Tag Archives: Root Cause Analysis

Root Cause

Attack on Hospital Staff Indicates Systematic Safety Issues

By ThinkReliability Staff

On July 13, 2015, a security counselor at a Minnesota psychiatric hospital was attacked and seriously injured by a patient. Even one injury to an employee is highly undesirable and should initiate a root cause analysis in order to reduce the risk of these types of events recurring. In the case of this hospital, this employee injury is one in a long line. In 2014, 101 staff injuries were reported at the hospital. From January to June of 2015, 68 staff injuries were reported. Clearly this is an extensive – and growing – problem at the site. According to Jennifer Munt, a spokeswoman for a union which represents 790 workers, “Workers at the security hospital feel like getting hurt has become part of the job description.”

An incident like this one can be captured within a Cause Map, or visual root cause analysis. The first step in the method is to define the problem in a problem outline. The problem outline captures the what, when and where of an incident, as well as the impact to the goals. Another important piece of information that is included is the frequency of similar events. Capturing the frequency helps provide the scope of the problem.

Understanding the details for one specific incident will likely reveal systematic issues that are impacting other similar incidents. That is definitely true in this case. Beginning with an impacted goal and asking “why” questions results in developing cause-and-effect relationships. Each cause that is determined to have contributed to an issue can lead to a possible solution. Each cause added to the Cause Map provides additional possible solutions, which, when implemented, can reduce the risk of future similar incidents.

In this case, we begin with the employee safety goal. An employee was seriously injured because of an assault by a patient at the hospital. The assault resulted from two causes, which were both required and so are joined with an “AND”. First, violent patients are housed at the facility. There were no other facilities available for the patient and the hospital is required to admit mentally ill county jail inmates because of a Minnesota law (known as the “48 hour rule” because of the time limit on admissions).

Second, clearly there was inadequate control of the patient. According to the union, limitations on the use of restraints, which are only allowed when a patient poses an “imminent risk”, mean that staff members feel that they cannot restrain patients until after they’ve been threatened – or assaulted. The union also says that inadequate staffing is leading to the increase in assaults. Specifically, union officials say at least 54 more staff members are required for the facility to be fully staffed.

The issues have caught the attention of state safety regulators and government. Multiple solutions have already been incorporated, including use of cameras, a separate admissions unit for new patients and protective equipment for staff. Additional staff is also being hired. The patient involved in the attack is isolated and under constant supervision. There’s no word yet on whether the use of mobile restraints, as requested by the union, will be allowed.

Says Jaime Tincher, Chief of Staff for Minnesota Governor Mark Dayton, “These are important first steps; however we will continue to assess what additional resources are needed to improve safety and treatment at this facility.” No less would be expected for ongoing issues that have such a significant impact on employee safety.

Root Cause

How One Hospital Improved Heart Attack Care

By ThinkReliability Staff

The heart is responsible for pumping blood through the body, but it also requires blood flow to continue functioning. When the blood supply to the heart is cut off, it’s known as a heart attack and it can be deadly. According to the Centers for Disease Control and Prevention (CDC), about 15% of people who have a heart attack will die from it. Time is of the essence when treating heart attacks. Again according to the CDC, “The more time that passes without treatment to restore blood flow, the greater the damage to the heart.”

Treatment to restore blood flow is generally a balloon (which pushes aside the blockage) and a stent (which holds the artery open). In the United States, this is performed in a hospital. Although hospitals can’t control the amount of time it takes to get a heart attack victim TO the hospital, they can control the time from when a patient enters the hospital until treatment is begun. This is known as the door to balloon (or D2B) time.

A national campaign to improve the speed of heart attack treatment was launched. At that time, the typical heart attack process went like this: a patient suffered a heart attack and (hopefully) 911 was called. An ambulance picked up the patient and delivered them to a hospital. Once the patient arrived at the hospital, an electrocardiogram (EKG) was taken and transmitted to a cardiologist, who determined whether or not the patient was suffering from a heart attack. If it was a heart attack, an interventional cardiologist and other members of the heart attack team were called and made their way to the hospital. The patient was taken through a consent and surgical prep process, and then then balloon and stent were installed. At this time, the national goal was for half of patients to receive a stent and balloon within 90 minutes of arrival at a hospital.

