Hey there, human — the robots need you! Vote for IEEE’s Robots Guide in the Webby Awards.

Close bar

IT Hiccups of the Week: At least 17.4 Million U.S. Medication Errors Avoided by Hospital Computerized Provider Order Entry Systems

Microsoft loses cloud, ISS comms go out, parking lot glitches and Facebook thinks everyone is under 100 years old

4 min read

IT Hiccups of the Week: At least 17.4 Million U.S. Medication Errors Avoided by Hospital Computerized Provider Order Entry Systems

This past week has seen a hodgepodge of IT-related uff das, glitches and snarls. However, we are going to start this week off with millions of human errors avoided by IT.

Computerized Provider Order Entry Systems Avoid an Estimated 17.4 Million Medication Errors Per Year

Last week, the Journal of the American Medical Informatics Association (JAMIA) published a study that estimated the reduction in medication errors in U.S. hospitals that could reasonably be attributed to their computerized provider order entry (CPOE) systems.  The study’s authors said that they “conducted a systematic literature review and applied random-effects meta-analytic techniques” to develop a “pooled estimate” of the effects of CPOEs on medication errors.

They then took this estimate and combined it “with data from the 2006 American Society of Health-System Pharmacists Annual Survey, the 2007 American Hospital Association Annual Survey, and the latter's 2008 Electronic Health Record Adoption Database supplement to estimate the percentage and absolute reduction in medication errors attributable to CPOE.”

Working through the data, the authors concluded that a CPOE system decreases the likelihood of error by about 48 percent . "Given this effect size," say the authors, "and the degree of CPOE adoption and use in hospitals in 2008, we estimate a 12.5% reduction in medication errors, or ∼17.4 million medication errors averted in the USA in 1 year.”

The study authors are careful to note that it is unclear whether this reduction in medication error actually “translates into reduced harm for patients,” although the research tends to lead one towards that conclusion.

The number of medication errors avoided because of CPOEs is expected to rise as more hospitals install them. Only about 20 percent of U.S. hospitals had deployed CPOE systems as of the middle of 2012.

Microsoft’s Azure Goes Off-line Because Of Embarrassing Oversight

Last Friday afternoon, Microsoft shot itself in the foot when it let an SSL (secure sockets layer) certificate expire, taking down its Azure cloud services, the AP reported. A Forbes story said the outage eventually lasted over 12 hours. And according to a story at ComputerWorld, Microsoft also admitted that the issue caused problems with its Xbox Music and Video Store services.

ComputerWorld quoted Microsoft as saying, “Beginning Friday, February 22 at 12:44 PM PST, Storage experienced a worldwide outage impacting HTTPS operations (SSL traffic) due to an expired certificate… We apologize for any inconvenience this causes our customers.”

Well, Microsoft was at least forthcoming about what the problem was, despite how embarrassing the admission may have been. As you may remember from last week, IBM’s New Zealand Virtual Server Services suffered an outage that lasted from early Monday morning into Wednesday. IBM, however, still hasn’t explained what caused the outage or even bothered with a pro forma apology to its customers for the inconvenience caused by the outage.

Software Update Takes Out Communications to International Space Station

On Tuesday morning, a routine software update to the International Space Station’s flight systems caused the ISS to lose communication with NASA’s Houston Mission Control. According to a story at the Houston Chronicle, “Houston flight controllers were updating software onboard the station's flight computers at 8:45 a.m. when one of the station's data relay systems malfunctioned. Although a backup computer took over critical station functions from the primary computer, a NASA press alert noted that the station was not able to communicate with NASA's Tracking and Data Relay Satellites.”

The crew was able to restore communications via a back-up computer by around 11:30 a.m. CST—about the time when the ISS flew over Russia. I suspect the crew was secretly happy for the brief comms blackout while it lasted.

Glitches Hit Parking Systems in Pittsburgh and Vancouver

What software glitches giveth, they also taketh away. The Pittsburgh Post-Gazette reported that a programming error in Pittsburgh’s 550 new parking meter pay stations inadvertently gave motorists free parking last Monday at 3500 parking spaces. Though Monday was a U.S. federal holiday (Presidents Day), and many cities don’t charge for parking on federal holidays, Pittsburgh had intended to get revenue from the new meters. But the programming glitch (or oversight) caused all the pay stations to display messages indicating that parking was free because of the holiday.

Pittsburgh’s parking authority corrected the error by late Monday morning. But by then, it was too late to figure out when motorists had parked in the spots. So the parking authority decided to forego the normal fees for the rest of the day at the spaces covered by the pay stations. However, motorists still had to pay at the 4000 or so metered parking spots throughout the city that were not affected by the glitch.

Pittsburgh’s parking authority didn’t know how much money it failed to collect, but it didn’t think it lost too much because traffic was light—because of the holiday.

Motorists in Vancouver weren’t so lucky.  According to the Vancouver Sun, about 1000 motorists who parked at Diamond-run parking lots last Wednesday and Thursday and paid by credit or debit card were charged multiple times because of a “glitch” in the payment system. Diamond told the Sun that the problem occurred after “Elavon [the company that processes payments for Diamond] updated its processing system.”

Diamond apologized for the inconvenience.

Facebook: No One is Over 100

Our final glitch concerns a programming misstep by the good (young) folks at Facebook. The social media site won’t accept a registrant’s age if it is greater than 99. According to an AP story last week, Marguerite Joseph, a Michigan grandmother who is 104 and an avid Facebook user, has been unable to list her correct age on her Facebook page.  Every time her granddaughter, who takes care of the page for Mrs. Joseph, puts in her birth date of 1908, Facebook changes it to 1928. So Mrs. Joseph has had to settle on being 99 years of age for the past two years.

Facebook initially had no comment about the story, but soon admitted to the AP in another story that there is a “glitch in the system” with accepting pre-1910 birth dates. Facebook is now working on a fix. It has also apologized to Mrs. Joseph.

Photo: Rafe Swan/Getty Images

The Conversation (0)