Lack of National Reporting Mandate for Hospital Infections Hurts Consumers

This article from Forbes brings together some compelling reasons for why we need more reporting of patient safety related data: Transparency & Accountability.

These principles seem to be on everyone's mind these days (see excellent recent blog by Dr. Kent Bottles in Hospital Impact on the topic). Transparency and accountability are like motherhood and apple pie - how can anyone disagree with these values? Yet we continue to struggle with achieving both as the industry grapples with just how much transparency there should be - should we report on complications related to four procedures or two? or should we not report at all.

Unfortunately, it takes regulatory mandates to get most providers to report, which has a dual edge to it. Regulatory mandates have a way of making people pay attention and try to comply; but at the same time, these mandates tend to invoke ferocious criticism - why should the government be involved in regulating quality? are the measures appropriate? why penalize instead of incentivize?

The article mentions a great potential solution - reward providers for reporting. This should be a sufficient incentive to get initial participants who can start learning from the comparative data, and hopefully start thinking about opportunities for improvement. The challenge will still be those providers who will forgo the additional payment incentive because of their philosophical disagreements with the data, methodology for collection, fear of how they will rank, and lack of commitment to performance improvement systems. The worry from a consumer or public health perspective is that the incentive winds up further dividing those providers who are motivated and therefore more likely to improve from those who have decided not to engage and therefore may become host to poor quality and patient safety processes. Will transparency and consumerism be enough at that point to get those providers to comply? or will competitive forces conspire to take some of those providers out of the market?

Patient Safety Data from Medicare Creates a Stir

This article from the Chicago Tribune lays out nicely some of the diverse viewpoints on some of the latest quality data to be posted publicly by Medicare - complication rates. At the heart of the matter is that the data being reported is derived from claims (bills) submitted to Medicare, and therefore it was never intended by the hospital to be a reflection of quality. Given that academic medical centers are more likely to find themselves on the list of "high complication rates", they are creating a stir in Chicago, Boston, and other medical meccas about the unreliability of the data. Not only is the data being publicly posted in order to drive consumer awareness and transparency, but CMS has announced that it will penalize hospitals who are in the highest tier of complication rates. There's clearly a lot at stake for all parties.

At the same time its refreshing to hear about the perspective of an institution such as Geisinger Medical Center - "We're perfectly fine with the way CMS does public reporting data," said Dr. John Bulger, chief quality officer at the Geisinger Medical Center in Danville, Pa., which is among the hospitals listed by CMS with substantially more complications than the average hospital. "At Geisinger, we would never shy away from the number and say, 'We don't need to get better than this.'"

This needs to be the key point for posting quality data publicly - how can we improve? what does this teach us about our patient care? how do we compare to other hospitals like us? do we really believe we are as safe as we think we are? Insight can often be painful, and it is tempting to rush to judgment both from the government side for thinking that the data is reliable enough to base payments on; and also on the part of hospitals who feel that they need to attack the data in order to defend their reputation or superiority.

Performance improvement starts first and foremost with a sense of humility, a desire to improve, and realization that improvement can and must be made. Much like the commendable approach taken by Geisinger,  it would serve all parties much better if the focus were placed on the improvement rather than on the strengths or weaknesses of data. 

Hospitals Decry Low Safety Rankings

This article in the Boston Globe lays out the concern emerging from the hospital community over the "safety" data that was recently released publicly by CMS (http://www.hospitalcompare.hhs.gov/), and is also planned to be included in the "value based purchasing" calculations that will penalize hospitals who have low "quality scores".

Even those who fared well on the rankings note that this metric was not intended for the purpose for which it is being used. One can certainly glean some insights from the AHRQ indicators that are used to calculate the patient safety scores. Having reviewed the respective data for a particular healthcare institution, I did find it to be helpful and a reasonably accurate reflection of our patient care. However, as the billing data from which these measures are derived are not generally constructed with the delivery of quality clinical care in mind, the data cannot be presented as a highly reliable picture of the quality of care being delivered at an institution. 

This isn't a question of using data, or comparing hospitals, or posting the data publicly - although each of these initiatives may independently inspire criticism as well. This isn't even about those institutions not faring well in a head-to-head competition crying foul. The central issue remains that one can't take major shortcuts in data gathering if one if trying to properly incentivize and motivate the system to improve. This doesn't work in an individual medical center or clinic - as one of the first tenets of performance improvement is to ensure that the data we share with our physicians and other clinicians is meaningful and reliable - nor does it work for the entire healthcare system.