One of the hospitals to take up the challenge was Our Lady of Lourdes Medical Center in New Jersey. In 2007, heart attack treatment was on par or better than other hospitals, with half of patients treated within 93 minutes. (In many locations it took more than 2 hours.) By 2011, treatment time was down to 71 minutes. The head of the cardiovascular disease program challenged the staff to continue to decrease the time and staff members set up a “D2B task force”. This task force looked at each step in the process for potential improvements. Some individual steps were shortened. The forms required for consent were reduced as much as possible. The time spent individually calling in all the members of the cardiac care team was reduced by having a single call ring to all their pagers. Those on the team that were on call were limited to being 30 minutes away from the hospital.

Other steps, instead of being performed one after the other, were performed simultaneously. Instead of waiting for the patient to arrive at the hospital for an EKG, it is taken in the ambulance and transmitted to the emergency room. Each step required for surgical prep is performed as much as possible simultaneously by a team. Additionally, one surgical room is reserved for heart attack patients and is kept stocked with necessary supplies.

Now the median D2B time is 50 minutes. This was demonstrated on March 29, when a patient arrived at the medical center at 1:54 AM and whose D2B time was 55 minutes. This was unusually long for the center. What caused the difference? Because the patient was a 49-year-old woman with ambiguous symptoms, the emergency room doctor waited until the patient arrived at the hospital for another EKG to verify the heart attack before the heart attack team was called.

From 2003 to 2013 the death rate from coronary heart disease has fallen 38%. Some of this drop is attributed to better control of cholesterol and blood pressure, but some is surely due to quicker treatment at most US hospitals.

The “before” and “after” process map that shows the flow of heart attack treatment at Our Lady of Lourdes Medical Center can be diagrammed visually to show how the process flows. To view the process map, the problem outline and timeline of the treatment of the heart attack patient on March 29, 2015, please click on “Download PDF” above. Or click here to read more.

Root Cause

U.S. Teen Dies from Plague

By Kim Smiley

Few people think of the plague as a present-day problem, but a teen boy died of the plague on June 8, 2015 in Colorado.  Officials believe he was bitten by a flea carrying the disease on his family’s farm although the exact source of exposure isn’t known. According to the Centers for Disease Control and Prevention, there are an average of seven cases of plague in the United States a year and a small percentage of these cases result in death.

A Cause Map, a visual root cause analysis, can be built to analyze this case and better understand how a patient died of the plague.  The first step in building a Cause Map is to fill in an Outline with the basic background information to define the issue.  The Outline includes a place to list the impacts to the goals resulting from an issue to help define the scope of the problem.  Focusing on the safety goal for this example, a death would be an obvious impact.  Next, “why” questions are used to build the Cause Map.

So why did the teen die from the plague?  There are two causes that contributed to his death; first, he was infected with the plague and second, he wasn’t treated for the plague.  When there are two causes that both contribute to an issue, both are listed vertically on the Cause Map and separated by an “and”.  So why was the patient exposed to the plague?  Officials believe that he was bitten by an infected flea.  The bacteria that causes plague lives in rodents and their fleas.  Investigators haven’t been able to identify which species of rodent was the culprit.

The teen wasn’t treated for plague because it wasn’t identified that he had the plague until it was too late.  All forms of plague can be successfully treated with antibiotics, but the window for treating the illness before it becomes life-threatening can be relatively short and plague can be difficult to identify.  It is suspected that this patient had septicemic plague which occurs when the plague bacteria enter the bloodstream directly.  Septicemic plague is caused by the same bacteria as the more common Bubonic plague, but the symptoms are different and more difficult to identify.  Rather than the telltale presence of swollen, discolored lymph nodes (also known as buboes) caused by the Bubonic plague, the main symptoms of the septicemic plague are fever, chills and abdominal pain which are very similar to the flu and other common illnesses.  In this heart-breaking case, the family of the teen understandably believed he had the flu and he wasn’t treated for the plague in time to prevent his death.

