Patient Safety Blog

On April 1, 2012, a patient at a university medical center in New York died from sepsis. The death was especially heartbreaking as the patient was 12 years old . . . and had been healthy just 4 days prior. However, he had contracted a bacterial bloodstream infection (sepsis), which has a high mortality rate (nearly 40%, according to the United Hospital Fund) that grows with every passing hour. (A study cited by the New York Times found that the survival rate decreases by 7.6% every hour before antibiotics are given.) With response time so crucial to patient outcome, rapid action at every step of the process is required.

We can look at this incident in a visual root cause analysis, or Cause Map. The purpose of this map is not to assign blame, but rather to discover and document causes in the hope of finding solutions to reduce the occurrence of this type of issue.

We begin with the impacts to the goals. In this case, the patient safety goal was impacted due to a patient death. Because of the high potential for emotional impact to providers, employees are also impacted. The potential for a lawsuit is an impact to the organizational goal, and the initial misdiagnosis of the patient is an impact to the patient services goal.

We begin with the patient safety goal and ask “Why” questions to develop cause-and-effect relationships that will show all the causes of the incident. The patient died of severe septic shock and insufficient intervention. (Had intervention come earlier, the patient may have lived.) The onset of the sepsis appears to have been a cut acquired at school, which was bandaged, but possibly not cleaned, likely due to the lack of severity of the initial injury. Delay of treatment allowed the sepsis to overwhelm the immune system. The treatment was delayed due to an initial misdiagnosis of dehydration. Sepsis is particularly difficult to diagnose because many of its symptoms mirror symptoms of other more common ailments. Information was not shared between providers – the child’s primary care pediatrician, parents, and the hospital staff, which may have contributed to the difficulty in diagnosis. Test results taken at the hospital came in after discharge and were not shared by phone with the primary provider or parents. Additionally, even after lab results from the hospital suggested that the white blood cell count was abnormally high, indicating infection, no action was taken.

From this very basic, high level map, at least four areas of specific improvement can be noted. Protocol at the school for injuries that involve cuts – even if they seem minor – should include cleaning or disinfection. The hospital should have – and follow – protocol for that specifies action to be taken upon receipt of lab results. This protocol should include documenting and sharing test results with other providers and caregivers. Because of the difficulty in diagnosing sepsis, and the importance of quick action, the United Hospital Fund is current developing a STOP Sepsis Collaborative, which aims to “reduce mortality in patients with severe sepsis and septic shock by implementing a protocol-based approach to case identification and rapid treatment”. Ideally, implementation of the results of this collaborative will reduce the risk of patient death from a situation like this tragic case.

To view the Outline, event Timeline, Cause Map, and Solutions, please click “Download PDF” above. Or click here to read more.

Accidentally Ingested Wire Bristles

July 12, 2012 Kim Smiley

There have been a number of documented cases of people accidentally swallowing wire bristles from the brushes commonly used to clean grills. If ingested, the bristles can cause significant pain and have the potential to puncture organs, including the intestine which can lead to dangerous infections. There isn’t a lot of data available to determine how often this occurs, but a recent report discussing six cases of ingested wire bristles at a single hospital in a one year period hint that this may be more common than many realize.

This issue can be analyzed by building a Cause Map or visual root cause analysis. The first step in creating a Cause Map is to determine how the issue impacts the overall goals of an organization. In this example, we’ll consider the organization the general public and the main impacted goal is the safety goal since there is a risk of serious health issues. Causes are now added to the Cause Map by asking “why” questions. Why is there a risk of serious health issues? Because there is the potential to swallow a wire bristle and the wire bristle can do a lot of damage within the body. (To view a high level Cause Map of this issue, click “Download PDF” above.)

Continuing the “why” questions, we would now ask “why” is there a potential to swallow a wire bristle? This possibilities exists because people may inadvertently swallow the bristles without realizing it, the bristles come from wire brushes that are often used to clean residential grills and the bristles sometimes fall out of the brushes and stick to the grill. People may inadvertently ingest the bristles because the bristles can stick to meat and the texture can hide the presence of the bristle. Many people are also unaware of the potential danger of eating a wire bristle so they aren’t looking to find bristles. Investigation into this issue has not found any defects that are causing bristles to fall out. No one brand or type of grill brushes has been singled out as the culprit in these causes, but a worn grill brush is more likely to shed bristles than a new brush.

All six patients in the report did make full recoveries after treatment. In three of the cases, the wire bristles were stuck in the throat and required only laryngoscopic removal; the other three cases required more invasive surgery to remove the object.

The best way to protect yourself from the possibility of this occurring is to inspect the grill after using a wire grill brush or to find another method to clean your grill. The Consumer Product Safety Commission has asked that medical professions or consumers report any incidents of swallow grill brush bristles to http://www.saferproducts.gov to help monitor this issue.