The intention may be proper, but displaying such data publicly, expecting consumers to make healthcare decisions based upon it, and furthermore penalizing hospitals for not performing better on the same scale has the risk of coming across as a desperate maneuver to reduce costs in the guise of quality.

Vigilance in the Face of Fatigue

The probe into ventilator deaths undertaken by The Boston Globe points to "alarm fatigue" as a key reason for failure of these systems. However, as the stories from the Detroit Free Press about patient safety concerns at nursing homes point out, there are numerous other "fatigue" issues involved with failure of ventilators and other clinical processes, including overwhelmed staff, under-competent staff, under-armed regulators, and an industry under financial strain.

Vigilance is often the byword we use in steeling ourselves to be watchful for patient safety, however, how do you ensure adequacy of vigilance in the face of fatigue - when the demand for it requires intensity, concentration, time, competence - all of which are at a premium even at the best of times. This is a complex problem which will not go away with exposes, fines, shame, root cause analyses, and disciplining of staff. Its a symptom of a system gone considerably awry, and a distressing illustration of the harm that can occur at the "sharp end".

Overwhelming as the task may seem, the solution will entail a careful redesign of care systems such that staff who care for patients on ventilators have the knowledge, training, experience, time, and support to ensure that they can deliver the care required. Relegating this to another case of "alarm fatigue" may be tempting, but it risks the thinking that we could solve it if only we had better alarms or more attentive staff. Staff will only be able combat the collective fatigue if they are placed into roles that they can manage, and provided with adequacy of workload such that patient care does not suffer.

Despite the mostly impressive drive over the recent decade to improve healthcare quality, we have largely failed to highlight and support adequacy of workload as a foundation of patient safety. All frontline nurses and doctors can readily point out that patient safety starts with a manageable workload. However, these repetitive patient safety failures demonstrate that we keep placing staff in "unsafe" working conditions and somehow expecting good outcomes. The accountability belongs to the leadership not to the staff. This complex problem cannot be addressed by the clinicians on the front lines; rather, it should be a call to leaders to design clinical systems more effectively and to put their staff in situations in which they can succeed.

The Regulator's Dilemma: Censure or Close a Nursing Home

This article points out a difficult dilemma, if you close a facility because its not performing well what do you do for patient access? Healthcare is a difficult business, with some barriers to entry, as well as limited supply of particular resources (including physicians, nurses, hospital beds, and/or nursing facilities). If you shut any of these resources down, you do so with the realization that this is not an entirely "renewable" resource, and also with the realization that the free market may not spring forth to supply an alternative - particularly for the poor, elderly, vulnerable members of society.

The article is part of the "Trust and Neglect" series by the Detroit Free Press which attempts to shed light on the challenges of the nursing home industry in Michigan. The first article in the series talks about the extent of the problem and adds names and stories to make the numbers real.



Many of the cases should disturb all of us:

• Staff tied a residents hands hands together with a garbage bag
• Respiratory therapist forgot to connect a patient to a ventilator; and also forgot to check his vitals
• Nurse fell asleep on the job allowing a patient to wander out of the facility

The reasons behind these mistakes are theoretically remediable:

• Nursing homes try to cover up errors
• Staff are overwhelmed
• Homes don't follow their own procedures
• Staff are not competent for the roles they are assigned 

Some healthcare systems related issues that appear to be contributing:

• Bureaucratic rules that burden homes and their staff with too much paperwork
• Differences among inspectors about interpretation and application of rules
• Regulatory systems dependence on "fines, surveys, and lists" to censure chronically underperforming homes 

The common thread here appears to be lack of staffing, which results from low operating margins, which are only going to come under increasing pressure as the industry has recently gone through an 11% cut in reimbursement. Where will the solution to protect patient safety come from when the staff are already overwhelmed and the industry is going to face increased financial pressures?

We need more aggressive redesign of care processes, simplification of protocols, reduction of bureaucracy and paperwork. Perhaps this can lower costs of operations, and increase the amount of time that staff can spend on patient care. 

Different business models have to emerge that are able to provide adequate, safe, quality care. Perhaps the "accountable care organization" model now being promulgated for Medicare patients - which rewards practitioners and facilities for reducing cost and improving quality - can provide a new financial model for facilities to improve their care processes.

A new regulatory framework also has to emerge to solve the "dilemma" posed by this report. In addition to fines, inspections, surveys, and lists, more aggressive action should be taken to ensure that staffing is appropriate to provide adequacy of care; or perhaps repeat offenders should be "mandated" to implement a structured, supervised "performance improvement system". 

Should Patients Get Direct Access to Their Laboratory Test Results?