As alarming as this case is, it is important to note that plague cases in the United States are very rare and occur primarily in two regions – northern New Mexico, northern Arizona, and southern Colorado and California, southern Oregon and far western Nevada. If you are planning to enjoy the outdoors in one of these areas, just remember that the best way to prevent plague is to prevent flea bites.

Click on “Download PDF” above to see a Cause Map and Outline for this example.

Root Cause

Multiple Potential Causes for Avian Flu Outbreak

By ThinkReliability Staff

An outbreak of avian influenza (flu) H5N2 centered around Iowa in the United States has resulted in nearly 47 million birds being killed in 21 states. There is a low risk that this outbreak could spread to humans as the 1996 avian flu did. The impacts on the poultry industry have been significant: the number of birds being killed has led to an increase in poultry prices. Says Phil Lempert, “We’ve lost 10 to 13 percent of the laying hens in this country, so we’re going to have this period of time where we have less birds and less eggs. That means higher prices.”

The financial impact isn’t limited to consumers. The United States Department of Agriculture (USDA) estimates it will spend more than $500 million fighting the outbreak. The impact on poultry producers is expected to be even higher. The USDA Animal and Plant Health Inspection Service (APHIS) is studying the outbreak and attempting to put into place measures that will reduce the spread of the outbreak. Finding the causes leading to the outbreak has proven to be challenging.

We can capture the information that is known in cause-and-effect relationships using a Cause Map to better understand what caused this outbreak. The first step in the Cause Mapping process is to fill in an Outline with basic background information, which includes listing how the overall goals are impacted by the issue. The Cause Map is than built by asking “why” questions to lay out the cause-and-effect relationships. In this example, the animal safety goal is impacted due to the deaths of nearly 47 million birds. These birds were killed because of an outbreak of avian flu. An outbreak results from an initial infection (believed to have been transmitted in this case to domestic flocks by wild birds) and the spread of the disease. Based on genetic analyses from APHIS, this outbreak appears to have multiple independent introductions within the outbreak area (i.e. the transmission from wild birds to domestic flocks happened in multiple locations).

According to their Epidemiologic and Other Analysis of HPAI-Affected Poultry Flocks: June 15, 2015 Report: “APHIS concludes that at present, there is not substantial or significant enough evidence to point to a specific pathway or pathways for the current spread of the virus. We have collected data on the characteristics and biosecurity measures of infected farms and studied wind and airborne viruses as possible causes of viral spread, and conducted a genetic analysis of the viruses detected in the United States.” This means that the cause or causes of the spread of the avian flu cannot be definitively determined due to lack of evidence. When an investigation has a lack of evidence, potential causes are included in the analysis with a question mark, indicating insufficient evidence.

In this case, avian flu was potentially spread by air, by wild birds, and by human movement. Data from APHIS research indicates that the virus has been able to spread on windy days up to a half mile. A solution under consideration is more advanced ventilation systems for poultry farms that would prevent transmission of disease from farm to farm. Previous outbreaks have indicated that wild birds can not only cause an initial infection, but can continue to spread the disease from flock to flock. This evidence supports this cause, but is not strong enough to rule out other causes so all should still be included on the Cause Map. Lastly, APHIS found inadequate biosecurity (primarily cleaning and disinfecting) measures on equipment and personnel that traveled from farm to farm, which could also potentially spread the disease.

The issues found with biosecurity are a particular concern. Says Michael T. Osterholm, the director of the Center for Infectious Disease Research and Policy at the University of Minnesota, “We used to think we had outstanding biosecurity in poultry. But, except for the outbreak in 1983, which was stopped quickly, we have never been tested before.”

Osterholm and other researchers say more research is needed to screen for viruses, and develop drugs and vaccines to ensure public safety. Although the virus has not yet been shown to infect humans, the Centers for Disease Control and Prevention has developed interim guidelines on testing and treatment. APHIS continues research on how to limit the spread and the USDA, in order to offer some relief on prices, has recently allowed poultry imports from the Netherlands.

