Home Pediatrics Pediatric literature plays catch-up on medical errors

Pediatric literature plays catch-up on medical errors

April 2009

Published in the April 2009 issue of Today’s Hospitalist

The Institute of Medicine’s 1999 “To Err Is Human” report blew the lid off health care’s dirty secret: We make a lot of mistakes. The report, which classified medical errors as the eighth leading cause of death in the U.S., jumpstarted a national discussion on how to prevent errors and adverse events.

Much of that discussion centered on errors related to adult hospitalizations, but last year, the pediatric literature started catching up. Several articles published in 2008 underscored the issues we face as pediatric hospitalists and suggested detection and prevention approaches we should consider. Here’s a review of several of those studies, with my comments.

Detecting errors
A study in the April 2008 issue of Pediatrics described the development of a pediatric-focused tool for detecting adverse drug events in 12 pediatric hospitals. The tool consisted of establishing “triggers” “which were events or scenarios, such as the use of diphenhydramine or naloxone, a rash, a serum glucose of more than 150 mg/dL “that would initiate a chart review to verify if an adverse event had occurred.

  • Results: The study sought to find a better way to uncover and report medication’related errors than through the typical avenues: occurrence reports, nontriggered chart reviews and administrative data analysis. By establishing triggers that would flag a case for chart review, investigators detected 1.23 adverse drug events per 1,000 medication doses. That, the authors noted, was “substantially higher than previously described.” About 22% of those errors were classified as preventable.
  • Bottom line: Hospital errors are more common than we thought. More importantly, we need to standardize what constitutes an “adverse event.”

    One effort to do exactly that came last year from the Centers for Medicare and Medicaid Services (CMS), which established guidelines for withholding payments to hospitals for care associated with a certain subset of errors: preventable complications including bed sores and blood incompatibility. Most of these measures are adult-centered. But they at least provide a template for establishing pediatric-specific adverse events that we should monitor and report, at least internally.

    As of now, pediatric hospitals are tracking measures related to these preventable complications. At Kentucky Children’s Hospital in Lexington where I practice, we’re tracking our rate of central-line infections as a quality measure. Other pediatric hospitals are also collecting data on urinary-catheter associated complications.

    The costs of adverse events
    While the above research tried to determine the incidence of errors, a study in the June 2008 issue of Pediatrics sought to quantify the excess charges and lengths of stay attributable to inpatient pediatric adverse events.

    The authors reviewed more than 400,000 discharges in 2006 from 38 freestanding pediatric hospitals. They then used pediatric-specific quality indicators, such as postoperative sepsis, iatrogenic pneumothorax and postoperative respiratory failure, to gauge the effect these events have on cost and length of stay.

    The bad news? Excess lengths of stay due to adverse events ranged from 2.8 to 23.5 days. Excess costs started at more than $34,000 (for accidental puncture or laceration) and soared to more than $337,000 for in-hospital mortality.

  • Results: This is one of the first studies to delineate pediatric-specific adverse events and establish their economic impact. The authors estimate, for example, that eliminating postoperative sepsis and pediatric health care-acquired infections could save about $700 million dollars and about 84,000 inpatient days a year. Although not all the adverse-event indicators are preventable, tackling them could significantly affect a medical system’s bottom line.
  • Bottom line: This type of study points out the opportunities that we have as hospitalists to take on error reduction and quality improvement “and use the financial data we should be collecting to affect hospital administrators’ decision-making.

    The time has probably come when administrators will look at subsidized hospitalist services and ask, “What have you done for me lately?” Quality improvement projects that show cost and safety benefits will continue to be the coin of the realm for hospitalist groups.

    Pediatric hospitalists are moving beyond basic quality measures, like length of stay, and are starting to track other quality variables, such as readmission rates and compliance with asthma medicines.

    In our hospital, for example, we recently instituted a clinical pathway to reduce the number of nebulized bronchodilators that patients with bronchiolitis receive. (We also enable respiratory therapists to make treatment decisions based on patients’ respiratory status.) This initiative has decreased length of stay and readmission rates.