I blogged recently about "breakdown in communication of test results" and subsequently saw this piece in JAMA referencing some of the same issues and discussing whether "patients should get direct access to their laboratory test results".

The authors do a great job discussing the pros and cons of this potential solution to "communication breakdown" from both patient and physician point-of-view. In the context of performance improvement, patient direct access could certainly help to reduce some communication issues. However, it may not improve outcomes if patients are not clear about how to interpret the results. It could set up a further discontinuity within the system if there is confusion about "who is responsible for following up on the results?"

If part of a well-designed system, this initiative could help simplify the process by reducing steps in the communication chain and provide redundancy to reinforce patient safety. However, if implemented without careful thought to many of the questions posed by the authors, it could increase the complexity of the communication dynamic and lead to breakdowns of a different sort. 

Dangerous Assumptions & Lack of Systems: Breakdown in Communication of Test Results

This article in the Washington Post points out a disturbing, yet unfortunately widespread, cause of medical errors: breakdown in communication of test results.

The Joint Commission addresses the issue as one of its National Patient Safety Goals for hospitals. However, systems at many hospitals and physician offices are not "airtight" enough to assure the reliable transmission of information and the delivery of safe care.

The "root cause" for this error generally comes down to one of a few common factors:

• Staff too busy with their "day-to-day" routine. 
• If the result reporting process is not well-integrated into their clinical workflow, there is a high likelihood of system failure.
• No clear process identified for handling of critical lab results.
• This is probably the "over-arching" root cause for the issue at hand. If the process for communication is not robust, not well disseminated, and not monitored, it will inevitably lead to errors. 
• Miscommunication about who in the hospital/ office is handling the communication.
• Results may get reported to a non-clinical staff member (such as a Unit Clerk or Office Manager) who is not the right individual to interpret the significance of the findings.
• Physician was not reachable despite numerous attempts.
• Sorry to say that this is probably one of the most common reasons for failure to relay critical test information, particularly in a busy hospital setting. 
• Lab techs, nurses, and other staff try to follow an "established process", but the process is doomed to repetitive failure because one of the key steps is not reliable.
• Results get communicated, but subsequent "changes to reports" are lost. 
• Unfortunately, this happens often for tests that have to be "incubated" (such as microbiology reports) or "reviewed" (such as radiology reports).
• Test results that cross clinical settings (such as hospital to provider office) are prone to getting lost. 
• Every provider and every hospital may have a different mechanism and process for communicating critical test results. This increases the complexity of the process exponentially, and makes it more prone to break down.
• When there is not "one standardized process for communication" across the local healthcare system, how do we expect individual physicians, staff, and even organizations to ensure reliability?

The system described by Dr. Eric Poon at Brigham and Womens Hospital is a constructive approach to this problem:

However, this system is mostly reliant on an electronic interface, which may not be applicable to all providers in a diversity of clinical settings. In addition to using electronic systems wisely, healthcare providers should remember the key principles of performance improvement for systematically reducing the likelihood of error in this vital clinical process:

• Undertake a root cause analysis for any specific errors at the respective institution
• Undertake a FMEA (failure modes effect analysis) to formally assess the likelihood of systems breakdown for key processes
• Ensure that a specific process is laid out and communicated widely to all staff and physicians
• Build in redundancies to ensure that an "adequate safety net" is in place to catch errors before they affect a patient
• Monitor data both on outcomes (errors reaching the patient) as well as critical steps in the process (such as success in test result handoffs)
• Report data widely to staff, physicians, leadership
• Provide feedback to staff and physicians on individual errors
• Incorporate education about process into staff orientation and ongoing training forums

LED Displays in Hospitals Improve Compliance through Feedback

I recently blogged about "The Power of the Feedback Loop" and then came upon this article from The New York Times about an interesting application of this concept that I thought I should share.

ICUs in North Shore University Hospital in Manhasset, NY are using LED displays to provide instant reminders and feedback to staff:




The initiative is based on a system developed by a private company called Arrowsight. It employs video-surveillance of staff as they enter and exit patient rooms. Random snippets of video are monitored by employees in India who rate each event as pass or fail. Nurse managers receive close-to-realtime information about hand hygiene rates, and the LED displays provide positive reinforcement when the hand hygiene rate for the shift meets the target.

The results:

• Hand hygiene compliance has improved from 6.5% to over 80%. 
• Rates of hospital-acquired infection are reported to have dropped. 
• The culture of the institution with respect to hand hygiene appears to be changing. 