To view a Cause Map, or root cause analysis presented in a visual cause-and-effect diagram, of the ongoing outbreak, please click “Download PDF” above.

Root Cause

Contamination found in NIH pharmacy

By Kim Smiley

The National Institutes of Health (NIH) has announced that production of drugs for use in clinical studies has been suspended after fungal contamination was found in two vials of product.  The exact source of the contamination has not been identified, but a recent Food and Drug Administration (FDA) inspection of the facility that prepares the contaminated product found multiple deficiencies, including issues with both the facility and work practices.

This issue can be analyzed by building a Cause Map, a visual root cause analysis that intuitively lays out the cause-and-effect relationships that contribute to an issue. The first step of the Cause Mapping process is to determine how an issue impacted the overall goals.  In this example, the safety goal is impacted because 6 patients were unknowingly given potentially contaminated drugs.  These patients received vials of product from the same batch as the 2 vials found to be contaminated prior to the contamination being identified.  None of the patients have shown signs of illnesses, but they will continue to be monitored. Additionally, the safety goal is impacted because some patients will knowingly be given potentially contaminated drugs.  These patients are due for treatment imminently with no alternative available and the risk of delayed treatment has been determined to be greater than the risk of using the products.  The schedule goal is also impacted as clinical trials are being delayed because the necessary medications aren’t available.

The next step is building the actual Cause Map by starting at one of the impacted goals and asking “why” questions.  So why were the drugs contaminated? It hasn’t been released what specifically lead to the fungal contamination and it may never be known, but the FDA found deficiencies within the facility that could lead to contamination. The inspectors observed workers working with sterile products with protective gear worn inappropriately so that skin and facial hair were exposed.  Issues with the facility itself was also noted, both in the design of sterile work spaces and in the cleanliness of the spaces.  Inspectors determined that the air handling system for the clean rooms wasn’t adequately designed to ensure physical separation from the other spaces.  Additionally, a filter was missing on the air handling system.  The problems with cleanliness of clean rooms included insects found in 2 of 5 clean room ceiling light bays.

The investigation into these issues is ongoing and officials are working to ensure the safety of all products.  As more information becomes available, it can easily be added to the Cause Map.  Once the specific problems with the work processes and facility have been determined, specific solutions can be implemented to address the many issues found by investigators. This problem is one that clearly doesn’t have “one root cause”, but rather many causes that contributed to the problem and more than one solution will be needed to reduce the risk of contamination to an acceptable level.

Root Cause

Measles Vaccine Provides Multiple Protections

By ThinkReliability Staff

For previously unknown reasons, children who received the measles vaccine were less likely to die from infectious diseases other than measles.   According to Michael Mina, a postdoc in biology at Princeton University and a medical student at Emory University, the difference is significant.  “In some developing countries, where infectious diseases are very high, the reduction in mortality has been up to 80 percent.  So it’s really been a mystery – why do children stop dying at such high rates from all these different infections following introduction of the measles vaccine?”

Based on epidemiological data from countries before and after the measles vaccine was introduced, scientists believe they may have an explanation for this mystery that is part correlation and part causation.  So what’s the difference (and why do we care)?

Correlation means that two or more events tend to occur about the same time and might be associated with each other, but aren’t necessarily connected by a cause-and-effect relationship.  Causation means that a specific action causes a second event to happen.  A cause-and-effect relationship results from causation.   Sometimes it’s very difficult to distinguish between the two.  This is where the importance of evidence comes in.

In this case, part of the decrease in death due to infectious diseases can be considered due to correlation.  In this case, children who received the measles vaccine must have had access to healthcare, including the measles vaccine.  If they received the measles vaccine, they were also likely to receive other vaccines and treatment for other infectious diseases, meaning their death rates from other diseases were also lower.  The measles vaccine did not cause the reduction in deaths from infectious diseases, the access to healthcare did.  Getting the measles vaccine also resulted from the same cause, access to healthcare.

