Posted by Sten Westgard, MS
While we recently got a study that estimated the (frightenly high) number of Adverse Events caused by US hospitals, it looks like other countries are not content to let us stand alone. Now Sweden is doing us one better. Guess what the Adverse Event rate is in one hospital in Sweden?
- 28.2%
- 20.5%
- 6.3%
- 3.32%
Which number would you choose?
As it happens, each number is correct in its own way.
The study, Characterisations of adverse events detected in a university hospital: a 4-year study using the Global Trigger Tool method, Rutberg H, Borgstedt Risberg M, Sjodahl R et al, BMJ Open 2014;4:e004879
In this study, 20 records were randomly reviewed each month bewteen 2009 and 2012. University Hospital in Linkoping in southeast Sweden, has 650 beds and 32,000 admissions annually. The Global Trigger Tool is a review method that involve record review by clinicians to determine if an Adverse Event occurred during a past hospitalization.
- Of 960 medical records reviewed, they found 271 adverse events, or 28.2%.
- This meant that 20.5% of all discharges experienced at least one adverse event (some experience more than one)
- If the adverse event rate is weighted per 1000 patient-days, that works out to 3.32%
- As it happens, of all these adverse events that occurred, only 6.3% were actually reported. the hospital has a voluntary reporting tool and is mandated to report any adverse events, but the vast majority of events either went unrecognized or unreported.
- What we would probably want to do is use the 3.32% rate of adverse events per 1000 patient days, which works out to a 3.4 Sigma-metric (short term scale). What's interesting is that this matches almost exactly what was reported for US hospitals and their adverse event rate.
This study, particularly the under-reporting detail, is a chilling finding indeed, because it implies that many of our metrics might be completely underestimating the extent of errors occurring in healthcare.
The cost of these errors is not trivial:
"If we extrapolate our findings of an AE rate of about 20% and 5.4 additional hospital days for patients with an AE to the 32,000 hospital admissions/year, approximately 35,000 additional days are used in the hospital for treating patients with an AE. Based on our own and other findings, we estimated that about 50-70% of the AEs were preventable, which indicates that annually around 17,500 - 24,500 additional hospital days are used for caring for patients with a preventable AE. According to the Swedish Association of Local Authorities and Regions, the average cost for a hospital day in Swedish healthcare is SEK 8700 (US$1260). Thus, the estimated cost for the hospital could be between SEK512 and SEK213 millions (US$22-31million) annually."
That's just one hospital. Think of the implications for an entire healthcare system.
We've got a lot of work to do.
Characterisations of adverse events detected in a university hospital: a 4-year study using the Global Trigger Tool method, Rutberg H, Borgstedt Risberg M, Sjodahl R et al, BMJ Open 2014;4:e004879
Comments