Checklists help reduce hospital deaths

by Michael Smith

The use of treatment checklists for 13 common diagnoses was associated with a dramatic reduction in patient deaths at three London hospitals, researchers said.

The year the checklists were introduced, the three facilities in the North West London Hospitals NHS Trust registered 255 fewer deaths than the previous year, according to Brian Jarman, PhD, of Imperial College London, and colleagues.

The targeted diagnoses accounted for 174 fewer deaths than the previous year, Jarman and colleagues said online in BMJ.

The hospital standardized mortality rate — which compares actual deaths in an institution with the average national experience — went from among the highest in English acute care settings to the lowest, they said.

The impact of the checklists became significant a month after the intervention started, Jarman and colleagues said in the journal.

The clinical implication, they said, is that so-called “care bundles” can reduce mortality in the targeted diagnoses and also overall.

The three hospitals introduced the checklists in April 2007, targeting eight clinical areas: central venous catheter/line sepsis, diarrhea and vomiting, stroke, ventilator acquired pneumonia, methicillin resistant Staphylococcus aureus infections, heart failure, surgical site infections, and chronic obstructive pulmonary disease.

The 13 diagnoses covered by the eight checklists were peritonitis and intestinal abscess, senility and organic mental disorders, pleurisy pneumothorax pulmonary collapse, aspiration pneumonitis food/vomitus, skin and subcutaneous tissue infections, acute bronchitis, urinary tract infections, acute cerebrovascular disease, other gastrointestinal disorders, septicemia (except in labor), pneumonia, chronic obstructive pulmonary disease and bronchiectasis, and congestive heart failure (non-hypertensive).

Aside from the treatment checklist itself, each care bundle had a medical notes component, a flag to mark the patients involved, and a tracer backing form, the researchers said.

The bundle was printed on a single sheet of paper, with a peel-off sticker that was stuck in the patient’s medical notes on the day treatment was started, they said.

In total, over the year of the study, about 1,200 care bundles were used, they said, most of them at the busiest of the three sites, the Northwick Park Hospital.

The intervention resulted in a dramatic change, Jarman and colleagues said: the trust’s hospital standardized mortality rate fell from 89.6 in 2006-2007 to 71.1 in 2007-2008. (A figure of 100 is the national reference value; hospitals with higher or lower adjusted mortality have values above or below 100.)

From 2006-2007 to 2007-2008, the trust saw a 5.7% increase in admissions and a 7.9% increase in expected deaths. However, Jarman and colleagues said, there was a 14.5% decrease in actual deaths.

The overall decline in deaths — from the 13 targeted diagnoses and 43 others — was statistically significant, as was the reduction in deaths associated with the 13 targeted diagnoses, the researchers said.

However, while there was a decline in deaths from the 43 non-targeted diagnoses, it did not reach significance, they said.

Jarman and colleagues also noted that the effect was only significant at the Northwick Park Hospital, which had 62% of admissions and also the greatest use of the checklists.

The researchers cautioned that it is difficult to establish a cause-and-effect link between the reduction in deaths and the introduction of the checklists. They also noted that “there could have been a Hawthorne effect related to the participants in the project knowing that they were being monitored.”

However, they said, the mortality reduction occurred:

* Only where the care bundles were predominantly used.
* In the year in which they were introduced, and started in the month of introduction.
* For the targeted diagnoses and not for the not-targeted diagnoses.

Taken together, those facts indicate that “the introduction of the care bundles is among possible causes of the reduction in death rates,” they concluded.

Michael Smith is a MedPage Today North American Correspondent.

Originally published in MedPage Today. Visit for more hospital news.

Comments are moderated before they are published. Please read the comment policy.

  • Lockup Doc

    I’ll have to admit that I’m always skeptical when I hear about things such as checklists in medicine–the fear being that we’re moving toward “cookbook” medicine.

    But I recently read a book, “The Checklist Manifesto,” which is written by a surgeon, Atul Gawande. He takes the reader through different industries (airlines and others) in which checklists are used. He then talks about attempts to study the use of checklists in medicine, including how to and how not to create good checklists. After reading this book, I am much more open-minded about the idea, but ultimately the ones who are going to be using such checklists MUST be involved in the process of creating and implementing their use. They won’t work if they are created and handed down by those who don’t know what they don’t know.

    Here’s the book:

    • Ed Waldron

      It’s not cookbook if you are the cook. Physicians need to be leaders and set quality process in place for documentation. Plan, Do, Check, Act, Document and repeat. Whining abt cookbook medicine is an excuse not to duplicate process and measure results. Technology is not an enemy. Physicians have the tools to get quality in place and regardless of age or interest these tool such as EMR/EHR must be adopted but it will require physician leaders to make this happen.Physicians must not give quality process over to executive admins who will write the cookbook if physicians opt out of the quality battle.

      • Lockup Doc

        Well said.

        It was a healthcare quality improvement expert who pointed me toward the book I mentioned. And after having read that book, I totally agree with everything you just said. If we as physicians don’t take charge of this issue, it will likely be handed down to us in a form that we don’t like and that doesn’t work very well.

Most Popular