In addition to this correlation, epidemiological data from several countries from prior to the introduction of the measles vaccine shows that the number of measles cases predicted the number of deaths from other infectious diseases two to three years later.  Their hypothesis, supported by studies in monkeys, suggest that the measles virus actually erases immune protection to other diseases.  So, if a child gets measles, he or she loses some of the immune system’s “memory” of how to fight diseases can also be wiped out.  Preventing a child from getting the measles (by getting a measles vaccine) is believed to prevent deaths from other infectious diseases as well.

Although more testing is needed to verify the causation, scientists hope it will provide more evidence for parents to vaccinate their children.  Epidemiologist William Moss, who studies the vaccine at John Hopkins University, says “The reduction in overall child mortality that follows measles vaccination is much greater than previously believed.  I think this paper will provide additional evidence – if it’s needed – of the public health benefits of measles vaccine.  That’s an important message in the U.S. right now and in countries continuing to see measles outbreaks.”

To view the cause-and-effect relationships (both correlation and causation) between the measles vaccine and decreased mortality from childhood infectious diseases, please click on “Download PDF” above.  To learn more about the epidemiological study, click here.

Root Cause

Listeria in Ice Cream Causes 3 Deaths

By ThinkReliability Staff

On April 20, 2015, the Centers for Disease Control and Prevention (CDC) announced a recall of all Blue Bell Creameries products due to possible contamination by Listeria monocytogenes.  While the company has not yet determined the source of the outbreak, they are working with outside agencies to determine potential causes and implementing solutions to reduce the risk of food-borne illness in the future.  Says Paul Kruse, the CEO and president, “We’re committed to doing the 100 percent right thing, and the best way to do that is to take all of our products off the market until we can be confident that they are all safe.  At this point, we cannot say with certainty how Listeria was introduced to our facilities and so we have taken this unprecedented step.  We continue to work with our team of experts to eliminate this problem.”

Performing a root cause analysis can help clarify the goals of an investigation, determine the causes of the problem(s) related to an issue, and provide ideas for action items to reduce the risk of the issue recurring.  We can gather the information known so far about the outbreak in a Cause Map, or visual root cause analysis.

The Cause Mapping process begins by capturing the what, when and where of an incident.  Here, the “what” is the Listeria outbreak.  The “when” in this case is believed to have started in 2010 and continued to the present.  It can be helpful to capture any noted differences about the particular investigation.  For example, most outbreaks don’t last 5 years.  The use of genome sequencing (starting in 2013) allowed investigators to tie Listeria cases from 2010 on to this particular outbreak.  An additional difference is that Listeria can replicate in very cold temperatures.  This is unusual because freezing foods generally reduces the risk of propagating food-borne contamination.  The “where” is across the US – all products have been recalled and all plants have been shutdown, with several having been implicated in spreading Listeria.  Another useful piece of information can be the task being performed.  In this case, the contamination was discovered during random sampling.

The next step is identifying the impacts to the goals.  For this incident, the safety goal was impacted due to the sicknesses and deaths.  The outbreak of Listeria can be considered an impact to both the environmental and customer service goal, while the loss of production (no Blue Bell products are currently available or being produced for consumers) is an impact to the production goal.  The disposal of the estimated 8 million gallons of ice cream covered by the recall impacts the product goal, and the response and investigation impacts the labor goal.

The analysis step begins with an impacted goal.  Asking “why” questions develops the cause-and-effect relationships that led to the impacts.  In this case, the sicknesses and deaths were caused by a Listeria outbreak.  In order to have a food-borne illness outbreak, the food needs to be contaminated AND it needs to be delivered to consumers.  In this case, the contamination was not known because ice cream is not tested for Listeria.  There is no history of Listeria outbreaks in ice cream and testing is difficult on perishable products because of the time required.  Once ice cream products are again manufactured for consumers, Blue Bell has said it will implement a test and hold process (holding product until testing comes back negative).