The concept of analyzing video tape for compliance with clinical processes and converting that into data and feedback is highly intriguing, although privacy advocates will be duly concerned about the monitoring implications. The cost of the technology may also prove to be limiting to widespread dissemination. However, I love the innovativeness of the solution, its incorporation into the clinical workflow, and its harnessing of the power of the feedback loop in changing behavior.

Leadership & Business Model Failing to Support Patient Safety & Quality

In this post by Paul Levy, former CEO of BIDMC, and staunch patient safety advocate, points out the failings of healthcare leadership in creating a culture of safety.

Juxtapose his call-to-arms with this Letter to the Editor written by the Illinois Hospital Association stating that "hospitals .. firmly believe in providing the highest level of safe, quality care to every patient" (which comes on the heels of a fairly damning report recently in the Chicago Tribune about lack of investigation about patient safety concerns at Illinois hospitals).

I can't think of any healthcare entity or practitioner who would not stand behind a statement similar to that issued by the IHA - after all its what we all believe healthcare should be about (and it is the politically correct statement to make). Healthcare institutions and leaders are not deceiving us when they say they believe in quality and safety, and that they are working towards improving these outcomes. Yet the industry is still collectively failing to meet the needs of patient safety and quality. How then do we explain the gap between "what we all believe" and "why we are failing"? I would point first to a few logical disconnects:

• belief does not equate to action
• belief does not equate to the right action
• belief does not equate to primacy of belief
• belief may not equate to sufficient action

We can justify our belief in patient safety by committing some resources to the activity, and undertaking a few choice initiatives. However, this does not necessarily lead to the outcomes that ultimately everyone is concerned about - demonstrable improvements in quality and safety, and organizations driven by the same. The rate of improvement that these activities can sustain may not even keep pace with the increasing complexity of the healthcare delivery system, which continues to yield unfortunately visible, dramatic signs of failing the patient.

As Paul Levy states, quality and safety have to be " internalized it into .. decision-making and process improvement efforts". However, this approach is also vulnerable to the following challenges:

• quality and patient safety are as yet not adequately measured
• institutional leadership is generally constrained by the rules of the industry
• societal leadership has to change the rules of the game 

If the primary objective for hospitals is revenues and volumes, healthcare executives will inevitably subjugate patient safety to these business drivers. Our healthcare institutions need to be rebuilt with the ideas of maximizing patient safety and quality first, and restructuring the business model around quality and safety. Pay-for-performance (P4P) programs, quality measurement, public reporting, value based purchasing are all initiatives that support the "quality and safety movement", however, are not sufficient to transform the industry and its leaders to "think quality first". They are seen by many within the industry at worst as "bureaucracy", "regulatory burden", "added costs", and at best as "nibbling around the edges". Attempts to measure additional dimensions of quality and to increase transparency are generally met with similar immediate reactions. The industry is not generally rushing forward to embrace and evolve into a quality-driven model. This is largely to be expected, as the current business model does not allow for the typical healthcare institution to be "built around quality".

The Accountable Care Organization (ACO) concept is generally a push in the right direction - at least it is an attempt to create more of a platform for quality while reducing the model's demand on volume for profitability. However, the dimensions of quality measurement as yet are significantly limited - is the patient's experience of safety truly measured by patient satisfaction surveys? are physicians' views about quality incorporated into the "core measures"? are nurses' concerns about safety incorporated into "hospital acquired complications"? And even though the ACO model changes the focus from "doing more for every patient" it subtly subverts the business incentive to "do as little as possible (within the limitations of measurable quality) for every patient" and "sign up more patients (through hiring more physicians) to join the ACO".

In order to drive real transformation of the industry - rather than just "payment reform" - a true "quality-driven" business model needs to be developed. As opposed to the "volume" or "frequency" turnstile, we have to develop a "quality and safety" turnstile. Quality and safety should be measured at the level of the individual patient, and drive the payment mechanism only if deemed adequate. Institutional leadership can only rise to the challenge if the rules of the game are changed at the societal level.

Let’s All Feel Superior: Individual Failures Require Leadership, Systems, and Culture to Bind

I read this op-ed piece from David Brooks in the New York Times with great interest. While talking about the Penn State scandal, Brooks identifies several fascinating psychological constructs that we use as individuals to avoid taking the "right action":

• Normalcy Bias - "shut down and pretend everything is normal" 
• Motivated Blindness - "don't see what is not in their interest to see" 
• Bystander Effect - "the more people are around to witness the crime, the less likely they are to intervene" 
• Self Deception - "we attend to the facts we like, and suppress the ones we don't" 
• Blind Spots - "when it comes time to make a decision, our thoughts are dominated by thoughts about how we want to behave; thoughts of how we should behave disappear" 

His key point is that we're in denial of "the underside of our own nature", which I would have to believe is likely true. However, as Brooks also points out "in centuries past people built moral systems that acknowledged this weakness .. they helped people make moral judgments and hold people responsible amidst our frailties." I would argue that this is the real weakness. Human nature has not changed, but society has evolved into such a complex organism such that the systems to keep "human nature in check" are not functional.