Medical Laboratory Errors

June 28, 2012 ThinkReliability Staff

Surprisingly, many of what are considered laboratory errors do not actually occur in the lab. But errors related to laboratory testing can negatively impact patient care. We can look at the impacts and causes of errors related to diagnostic testing in a Cause Map, which allows us to visually diagram cause-and-effect relationships.

We begin this type of root cause analysis by determining the impacts to the organization’s goals. In this case, because we want to consider all possible sources of diagnostic errors in a proactive analysis, we will look at the generic goals for an organization that provides healthcare. Diagnostic errors can cause an impact to the patient safety goal because of the risk of impact to patient treatment. Employees’ abilities to do their job is impacted because they may be receiving incorrect information from lab testing. There is a risk of impact to the patient’s treatment, which is an impact to the patient services goal. Additionally, there is a risk of performing unnecessary treatment as a result of incorrect testing results, which could impact both the property and labor goals.

Once we have determined the impacts to the organization’s goals (and there may be more impacts for specific incidents involving diagnostic testing errors), we can ask “Why” questions to determine the causes that result in these impacts. We will begin with the patient safety goal impact. The patient safety goal is impacted because of the risk of an impact to a patient’s treatment. This includes the possibilities of a risk of delayed treatment, risk of not receiving needed treatment, and a risk of unnecessary treatment. Delayed treatment can occur from a delayed diagnosis, which could result from either delayed or incorrect testing results.

Delay of testing results can be caused by delayed reporting of results, potentially due to a lack of time requirement for reporting results and/or a lack of tracking these results. A possible solution to delayed reporting of results can be to implement a standardized process for reporting results, which may include time limits or guidelines for reporting results.

Incorrect treatment – whether that is not getting needed treatment or receiving unneeded treatment – can result from an incorrect diagnosis. An incorrect diagnosis can result from an incorrect assessment of diagnostic testing. An incorrect assessment can result from either an incorrect interpretation of laboratory test data or incorrect data from the lab testing.

Incorrect interpretation of lab testing can result from reports that are difficult to interpret, either due to a confusing layout or illegibility. A solution to this is to have a standardized reporting form. Other potential causes of incorrect interpretation include confusion of verbal reporting (such as over the phone) or results not being interpreted by a specialist. Solutions that can reduce this confusion include providing reports electronically when available or repeating results when provided verbally, and making lab experts available for interpretation.

Three main reasons that incorrect data is provided as a result of lab testing is that the specimen is associated with the wrong person, possibly because a patient is misidentified, a specimen is mislabeled, or information is entered incorrectly into the computer. Possible solutions are to use two patient identifiers and label the specimen in the presence of the patient.

Contaminated specimens can also cause incorrect testing results. Specimens can be contaminated at collection, handling, or testing. Any of these issues can be caused by insufficient quality control. The risk of contamination can be minimized by a standardized quality control procedure.

Lastly, incorrect diagnostic data can result from the wrong test being performed. This could occur due to equipment failure, an incorrect entry into the computer, or the wrong test being ordered. More details about any specific incident can be added to the Cause Map based on evidence gathered in the course of an investigation.

To view the Outline and Cause Map, please click “Download PDF” above. Or click here to read more.

Never Event, Root Cause Analysis

Use of Contraindicated Clip Leads to Death of Kidney Donor

June 22, 2012 ThinkReliability Staff

In 2011, a kidney donor in Texas bled to death after her renal artery became open. Sadly, her death was associated with the use of clips to close the artery – rather than staples – even though the use of clips was contraindicated for this purpose. The instructions that came with the clips said this, as did several warning letters sent from the manufacturer in previous years.

We can look at this tragic issue in a Cause Map, or visual root cause analysis. We begin with the impacted goals. Because of the patient death, the patient safety goal is impacted. Emotional impacts from employees resulting from a patient death can be considered an effect to the employee impact goal. The use of a device other than intended is a result to the patient services goal and is considered a “never event” (an event which should never happen), resulting in an impact to the compliance goal. A lawsuit resulting from the patient death is an impact to the organization goal. A total of four kidney donors are known to have died as a result of using these clips.

We begin with the impacted goals and ask “Why” questions to understand the cause-and-effect relationships resulting in this tragedy. The patient died from a massive, sudden bleed caused by the bleeding of the renal artery which was open. The renal artery had been opened as part of the kidney donor surgery, and had been closed using clips that slid off the renal artery. The stump remaining on the renal artery after this kind of surgery is too short to allow the clips adequate purchase, and the clips slid off. The hospital staff was unaware that these clips were contraindicated for this use. Although a warning was placed on the instructions for the clips, these instructions were not kept in the operating room. Additionally, the manufacturer sent out several letters to hospitals warning them not to use these clips for kidney surgery. However, at that time, this hospital was not using the clips, and had forgotten about the letters when the clips were purchased.

