Monthly Archives: October 2014

Root Cause

The Race to Develop an Ebola Vaccine

By Kim Smiley

Traditional public health methods have not been able to stop the Ebola epidemic raging in West Africa and some experts are speculating that a vaccine may be necessary to quash the outbreak.  The only problem is an approved vaccine against Ebola doesn’t exist.

A Cause Map, a visual root cause analysis, can be built to analyze this issue by intuitively laying out the causes that contribute to the problem.  A Cause Map is built by asking “why” questions and documenting the answers on the Cause Map to show the cause-and-effect relationships.

So why isn’t there an approved Ebola vaccine?  There are several promising vaccines in development, but some of them are newer efforts that haven’t had time to go through the lengthy approval process.  A few potential vaccines have been around for years, but development stalled prior to the necessary human trials.  Prior to this year, there was limited potential revenue from an Ebola vaccine because of the limited demand so it has never been a high-priority product. Demand for, and interest in, producing an Ebola vaccine, has of course skyrocketed as a result of the ongoing epidemic in West Africa and Ebola cases popping up in other countries.

Now that companies are putting significant resources in the race to produce an Ebola vaccine there are still huge logistical obstacles that must be overcome.  At least two different vaccines should be ready for large-scale human trials early next year, but actually distributing the vaccine and tracking volunteers will require significant resources.  The vaccine must be kept at a constant temperature which can be difficult in regions of West Africa without reliability electricity.  Keeping track of thousands of volunteers is always a massive undertaking, but will be even more challenging in the middle of an epidemic in a region where the medical systems are overtaxed.  There is also a chance for significant political fallout if the vaccine created by Western countries and given to poor African nations turns out to have harmful side effects.

This topic raises some really difficult ethical issues.  How much do you fast-track a vaccine?  People are dying and an effective vaccine would save lives, but distributing a vaccine prior to the normally required testing could also result in significant human suffering if there is an unexpected side effect.  When has a vaccine been tested “enough” to justify giving it to people in a high-risk environment?  Even designing the human trials requires some hard decisions.  Do you conduct a blind study with a control group, knowing that some of that group is statistically likely to be infected with a deadly disease? There is a lot of gray area and it’s difficult to know what the right answer is.  Thousands of lives may hang in the balance and there isn’t a lot precedence in how best to respond to the challenge of this Ebola outbreak.

If you’d like to learn more about this epidemic, you can read our previous blogs:

Patient With Ebola Sent Home From ER

Ebola Outbreak Claiming Lives of Medical Staff

Hundreds Affected in ‘Unprecedented’ Ebola Outbreak

 

Root Cause

Humidity May Cause Loss of Hospital Accreditation

By ThinkReliability Staff

A hospital in California has received a preliminary denial of accreditation from The Joint Commission, which may lead to a loss of Medicare and Medicaid funding.  This move is rare – less than 1% of hospitals are denied accreditation.  The ruling was made during an onsite survey after reports that four patients had developed surgical-site infections after hip replacement surgery within a month.  (The patients all recovered after administration of antibiotics.)

Says the hospital’s Associate Director of Emergency Services, “Surgical infections are extremely rare. When we saw the cluster we stood up and took notice. We stabbed every corner and every crevice.”

However, that may not be entirely true.  First of all, surgical infections are probably not what’s considered “extremely rare”.  In a recent study, the Centers for Disease Control and Prevention (CDC) found that 1 in 25 patients acquires an infection while in the hospital  and infections are among the leading cause of death worldwide.  (See our previous blog for more details.)  Only slightly more than 20% of those infections were found to be surgical site infections from inpatient surgery, but that still results in about 157,500 surgical site infections a year.

There’s also a question about how quickly action was taken to address the issues identified by The Joint Commission, specifically high humidity.  After the four patients were infected with enterobacter (a bacteria that is harmless in the gut but can cause illness in a wound), the hospital cultured equipment and instruments and found no contamination.  However, three of the surgeries which resulted in the surgical site infections took place in an operating room that had been identified in February or March to have high humidity due to issues with the heating and ventilation system.  Although a new system would not be installed for several months, surgeries continued to take place in that operating room until The Joint Commission identified problems with heat and humidity.

The germs that caused these infections can also be introduced to a surgical site by medical staff – usually due to improper hand washing or surgical prep procedures.  It’s unclear if this may have been the case here – there has been no discussion of this possibility – but handwashing issues are a constant source of infections at healthcare facilities.  According to studies, more than 50% of infections are largely preventable with good hygiene and technique.

