potentially testable hypothesis….a sort of heisenberg uncertainty/butterfly effect at work. the simple act of observing complex behaviors, and being observed, affecting the outcome of those behaviors. anyone know the JCH visit schedule?

  6. Benjamin Littenberg November 2, 2008 at 3:05 pm #

    Your analysis of the regulatory pendulum is spot on. However, you omit one of the forces driving the extreme swings of the pendulum. Those regulators work for politicians who need issues to attract the voters’ attention. Safety can be a hot-button just like release of sex offenders or a disease outbreak, and lead to the same kind of “do something now” orders from a Governor or Legislature that even the wisest regulator cannot resist.

Leave a Reply