Once the causes related to the issue have been captured, possible solutions can be brainstormed. In this case, there are solutions for all the stakeholders in the event. The operating team should use staples instead of these clips to close the renal artery. The FDA has issued a safety notification to attempt to provide additional warnings against using these clips after kidney donation. The hospital has implemented a system to track and document warnings and recalls related to medical equipment. Some personnel in the medical community have requested that the warning not to use the clips after kidney surgery are printed directly on the clips, rather than on the operating instructions. Dr. Amy Friedman, the Director of Transplant Services at Upstate Medical University in New York, who had raised concerns about using clips in kidney donors starting in 2004, would also like the warnings to include information that donors have died as a result of using these clips. Although the FDA believes that the warnings up to this point have been sufficient, hopefully the additional actions will prevent another death from the use of these clips.

To view the Outline, Cause Map, and Solutions, please click “Download PDF” above. Or click here to read more.

Abuse of “Good Grade” Pills

June 14, 2012 Kim Smiley

A recent New York Times article, “Risky Rise of the Good-Grade Pill” talked about the dangers associated with abuse of prescription stimulants. These stimulants are prescribed to treat attention deficit hyperactivity disorder (ADHD), but they cause individuals without the disorder to become hyper focused and better able to concentrate for long periods of time. There isn’t good data on how many high school students are abusing the stimulants, but anecdotal evidence implies that a significant number of students, especially those at highly competitive high schools, use the prescription stimulants to help improve test score and grades.

This issue can be analyzed by building a Cause Map, a visual root cause analysis. The first step when beginning a Cause Map is to determine the impact to the overall organizational goals. In this example, the Safety Goal is impacted because there are risks to the students abusing the prescription stimulants. Once the impact is defined, “why” questions are asked and causes are added to the Cause Map.

Why are there risks to students abusing prescription stimulants? Students are at risk because students without medical need are using the stimulants, the stimulants can be dangerous and illegal sale or procession of these drugs can have severe consequences.

Students are using the stimulants because they are available and they can aid in earning good grades. The stimulants are available because some people sell them and there is anecdotal evidence that some students fake the symptoms to get prescriptions for them.

Many of the students using these stimulants don’t realize that there can be serious health issues with using them. First off, these drugs are one of the most addictive substances that have a medical use. There is little known about the potential long term health issues of abusing prescription stimulants. There is also a concern that these drugs may act as a gateway drug which has the potential to open the door to more drug abuse as students get more comfortable with the idea of taking pills. In the short term, there are a number of health issues to consider including heart irregularities, exhaustion and even psychosis during withdrawal.

Another important fact many students are unaware of is that giving these drugs to a friend can have heavy legal consequences. Distribution of prescription stimulants is considered a felony. Adderall, Ritalin and other name brand stimulants used to treat ADHD are controlled as Class 2 substances, the same as cocaine and morphine, and the sale of them carries severe penalties.

This issue is just beginning to come to light and more information is needed to fully understand how many students are involved and the best way to stop the abuse of the prescription stimulants. As more information is available it can easily be added to the Cause Map.

To view a high level Cause Map of this issue, click on “Download PDF” above.

Study Finds that Diabetics are at Risk of Medication Errors

May 31, 2012 Kim Smiley

A new study found that nearly a third of diabetic hospital patients experienced a medication error in a one week period. The study examined bedside data for 12,800 patients and 6,600 patient questionnaires for hospitals in England and Wales. Medication errors when treating diabetics can have severe consequences because many diabetics require medication to maintain healthy blood sugar levels. Blood sugar levels that are either too high or too low can result in significant illness and even death if untreated.

The two most common errors found by the study were failing to properly adjust medication when a patient’s blood sugar level was high (23.9%) and failing to sign off on the patient’s bedside information chart when insulin was given (11.1%).

This issue can be examined by building a Cause Map, an intuitive, visual root cause analysis format. The first step to building a Cause Map is to determine the impact to the overall organizational goals. In this example, the risk to diabetic patients is an impact to the safety goal. The next step is to ask “why” questions and add the cause boxes to the Cause Map to illustrate the cause and effect relationships between all the factors that contributed to the issue.

In this example, the risk to the diabetic patients occurred because medication errors occurred and the patients required medication to maintain healthy blood sugar levels. The study did not provide details on why the medication errors were made by hospital staff, but that information could be added to the Cause Map if it becomes known. A Cause Map can be still be useful when only a high level map can be built because it can help identify an at risk population and a common problem, the diabetic patients and the medication errors, which could help identify where more research is needed or where resources could be directed. To view a high level Cause Map of this issue, click on “Download PDF” above.

A potential solution that has been suggested for this problem is to improve training for hospital staff on how to properly treat diabetic patients. A more detailed look at understanding exactly why the staff is making errors could help direct the training plan to the most needed areas.