The Listeria contamination results from the introduction of Listeria into the ice cream.  As discussed before, Listeria can replicate in cold temperatures.  The contamination source is likely surfaces in the production facilities or cross-contamination from other food products.  Because multiple plants are contaminated and cleanliness issues have been a concern in the past, it is likely that the outbreak is due to contamination of surfaces, on which Listeria can remain for a long time if not properly sanitized.

In addition to the test and hold process, Blue Bell is in the process of implementing a number of other changes to reduce the risk of future contamination.  Employees are being trained in microbiology and an expanded cleaning and sanitation program.  Prior to production resuming, equipment is being disassembled, cleaned, and tested for contamination and design changes that would make cleaning easier (reducing the risk of future contamination) are being considered.

While it is sometimes difficult to determine the success of solutions, the test and hold process to be used for future ice cream products should provide almost real-time feedback on the success of the programs and ensure that future problems are quickly identified.

To view a one-page PDF of the analysis and solutions, please click on “Download PDF” above.  To learn more about the ice cream Listeria outbreak, click here.  To read our previous blog about the 2011 fatal Listeria outbreak in cantaloupe, click here.

Root Cause

With $16.3B, Why Are Veterans Still Waiting for Care?

By ThinkReliability Staff

Concerns regarding the timeliness of treatment within the Veterans Administration (VA)’s network of hospitals and clinics have been around nearly as long as the VA itself. In 1995, a goal was set to have veterans seen for appointments within 30 days. VA doctors’ and executives’ bonuses are based at least in part on meeting timeliness targets. Many believe this is a key reason that waiting lists were doctored (by being kept on a separate “secret” waiting list, before being moved onto the real, computerized waiting list within 14 days of their scheduled appointment). The scandal, which is believed to have contributed to the deaths of dozens of veterans while they waited for appointments, led to much consternation and a call for significant reform to improve the waiting time of veterans.

It was found that veterans were waiting too long for appointments not only in Phoenix (where the “secret waiting list” scandal was discovered) but at many VA sites around the country. This was determined to have significant (though not always easily quantifiable) impact on patient safety as well as patient services to the large numbers of veterans who were unable to get timely appointments. (Read our previous blog about a veteran who lost much of his nose after waiting more than 2 years for a biopsy.)

In order to lessen the waiting times, $16.3 billion in spending to hire more doctors, open more clinics, and create a program that allows veterans to seek private-sector care was approved July 31, 2014. However, a study by the Associated Press has found that from August 1, 2014 to February 28, 2015, over 890,000 appointments failed to meet the timeliness goal. More than 230,000 appointments were delayed more than 60 days. While the number of vets waiting more than 30 and more than 60 days has stayed about flat, the number of appointments that take more than 90 days has nearly doubled. Some specific problem areas have been identified.

Challenges remain with the “Choice Program”: The Choice Program began to cover non-VA care for eligible veterans November 5, 2014. However, eligibility remains limited to those who have to wait more than 30 days from their “preferred date” or a date medically determined by their doctor or those who are more than 40 miles (straight line) from the nearest VA facility or face an unusual travel burden to access it.   Only some private physicians participate. The program is being expanded so that the 40 miles is based on driving distance rather than a straight line calculation, and telephone lines and other programs are being implemented to assist veterans using the program to seek care.

Medically underserved areas have the worst delays: During the government’s investigation, it was found that many VA facilities have inadequate providers for the number of veterans in their care. These areas tend to be areas that are medically underserved, which compounds the problem because civilian options in the area are also limited, limiting the effectiveness of the program that allows veterans to seek private-sector care. Says Dr. Kevin Dellsperger, chief medical officer at Georgia Regents Medical Center and former chief of staff at the VA medical center in Iowa City, Iowa, “Not a lot of medical students want to go work for the VA in a rural community medical clinic.” While 8,000 employees were added to the VA between April and December 2014, it’s hoped that increasing salaries in the underserved areas will attract more providers.

Physical space is also an issue: Any government contracting and building process can be cumbersome, and the VA has been identified as having particular difficulty managing the contracting process. When buildings are (finally) constructed, they’re usually already too small.