But this is ultimately what is means to be a leader - whether spiritual, political, or organizational - and to create an organizational culture: help to guide human nature - despite its many failings, blind spots, self deception, and biases - to achieve a purpose and pursue actions that no individual would undertake on his/ her own.

Many of us see these individual failings consistently in our healthcare organizations, which have largely been build around the notion that failure of behavior is an individual failing. Sometimes these individual failings are just as egregious as the examples used by Brooks when staff, managers, and executives "stand by" and allow for actions to be taken that lead to patient harm in the guise of "its just the way things work around here".

However, as many industry leaders have been pointing out - particularly the stalwarts for patient safety - individual behavior may contribute to a failing, but it is not the cause. The system (i.e., culture or society) that allows that behavior to be manifest is the cause. If we want better behavior, decisions, actions or ultimately the outcomes which result from these behaviors, then we have to work with greater diligence to build the leadership, systems, and culture that will allow individual failings to be overcome or at least held in check. The likelihood of harm is just as great, and not at all excused because the victims are sometimes unknown, or because it happens within a building and not on the street.

Hospitals Reduce Bloodstream Infections: Performance Improvement in Action

This article from The Tennessean provides a nice overview of gains made by hospitals in the state to reduce bloodstream infections related to catheters - a hospital-acquired condition that can be deadly. The secret to their success? They applied principles of performance improvement:

• The State Department of Public Health started collecting data across institutions.
• US Department of Health and Human Services established a benchmark for all hospitals to aspire to.
• Hospitals started following a "safety checklist", including the following elements:
• thoroughly wash hands
• wear sterile protective garb
• avoid putting catheters in the garb
• cleanse the infection site 
• remove catheters as soon as they are no longer needed
• Centennial and Methodist University Hospital committed to transparency by posting their numbers on their own websites.
• Vanderbilt increased awareness about the initiative by posting signs counting "days since the last infection".
• Vanderbilt developed an "antibiotic stewardship program" to closely monitor use of antibiotics which has helped to reduce the incidence of multi-drug resistant pathogens.

It is gratifying to see initiatives like this taking hold as other states have also been reporting positive outcomes with their collaborative efforts. However, there is a lot of improvement work to be done! Federal and state officials should provide a greater impetus to such initiatives - lead more collaborative efforts, aim for higher targets, and try to achieve the gains faster. The hospitals that have not achieved gains at the same rate as others may need more assistance from the state or other agencies in setting up their performance improvement systems.

Performance Improvement Tip of the Day: A "SMART" PI Project Checklist

In this "list-obsessed" world of ours, its hardly a surprise that one would create a "Checklist for Checklists". In healthcare, the call has gone out from Drs. Peter Pronovost and Atul Gawande to introduce Checklists as "patient safety tools". They are particularly useful when dealing with complex systems, and for reminding personnel of "basic operations" that can't be forgotten. When used properly, Checklists can ensure that necessary aspects of care are delivered, and free up highly trained personnel to think about when to deviate from a Checklist, and focus on higher level decision-making.

As we make our performance improvement teams more results-oriented, the Checklist becomes a useful tool to help keep the projects themselves on track. This helps to remind our project leaders about the steps they need to take in order to ensure that project goals will be met.

The "SMART" PI Project Checklist below - in keeping with our "SMART" theme - helps to provide a relatively simple, easily implementable framework for delivering improvement. In this context, "SMART" reminds us of the following:

• When designing Goals, Feedback, and Profiles, keep them "SMART" in order to ensure efficiency and minimize the likelihood of resistance or conflict. See prior blogs on each of these topics for more details.
• The "SMART" PI Project ensures that we:
• "Specify" ownership of the project.
• "Monitor" results in an ongoing fashion, and ensure that others in the organization help us with the monitoring.
• "Analyze" the current system in order to properly develop Goals, identify Barriers, and identify the Critical Processes that need to be undertaken in order to deliver results.
• "Redesign" the system intelligently building in decision support to ensure reliable outcomes, and enhancing safety mechanisms to catch problems real-time.
• "Train" ourselves, our staff, and our leaders to deliver better results using proven "behavioral management" techniques of Feedback and Profiling.