    The next step: disclosure
    In a study published in the October 2008 issue of Archives of Pediatrics & Adolescent Medicine, researchers set out to find out how pediatricians feel about disclosing medical errors. \The authors looked at responses to an anonymous Web- and mail-based survey of attending physicians and pediatric residents at Washington University/BJC Health Care in St. Louis.

    The survey asked a set of questions about two different scenarios. In one, a child had to be admitted to the ICU after an insulin overdose. In the other, the failure to follow up on a lab test for antibiotic susceptibility resulted in symptomatic bacteremia and hospitalization. Respondents were asked to give their opinion on the severity of the error, how responsible the physician was for that error and whether the family should be told.

  • Results: Most respondents rated both errors as serious and said they would feel responsible for “and upset about “either one. Surprisingly, only 53% claimed they would definitely disclose the error to the patient’s family. They were more likely to report the insulin error than the bacteremia error.

    Even more troubling, only 24% of residents said they would provide a detailed description of why the error occurred, compared to 66% of attending physicians.

  • Bottom line: The communication gap between patients and physicians is greatly exacerbated when dealing with medical errors. But isn’t it better that patients find out about errors from their own physician, not from their lawyer? Studies have shown that engaging directly with the family, openly discussing mistakes and recognizing failure in the process of providing care decreases the risk of litigation.

    We need to do away with the paternalistic attitude that patients don’t know any better and realize that they are our allies. Not surprisingly, my current medical center has multiple policies that create venues for reporting adverse events, but none addresses disclosing errors directly to patients. The University of Kentucky Medical Center, however, is tackling that issue by holding educational activities that target medical error reporting.

    A significant step forward is the advent of family-centered rounds, which provides a much-needed platform for residents and medical students to understand the importance of communication between clinicians and patients.

    Medical schools are starting to broach this subject with courses on bioethics and medical humanities. But it is during residency that doctors-in-training will likely deal with errors and their disclosure. Academic physicians need to recognize the importance of talking about errors “both among themselves and with patients “and of serving as role models.

    The problem with drug samples
    Finally, researchers analyzed data from a nationally representative consumer survey taken in 2004 that included questions on how many children receive free drug samples.

  • Results: The study revealed that while 10% of children were given prescription medications, 4.9% of all children in 2004 received one or more free drug sample.

    Children who were uninsured for part or all of the year were no more likely to receive those samples than were those who were insured all year. Instead, the leading factor associated with receiving free samples was having routine access to health care, defined as three or more provider visits in a year.

    Two of the most common samples distributed to children were the schedule II medicines atomoxetine (Strattera) and amphetamine/dextroamphetamine (Adderall).

  • Bottom line: Samples are supposed to be given to those who cannot afford prescription medicine. However, survey results indicate that poor and uninsured children are not the main recipients of free samples.

    Responses do seem to indicate that there is a tiered health care system based on insurance, but that what we think of as safety net provisions aren’t really working. Instead, the survey seems to confirm what we pretty much know: that pharmaceutical companies use samples to advertise their products, not get essential medications to needy children.

    We certainly need to think long and hard about the drug reps in our offices and hospitals, and what their good-looking, lunch-providing, ubiquitous presence means for medicine. And giving out samples of Adderall is, in my book, questionable, not only from the standpoint of patient safety, but because of the potential for abuse.

    While most hospitals don’t dispense samples, patients show up with them all the time and as hospitalists, we frequently encounter patients on unnecessary meds. That underscores the fact that one of the most important aspects of our practice is good communication with referring physicians.

    We need to work with those physicians to coordinate discharge medications, which includes mentioning, in a professional way, which meds we think need to be discontinued. I try to assess each situation individually to see if a sample a patient is taking is appropriate for a specific condition.

    When there is a cheaper, equally effective therapy available, I tell patients to throw out the sample, and I give them the appropriate med. Just because a sample is free for the patient doesn’t mean it is free for the health care system.

    Ruben J. Nazario, MD, is a pediatric hospitalist practicing at Kentucky Children’s Hospital in Lexington and a blogger for Today’s Hospitalist.