Amish have few allergies

May 17, 2012 Kim Smiley

A new study has found that Amish children living in Indiana have far fewer allergies than the general population and even significantly fewer allergies than other children living on farms. As high as 50 percent of the general population has evidence of allergic sensitivity when tested and only seven percent of the Amish children had allergic sensitivity. The study also looked at Swiss children living on farms and found that they had half the allergic sensitivity of the general population, but still more than Amish children.

Why is this finding significant? Scientists hope that studying the Amish will help them understand what factors are causing the large increase in allergies in the general population in Western Countries over the past few decades.

This issue can be built into a Cause Map, an intuitive, visual root cause analysis, to help illustrate the-cause-and-effect relationship between the factors involved. As more research is done and more information on this issue becomes available, it can easily be added to the Cause Map. In this example, researchers aren’t sure why the Amish have such low levels of allergic sensitivities, but there are a few factors that are likely involved. These factors could be documented on the Cause Map, but a question mark would be added to note that more information is needed to verify the accuracy of the cause and to ensure proper placement on the Cause Map. To view a high level Cause Map of this issue, click “Download PDF” above.

One fact worth adding to the Cause Map is that Amish are exposed to a wide variety of animals and the germs that go along with them from a young age. Many Amish live on farms and nearly all own horses for transportation. Additionally, Amish children help care for the animals from a young age. Pregnant Amish women are also typically around large animals and the prenatal exposure may play a role. Many Amish also consume unpasteurized milk and the impact of this on development of allergies is an ongoing debate.

In addition to environmental factors, some researchers also think that genetic plays a role in allergies so it is also worth noting that the Amish are relatively isolated genetically with limited mixing with the general population.

Understanding the factors that contribute to the low allergy rates of Amish children would hopefully help scientists both understand why allergies in the general population are increasing so dramatically and the best actions to take to treat them, maybe even before they develop.

Hospital Acquired Condition, Root Cause Analysis

New Research on the Impact of Hospital-Acquired Infections

April 20, 2012 ThinkReliability Staff

Recent research has shown that in-hospital mortality for patients who acquire an infection in the hospital increases from 4.5% to 18.5%. Hospital-acquired infections are infections obtained while a patient is hospitalized. The three main hospital-acquired infections (or HAIs) are bloodstream infections (28% of HAIs), pneumonia (21%) and urinary tract infections (15%).

Not only does an HAI increase the mortality rate, it has other impacts as well. We can look at these impacts, and their causes, in a root cause analysis demonstrated visually as a Cause Map. For the purpose of this root cause analysis, we will limit our investigation to HAIs that occur during hospitalization in an intensive care unit (ICU). We begin with determining the other impacts to the goals. The patient safety goal is impacted due to the increase in mortality. The organization goal is impacted because many insurers (including Medicare and Medicaid) will not reimburse for some infections obtained during hospitalization. Additional treatment is required to treat the infection, resulting in an impact to the patient services goal. The treatment for these infections normally results in an increased stay in the ICU (from an average of 8.1 days to 15.8 days), at a cost of $16,000. It is estimated that 26.7% of all ICU stays result in at least one HAI.

Beginning with the impacted patient safety goal, we can ask “Why” questions to demonstrate the cause-and-effect relationships leading to the increase in mortality. Increased mortality is due to the acquiring of an HAI. HAIs result from the exposure to a pathogen and frequently occur in the ICU partially due to the increased risk of infection due to the underlying condition for which the patient is in the ICU. There are two types of pathogens to which patients can be exposed: endogenous (essentially, from the patient’s own body) and exogenous (from visitors, healthcare providers, equipment, the environment, etc). HAIs are highly related to the use of invasive support measures, which provide a path for either kind of pathogen directly into the patient’s body. Specifically, the use of a central intravenous line is cited in 91% of bloodstream infections, mechanical ventilation is cited in 95% of hospital-acquired pneumonias, and urinary catheters are cited in 77% of urinary tract infection.

Because these invasive support measures are generally required for patient care, it’s difficult to see how these infections can be reduced. However, some programs have been shown to substantially reduce HAIs – and the cost associated with them – by improving the culture of safety and compliance with preventive methods. One such program in Michigan has reduced the rate of bloodstream infections associated with central lines from 7.7 to 1.3 per 1,000 catheter days. Even without a dedicated safety program, insisting on hand washing and proper cleanliness procedures during the insertion, checking, and removal of invasive support measures can reduce the risk of HAIs. Additionally, because the use of invasive support measures is so strongly correlated to HAIs, removal of these measures as soon as possible can also reduce the risk.

To view the Outline and Cause Map, please click “Download PDF” above. Click here to read more about hospital-acquired conditions. Or click here to read more about the latest research.

Never Event, Root Cause Analysis

Patient Deaths Caused by Defective Defibrillator Wires

April 5, 2012 Kim Smiley