While this list can be individualized for the needs of particular institutions, the elements included are generally those that I find essential in delivering results. As you work through the list, you are introducing greater reliability into the system and establishing an increasingly mature culture of improvement.

Performance Improvement Tip of the Day: Standardization of a PRN Protocol

A simple example of standardization - a short list of PRN (as needed) medications that were most likely to be ordered for inpatients.



What led to the development of this initiative?

Physicians were frustrated that they were getting calls at all hours of the day - and particularly night - for relatively "simple issues".
Nurses were frustrated that they could not give their patients relief for "minor" and common symptoms - for which many of the medications were over-the-counter, until they heard back from the physicians. The delay in some cases could be hours.
Patients naturally bore the impact of this, not getting timely relief.

It took some work to come up with a set of medications that the medical staff and nursing staff could agree upon to be administered for most patients without a direct medical evaluation. However, once the "PRN protocol" was created, and integrated into the General Admission Order Set, it immediately led to a reduction in what was perceived by all staff to be "unnecessary phone calls".

Patients received care more promptly. The "first-line" treatments for "minor" symptoms could now be initiated by nursing judgement, and rarely led to a request for "second-line" treatments.

Standardization by using a common set of medications also improved safety by reducing the complexity of the system, by increasing clinical familiarity, and by having parameters and precautions "built-in" to the order.

It allowed the communication between nurses and physicians to rise to a "higher level of practice" rather than to consume their valuable time chasing each other about the routine.

Caveat: Even "minor" symptoms can turn out not to be minor, particularly in patients admitted to a hospital. Clinicians are right to be worried about over-standardizing care through the use of protocols, however, a well-developed protocol when balanced with continuing exercise of clinical judgment can streamline the care process dramatically.

Medication Errors Affecting 2000 Patients at Lifespan Hospitals: Need to Improve Care Transitions

The issue of medication errors affecting 2000 patients at Lifespan Hospitals in RI is understandably catching the media's and public's attention.

Although the apparent explanation offered by the hospitals is "software error", the article in FierceHealthcare quotes RI State Health Director, Dr. Michael Fine: this represents a "risk in the handoff process" and requires a "more robust team approach for care transitions".

It may be a relatively subtle mistake that was made - substitution of time-release medications for shorter acting formulations, however, a robust "care transitions process" that the article alludes to could have provided a safety net to catch mistakes made by a software glitch:

• At the time of discharge, it is a common expectation at US hospitals that a formal "medication reconciliation process" be undertaken. 
• Many, if not all, hospitals struggle with medication reconciliation due to the complexity of the task. 
• However, the minimum requirement that all institutions should be able to uphold is that the physician discharging the patient review all discharge medications explicitly, and indicate whether each should be continued or not. 
• The nurse discharging the patient should also review the medications ordered by the discharging physician, providing a "double-check" of the medication reconciliation process. 
• Prior to discharge, the discharging nurse (or delegate) should review the discharge medication list with the patient in order to ensure that the patient understands which medications to take and the indication for each.  
• While many patients may not be in a position to question a subtle substitution of formulations, the aware patient and/or representative family member can often be a "triple-check" in the medication reconciliation process. 
• Subsequent to discharge, the patient's primary care physician and visiting nurse (if involved) should review the discharge medication list to ensure clarity regarding each medication prescribed. 
• These clinicians may be at a disadvantage at determining which medications may not be appropriate since they were not involved in the inpatient care, however, their astute judgment can be a "fourth check" to protecting patient safety. 
• The corollary discharge documents - usually a discharge summary, medication list, and "referral form" - should be completed in a timely manner by the inpatient clinicians and communicated reliably from the hospital to the outpatient care providers. These documents can provide an essential link in the communication chain to ensure safe patient care. 

For an error not to be caught by any of these "safety mechanisms" does imply that there are more serious issues than a software glitch, and Dr. Fine is right to identify that the "care transitions" process needs to be carefully reviewed and may need to be redesigned.

Performance Improvement Tip of the Day: Embed the Desired Practice into the Clinical Workflow

This document illustrates a very simple improvement principle - in order to ensure that the desired practice is undertaken, embed it within the clinical workflow.

In this example, ordering of DVT prophylaxis was embedded into the tool that clinicians used to admit all patients - the General Admission Order Set. Multiple prior iterations of a separate "DVT order sheet" were tried but failed, because it was "yet another document that clinicians had to remember to grab".