The hospital has 23 days from the October 8th preliminary denial of accreditation to meet federal standards.  The hospital is installing new heating and air conditioning equipment and is reviewing its procedures for “hospital documentation and staff communication”.   Hopefully other processes will be reviewed as well.  The hospital expects to reopen its operating rooms soon.

To view an overview of the investigation of this issue, please click “Download PDF” above. The downloadable PDF shows the goals impacted by this issue, a visual layout of the cause-and-effect relationships that led to the impacted goals, and the solutions being discussed by the hospital to improve patient safety regain their accreditation.

Root Cause

FDA To Device Maker: Complaints Not Properly Investigated

By ThinkReliability Staff

The Food and Drug Administration (FDA) issued a warning letter – a possible precursor to regulatory action – to a manufacturer of pulse oximeters challenging its response to violations identified during a facility inspection in 2013.  Specifically it found that:

1) Complaints of device failures were not adequately investigated,

2) Investigation records did not include required information,

3) Procedures did not identify all potentially affected product, and

4) Causes of failures were not investigated.

The company’s response to the initial violations found during the FDA inspection was found to be unacceptable.  Some of the comments to the FDA regarding the violations were: “The reported event poses no risk to user or patient safety, therefore, no Corrective and Preventive Action required”, “The cable and sensor that were returned with the instrument were also investigated and no problem was found”, “Through the investigation process, it was determined that Masimo’s product did not malfunction”. To each of the violations, the company claimed that either no problem was found, or that the problem did not involve any failures of the pulse oximeters identified in the complaints.

This doesn’t ring true to Diana Zuckerman, the president of the National Center for Health Research.  She says, “It may well be that it’s a user error. But you have to investigate that and show that it’s a user error and not a device error… When a company refuses to respond in any way to the FDA other than to say that the FDA is wrong on every issue, that’s not very credible.  Especially when users made complaints that the company’s product put patients at risk.”

Included in the complaints are that the manufacturer’s pulse oximeters failed to alarm in a patient that died and that they resulted in burns on both an 11-month old and 33-year-old and a skin tear on another patient.  Complaints also indicated issues with consistency of readings and exposed wires at the connection points.  The failures are getting attention not only because FDA warning letters are fairly rare – this is the first such letter to the company in its 25-year history – but also because of the company head’s views on medical errors and funding of the nonprofit Patient Safety Movement Foundation.

Pulse oximeters are used to monitor a patient’s pulse and blood oxygen.  Abnormal readings can indicate a change in a patient’s condition which may require medical intervention.  If the device fails to function, it can lead to patient injury or even death.  Because the devices are intended for use in diagnosis, they are regulated by the FDA and have to conform with FDA practices.  If the violations identified by the FDA in the warning letter are not corrected, the company may face regulatory action.

For now, the company has not released its plans to ensure compliance with the FDA requirements.  The FDA is clearly looking at updates to the investigation process used to respond to customer complaints and ensuring that causes identified as part of those reviews result in changes to other processes, such as manufacturing and quality control.

To view an outline and analysis (in the form of a Cause Map, a visual form of root cause analysis) of the FDA’s findings, click on “Download PDF” above.

Root Cause

Patient With Ebola Sent Home From ER

By Kim Smiley

If you have been paying any attention to the news lately you have heard that a patient was diagnosed with Ebola in the United States for the first time.  The fact that the patient sought treatment at an emergency room (ER) and was sent home is particularly alarming and people are naturally very interested in how such a thing could happen.

The media has been flooded with breaking news about this case. In situations like this, keeping track of what information is current and reliable and what is already outdated or has been determined to be inaccurate can be a moving target.  A Cause Map, a visual format for performing a root cause analysis, can be useful in these situations as a way to document the available information.  A Cause Map visually lays out the causes that contributed to an issue in an intuitive format.  It is relatively easy to quickly expand a Cause Map as more information becomes available.

Generally, a Cause Map is built in an Excel workbook so old versions of the Cause Map can be easily saved as different tabs and that the evolution of the investigation isn’t lost.  Additionally, it’s easy to add evidence supporting each individual cause onto the Cause Map itself so that all relevant information is documented in one location and easily referenced.  There are often things that are being considered that may have played a role in a problem, but lack evidence to determine whether or not they are actually relevant to the investigation.  This situation can be documented on a Cause Map by listing the potential cause and adding a question mark to show that a cause is being considered, but that it needs more evidence.  If that cause is later determined to not have been a factor it can be crossed out to document that it has been considered so that no work is duplicated.