Enrollment is increasing: Enrollment in VA programs has been expanding rapidly. From 2002 to 2013, enrollment increased from 6.8 million to 8.9 million and spending increased from $19.9B to $44.8B.   Says Robert McDonald, Secretary of Veterans Affairs, “Today, we serve a population that is older, with more chronic conditions, and less able to afford private sector care.” It’s hoped that the increased enrollment is actually a positive, buoyed by the efforts made to increase access and shorten waiting times. “I think what we are seeing is that as we improve access, more veterans are coming, ” says Sloan Gibson, the Deputy Secretary of Veterans Affairs.

It may get worse: “The cost of fulfilling those obligations to our veterans grows and we expect it will continue to grow for the foreseeable future. We know that services and benefits for veterans do not peak until roughly four decades after conflict ends . . . we project the benefits for recent veterans in recent conflicts will peak around 2055,” testified VA Secretary McDonald.

The VA administration is asking for patience. Deputy Secretary Gibson says “We are doing a whole series of things – the right things, I believe – to deal with the immediate issue. But we need an intermediate term plan that moves us ahead a quantum leap, so that we don’t continue over the next three or four years just trying to stay up. We’ve got to get ahead of demand.”

To view an overview of these issues in a visual cause-and-effect diagram (or Cause Map), as well as some of the associated solutions, click on “Download PDF” above. To read more about the AP’s analysis, click here.

Root Cause

Prisoner escapes from hospital

By ThinkReliability Staff

A recent prisoner escape from city custody in Virginia was only one of four attempted escapes in the US over 8 days related to seeking medical care.  Examining the cause-and-effect relationships shows what led to the prisoner escape and can provide insight into improvements to reduce the risk of it happening again.  These cause-and-effect relationships can be diagrammed visually in a root cause analysis, or Cause Map.

The analysis begins by capturing the what, when and where of the problem.  In this case, the issue being analyzed is the escape of a prisoner from a public hospital in Alexandria, Virginia March 31, 2015 at about 3:00 a.m.  Along with the where, we capture what was happening at the time.  In this case, the patient was receiving medical care after a suicide attempt.  It’s also helpful to capture any differences.  Differences could be in the location, date, time or task being performed.  In this case, a few things stand out from a summary reading of the media reports available.  First, the city jail prisoner was being treated at a public hospital, and second, one of the guards responsible for the prisoner was taking a bathroom break.

These differences may or may not be causally related to the issue, but provide potential causes to consider. As mentioned, there were four prisoner escapes during a week related to medical care.  On the same day, a New Orleans prisoner escaped from a van transporting prisoners to a hospital.  The previous day, a New Jersey prisoner escaped from a hospital, and a week prior, a West Virginia psychiatric hospital patient facing murder charges escaped.

As physical and procedural security at prisons improve, fewer prisoners are escaping from the facilities themselves.  Many times, being removed for medical care is the best opportunity.  Federal prisons, which provide on-site medical care, have far fewer escapes than other facilities.  From 1999 to 2001, only one of 115,000 federal prisoners escaped.

A single trip for medical treatment itself may not be to blame for the escape attempts, but repeat trips to the same medical facility may increase the risk.  Says Kevin Tamez, inmate advocacy consultant, “Very rarely do these guys go to the hospital for treatment and all of a sudden they decide they’re going to escape.  What happens is, traditionally, inmates go to the hospital for treatment . . . they come back to the facility and they start telling other inmates . . . There is nobody more ingenious than an inmate.  They have nothing to do all day but sit around and think things up. There are ways of minimizing it, but there’s never a way to prevent it.”

Having only one guard instead of two, due to a bathroom break, is problematic for obvious reasons.  It’s far more difficult to overwhelm two guards than one.  “From a safety perspective it’s always good to have two people there,” says Gary Klugiewicz, a consultant/ trainer for law enforcement & correctional officers.  The amount of time the guards were watching this prisoner at the hospital (4 days, for reasons that are unclear) may have also played an impact.  It’s hard to keep your guard up for that amount of time.