Once the desired practice was built into this order set, the resultant ordering of DVT prophylaxis improved considerably. The issue in this case was not that the clinicians were unaware of the need for DVT prophylaxis nor that they were resistant to the same, but simply that the appropriate reminder (AKA checklist) was not presented to them at the time when they were most likely to write this order. A simple reminder, including the most common orders for DVT prophylaxis, provided at the most clinically opportune time gave the clinicians the precise "decision support" they needed to ensure safe practice.

The DVT Prophylaxis section is highlighted here:


We also included "decision support" on the reverse-side of the paper-based orders so that staff had an immediate reference handy to support their decision-making:

Could Performance Improvement Systems Reduce Communication Errors, Malpractice Payments?

From this recent article in FierceHealthcare, failures to communicate have resulted in 40% increases in malpractice premiums from 1996 to 2003.

Some of the common reasons cited for communication breakdown: "Physicians and patients might not receive results, report findings may be delayed, and there may by lengthy turnaround time. Together, these three communication failures across all specialties totaled $91 million in payouts in 2010, compared to $21.7 million in 1991."

Automation is touted as one of the potential solutions to this issue, which it certainly is. Automation can take out the human element from the workflow equation, which generally improves the reliability of a process (as long as the workflow is otherwise designed properly). However, automation has its limitations as well:

• The wrong fax numbers may be entered in the system, thereby sending documents either to wrong parties, or to a document neverland. This can create more quality and risk issues.
• Changes to the system can lead to breakdowns in other parts of the system, requiring constant vigilance.
• Not all parties use the same technologies thereby making the connectivity more problematic.
• Automation can convey the false-sense of reliable transmission, but it is generally unidirectional, and as such, may not be able to ensure that the recipient has actually received the document (no matter if the fax machine confirms that it was sent) nor that the matter is handled with the same urgency as the sender intended. 
• In the "clinical emergency arena", nothing beats the direct phone call, provider to provider, to ensure that the communication and intended actions do not fall through the cracks.

Beyond issues with automation, however, the biggest opportunities to improve communication lie in developing and implementing reliable systems for communication:

• Are there clear protocols for which tests and what test results are considered critical, thus, requiring "critical communication"?
• Is this list as appropriately narrow as it can be, and are the cutoffs for the test results truly clinically meaningful? 
• Having a comprehensive list can dramatically increase the number of times "critical communication" is expected to take place, and increases the likelihood of "communication fatigue" or "overload".
• Is there a clear process for "critical communication"?
• Is there a reliable system for training and re-training all staff in the "critical communication" process?
• Is the process for "critical communication" as streamlined as it should be?
• Are there staff involved in the process that don't need to be?
• Are the key steps in the critical communication process being monitored and measured?
• How often do the steps breakdown in the critical communication process, and when they do, what action is taken to address the systems or staff issues related to the breakdown?

Unless an organization develops a performance improvement framework for critical communication - implementing systems, defining protocols, redesigning processes and workflow, and monitoring data - it is unlikely that any technology or automation will resolve the communication breakdown.

Tackling Delirium in Hospitalized Patients with a Structured Approach

This is a great article in the New York Times by Susan Seliger about the hazards of delirium in hospitalized patients. Delirium in this setting is not uncommon, but is poorly understood and addressed by clinicians and organizations. Unfortunately it is fraught with perils for patients.

Fortunately, there are protocols and systems that can be implemented to reduce the likelihood of delirium.

Dr. Sharon Inouye, who is referenced in this article, helped to pioneer the HELP (Hospitalized Elder Life Program) that seeks to reduce delirium by increasing social interactions with hospitalized elders. The goal of this program is to return patients to home close to the functional capacity that they came in to the hospital with. One of the hospitals that I have worked with implemented this program very successfully.

Other strategies to improve delirium include reduction of the medications that can lead to delirium, improving ambulation, and attention to hearing, visual, and eating impairments.

Alcohol withdrawal - unfortunately also quite frequent in elders - is a fairly common cause of delirium in the hospitalized patient, and can be tended to with careful history-taking, as well as protocols for close monitoring.

Undiagnosed or untreated pain is often a common cause of delirium, which somehow continues to evade clinicians. Sometimes clinicians get stuck between the "rock and hard place" of treating the pain with narcotic medications which can then in turn also lead to delirium.

Ultimately, this "hospital acquired condition" can also be mitigated by taking a systematic, performance improvement oriented approach to identifying patients at risk, implementing strategies to reduce the risk, and monitoring outcomes closely to ensure that the intended interventions are indeed in place.