So what information is known at this point about the patient with Ebola who was sent home from the ER?  A man has been diagnosed with Ebola in Dallas, Texas.  Statements by friends and family indicate that he helped transport a sick woman prior to traveling to the US.  It’s not clear whether he knew that she had Ebola or not.  After being exposed to Ebola, he passed through the airport screening because he did not have a fever at that time and he did not indicate possible exposure on the required questionnaire.   He arrived in Dallas, Texas on September 20, 2014.

His family has stated that he started feeling ill on the 24th and he sought treatment at a hospital on September 26.  He was released with a prescription for antibiotics.  Ebola can be difficult to diagnose because the early symptoms, such as fever and muscle pains, are very nonspecific.  The only information the hospital would have had to indicate that patient might have Ebola is his travel history.  Statements by the hospital indicate that the patient told the nurse he had recently been in Liberia.  The travel history information doesn’t seem to have been known by physicians that treated the patient.  Initial statements by the hospital indicated that issues with electronic health records may have played a role in the confusion, but later released more information that modified their position.  Exactly how the risk of Ebola was missed isn’t clear and this portion of the Cause Map will need to be expanded as more information is available.

It is not shocking that an Ebola case would be diagnosed in the US with travelers still arriving from West Africa and the long incubation period.  As long as Ebola is still a problem in Africa, US hospitals could see more patients with Ebola and need to be prepared for the possibility.  But it is concerning that someone who had recently traveled from West Africa with a fever wouldn’t trigger any alarms at an ER.   Individuals with Ebola must be quickly identified and isolated In order to prevent the spread of the deadly disease and hospitals in the US need to ensure that no other potential Ebola patients fall through the cracks.

Click on “Download PDF” above to see an initial intermediate level Cause Map for this example.

Root Cause

Healthcare Worker with Active TB May Have Infected Infants

By ThinkReliability Staff

A healthcare worker may have exposed more than 751 infants and 55 coworkers to tuberculosis (TB), a potentially deadly disease, during the 11.5 months she worked with newborns in the nursery and post-partum unit.  The employee was found to have symptoms of active TB (the only kind that is contagious) in July 2014 and was tested on August 21, 2014.  Infants and coworkers who were in those units between September 1, 2013 and August 16, 2014 are potentially affected.

The impacts of the potential exposure are significant.  The goal of ensuring patient safety has been impacted because of the potential exposure of 751 infants (so far identified).  Additionally, the goal of ensuring employee safety has also been impacted as 55 coworkers are also potentially impacted (those already screened have not shown any signs of infectious tuberculosis).  Because of the close contact required to spread TB, the issue has not been identified as a public health threat.  Regulatory agencies are still interested – the hospital has been cited by the Department of State Health Services for issues relating at least in part to infection control that are “an immediate jeopardy to patient safety”.  (It wasn’t clear how or if these issues are directly related to the TB exposure but were found in an inspection that occurred as a result of it.) All the patients who have been exposed may potentially require a course of antibiotics, typically six to nine months.  Infants under six months are being given the course as a preventive measure.  This impacts the patient services goal.  All the screening and treatment is being provided by the hospital free of charge, which is an impact to the hospital’s labor and time goal.

Developing the cause-and-effect relationships that led to the goals that were impacted can provide clarity to the investigation and potential solutions.  Exposure to TB occurred when an employee with an active infection came to work and was potentially aided by the hospital’s infection control policies or procedures, which may not have been effective in preventing the spread of the disease.  It’s unclear how the employee contracted tuberculosis, but she was likely not vaccinated. Although a vaccine against tuberculosis exists, it’s rarely used in US.

A question raised by this issue is why the long period of time during which there was a potential exposure?  Tuberculosis can remain latent in the body for months or even years before turning into an active case.  (It is only contagious when active.)  The employee appears to have passed a routine annual health screening in July 2013 and started showing symptoms at or near her next annual health screening in July 2014.  However, she was not tested for the disease until August 21st and appears to have continued work until August 16th.  It’s unclear why the delay occurred, and the hospital will surely be looking to ways to minimize patient exposure to workers who may be sick.

The hospital is working with the Texas Department of State Health Services and the Centers for Disease Control and Prevention to screen and treat patients and other healthcare workers that came into contact with the infected worker based on employment and medical records in an extensive outreach campaign.  The employee has been placed on leave and is being treated.  The hospital is required to submit a corrective action plan to the Texas Department of State Health Services, which should identify corrective actions to issues raised as a result of the state’s inspection and by the Centers for Medicare & Medicaid Services.

To view the impacted goals, cause-and-effect relationships and potential solutions in a Cause Map, or visual root cause analysis, please click “Download PDF” above.  Or, click here to learn more.