The U.S. Marshals, who had responsibility for the prisoner at the time, will be reviewing their procedures to look for opportunities for improvement.  Experts suggest that enlisting hospital security to fill in, rather than leaving just one guard in place, may help.  Because the secure healthcare facilities in federal jails allow so many fewer escapes, using these instead of public hospitals may reduce the risk of escape.  However, there’s still the problem of transporting inmates, which is another high escape potential.

To view the Cause Map of the prisoner’s escape, click on “Download PDF” above.  Or click here to learn more.

Root Cause

Disabled resident dies when caregiver falls asleep

By ThinkReliability Staff

A physically disabled resident in a New York state-run care home required checks every two hours to ensure he was receiving adequate oxygen.  On the night of September 10, 2013, his nurse fell asleep, and he went more than 8 hours without the checks.  During this time, his oxygen level dropped to 40% (anything below 90% is considered dangerous), and he later died of hypoxic brain injury.

Says Patricia Gunning, prosecutor for the New York State (NYS) Justice Center for the Protection of People with Special Needs, “This case serves as a tragic reminder of the serious risk posed by an all too common workforce problem of caregiver fatigue or workers sleeping on shifts.”

Sadly, “all too common” turned out to be all too true.  The NYS Justice Center for the Protection of People with Special Needs was formed in mid-2013, and oversees agencies responsible for more than a million people in state care or state-funded nonprofits.  During its first year, it found 458 reports alleging abuse or neglect that cited a caregiver sleeping on the job.  This included caregivers who slept through a resident’s grand-mal seizure and a resident’s elopement, residents with unattended access to medications and food, and residents who were in a car driven by a caregiver who fell asleep at the wheel.

Even with a seemingly overwhelming problem such as this, progress can be made by looking at the specifics of one case, identifying causes that led to the problem, and developing solutions.  These solutions can then be considered for individual or widespread application.  We will examine the specifics of this case in a Cause Map, or visual root cause analysis, which lays out the cause-and-effect relationships leading to a problem.

The problem being examined is determined by the impact to an organization’s goals.  In this case, the resident safety goal was impacted because of the death of the resident.  The resident services goal was impacted because the resident did not receive adequate oxygen.  The compliance goal is impacted because of the felony charges against the nurse, who was sentenced to 90 days in prison.

Beginning with the most prominent impacted goal – in this case the resident safety goal – and asking “why” questions develop the cause-and-effect relationships that led to that impact.  In this case, the resident died from hypoxic brain injury (per diagnosis), from a lack of oxygen.  Due to the resident’s physical disability, his oxygen delivery equipment was required to be checked every 2 hours around the clock.  On the night of September 10 to September 11, more than 8 hours passed between checks, at which point the patient was found unresponsive.  (He died two weeks later.)

The resident’s oxygen delivery was not checked for more than 8 hours (as opposed to the required two) because the caregiver on duty had fallen asleep.  Testimony from the nurse in question as well as others from the facility describing sleeping on overnight shifts as a common occurrence.  Later research from the NYS Justice Center for the Protection of People with Special Needs found that many incidents involving caregiver sleeping on duty involved staff working extended or otherwise non-traditional work shifts.  The nurse who fell asleep on duty worked 12-hour night shifts at a site where many signed up for overtime and just barely passed duty hour requirements.

In response to the numerous caregiver sleeping events it discovered, the NYS Justice Center for the Protection of People with Special Needs has provided a toolkit aimed to protect people with special needs from caregiver fatigue.  The Center recommends that care provider agencies implement & regularly review policies meant to deter and detect sleeping on the job, establish contingency plans to relieve staff found unfit for duty, and provide assistance to residents in calling for help if caregiver is unresponsive.  Due to the myriad issues associated with caregiver fatigue, the American Nurses Association (ANA) continues to fight to reduce nurse fatigue, and possible harm to patients.

To see a one-page PDF with an overview of the investigation related to the resident lack of oxygen due to caregiver sleeping, click on “Download PDF” above.  Or, click here to learn more.