Respectful Management of Serious Adverse Events + Daily Safety Check-in

Combine the thinking of: 

• the paradigm proposed by Jim Conway, Frank Federico, Kevin Stewart, and Mark Campbell of the Institute for Healthcare Improvement about a framework for respectful management of serious adverse events in this article from HealthLeaders Media 

with: 

• the concepts proposed in this article from Patient Safety and Quality Healthcare by Carole Stockmeier and Craig Clapper about instituting a "daily safety check-in" 

and you start to get the makings of a powerful real-time and retrospective risk mitigation and patient safety system, which incorporates the following elements:

• Proactive, immediate, and retrospective responsiveness
• Heightened awareness of risk
• Early resolution of problems
• Accountability
• Leadership presence at the frontline
• Involvement of the frontline staff in safety
• Furthering the culture of safety
• Team-based approach to improving safety
• Development of a structured plan for dealing with safety issues
• Structured risk assessment and mitigation

Thanks to the vision and insights shared by the authors above and for their leading the charge to develop a patient safety system we can all be more proud of.

USA Today: Medicare Inspector General - "More tracking of serious errors needed"

Have to applaud this report since reporting and tracking are often the first steps towards improvement:

In an excerpt from the article - "That means that those hospitals ... don't learn from their mistakes, Inspector General Daniel Levinson writes .. No one tracks the effectiveness of policy changes or how the hospitals actually correct mistakes." 

These statements get at the real crux of the matter - serious adverse events are happening everyday throughout the nation's hospitals. However, are we learning and responding appropriately to the same in order to reduce the risk of injury for subsequent patients? 

In addition to reporting and tracking of the errors, we need robust risk mitigation and performance improvement systems to ensure that patient safety is being improved:

• Root cause analysis of these events to get to the incipient causes of the errors.
• Corrective action plans that hospitals are held accountable to in order to ensure that the root causes are mitigated.
• Failure modes and effects analyses (FMEAs) to ensure that we understand all the myriad ways systems can fail.
• Systems redesign to ensure that the culture, workflow, clinical actions, and behavior that led to the error are truly modified.
• Auditing and Monitoring of systems and outcomes to ensure that the intended changes have gone into effect.
• Real-time assessments of adverse events in order to mitigate immediate risk to the patient in front of us - in addition to retrospective analyses or prospective improvements.

As outlined above, many actions can be taken to improve patient safety. Unfortunately, as hospitals continue to be impacted by the recession - reductions in volumes, cut-backs in budgets, and cuts in reimbursement, patient safety staffing and patient safety systems are not as robust as they need to be.  We can mandate more monitoring, reporting, and tracking of errors, but unless we work with hospitals to develop the infrastructure required to build patient safety systems, patients will continue to be harmed by the very systems intended to help them.

How to Address Cleanliness Violations at Hospital? New Building or Performance Improvement

The LA Times posted this article about UCLA Harbor Medical Center's safety violations stemming from "lack of cleanliness".

The plan for correction: "Los Angeles County is spending nearly $323 million to construct a 190,000-square-foot building at the hospital that will replace both the surgical facilities and the emergency room. "

As some of the comments to the story point out, its not all about the facility - its also about management and leadership. I've worked with organizations that have faced even older infrastructure, and yet we have improved their cleanliness with good old fashioned management and systems:

• Leadership recognized the difficult situation but resolved to address it.
• Management threw its shoulder "to the flywheel" and made it turn.
• A voluntary team of "cleanliness inspectors" toured the hospital on a regular basis and documented cleanliness of high risk areas - in a manner that could be measured, easily disseminated, and compared.
• Performance improvement systems were installed to ensure improvement:
• Pulled together a "performance improvement team" to oversee the issue.
• Measured the results.
• Made the measurement consistent and reliable.
• Disseminated the findings.
• Developed policies for cleanliness that could be adhered to by all.
• Ensured accountability by reporting on the team's findings to the governing body through the appropriate oversight committees.
• Made the "clicks of the fly wheel visible to all" by showing that some departments (that were just as resource-starved as the others) were able to make improvements.

Perhaps a new building is necessary at UCLA Harbor for many reasons, but in order to reform healthcare, we as a society have to start recognizing that adding costs (staffing, equipment, and buildings) to solve a problem are not sustainable interventions. Improving leadership, management, and systems are.


ADDENDUM 11/1/11: This story published subsequently in the Daily Breeze provides further details about a corrective action plan being put into place at UCLA Harbor. "Those fixes include a reorganization and restructuring of the hospital's infection prevention and control unit, identifying problems with the physical plant and enhancing efforts to assure staff members are washing their hands and practicing good hygiene." All of these actions appear to be appropriate, however, the key to real and sustained improvement from the "corrective action plan" is to ensure that the actions are monitored regularly, and that comprehensive performance improvement systems are implemented.