In the last few days we have had a lot of discussion of the following report.
Discharge reports found wanting
Half of all hospital discharge summaries omit significant clinical information and one in 10 lack the main diagnosis, according to an audit of a tertiary hospital.
Researchers took a random sample of 150 summaries at Maroondah Hospital in Melbourne and found 12% were missing the principle diagnosis — including the diagnosis of sepsis in a patient treated for a UTI.
Acute renal failure, anaemia and electrolyte disturbances were the most commonly omitted comorbidities in discharge documentation.
One patient presented with fever, rigors, hypotension and elevated inflammatory markers and was treated for Proteus mirabilis urosepsis. On discharge, his principal diagnosis was documented as "UTI".
In an article soon to be published in the Internal Medicine Journal, the authors said the study was designed to track whether clinical activity in the hospital was being accurately coded to ensure the hospital attracted all the funding available.
They said associated diagnoses and a full list of complications were often missing because the interns writing the summaries did not think them serious.
More here:
Here is another report:
Patients urged to check medical records after errors havoc
Date February 1, 2013
Julia Medew
The study of 150 patients' discharge summaries at Maroondah Hospital in Melbourne during 2011 and 2012 found half were missing significant clinical information and one in 10 had the wrong diagnosis.
PATIENTS are being urged to keep a close eye on their medical records amid growing evidence that hospital staff are regularly making costly and dangerous mistakes .
Epworth Freemasons acknowledged on Thursday it had accidentally sent a Melbourne mother home this month with records showing she had had a baby boy instead of a baby girl.
The records listed the correct name and address of the mother, but said her male baby's testes had been checked, among other things. The documents also included the wrong doctors' name and said the mother had suffered a fractured pelvis when this was not the case.
The mother, who did not want to be named, told Fairfax Media the erroneous records had caused havoc in the first three weeks of her baby's life, making it difficult to register her infant with the Office of Births, Deaths and Marriages.
She was also shocked to hear a maternal child health nurse lament her fractured pelvis and use male growth charts for her baby girl.
While the hospital has apologised and corrected the records, the mother said it made her doubt the care she and her baby received at the hospital in East Melbourne.
''The fact that it wasn't just missing information but recorded that the testes were checked puts a question mark over everything,'' she said.
Lots more here:
And from the US we have reports like this.
EHR Accuracy Remains Problem, CHIME Says
College of Healthcare Information Management Executives warn technology and workflow burdens make EHR reporting nearly impossible.
By Michelle McNickle, InformationWeek
January 31, 2013
URL: http://www.informationweek.com/healthcare/electronic-medical-records/ehr-accuracy-remains-problem-chime-says/240147454
January 31, 2013
URL: http://www.informationweek.com/healthcare/electronic-medical-records/ehr-accuracy-remains-problem-chime-says/240147454
Recent comments made by members of the College of Healthcare Information Management Executives (CHIME) make clear the organization doubts hospitals' ability to submit accurate and complete data through electronic health records (EHRs). CHIME members' comments were made in response to a Jan. 3 Request for Information (RFI) issued by the Center for Medicare and Medicaid Services (CMS).
In a statement, CHIME members commended federal efforts made toward "reaching a harmonized approach" for electronic clinical quality measurement and reiterated its support for aligning EHR-based reporting and hospital quality reporting programs. But the "number one thing" the organization wanted to convey to CMS is that quality measurements through EHRs are "extremely time intensive and difficult," said Jeffery Smith, assistant director of public policy at CHIME, in an interview with InformationWeek Healthcare. 
"We want to make sure they understand that as far as data coming together in an electronic format, it doesn't seem like sending data electronically will be difficult," Smith said. "But getting accurate and complete measures is really difficult."
See link in text.
The bottom line in all this is the old chestnut of ‘garbage in, garbage out’. The other issue is that data quality can only really be assured if those recording the data 1. Know what they are doing and 2. Have some ‘skin in the game’ in terms of needing to use and being accountable for the information captured.
There is zero motivation for a GP to record wrong information in the records they rely on to treat their own patients but similarly there is less pressure on the intern who is leaving the rotation in the next week to be as thorough as might be desired with the discharge summary.
It is my belief that sadly we need a mix of both carrot and stick - as well as ongoing serious investment in education - if there is ever going to be able to be real trust in shared health information. Additionally patient access to their record can help locate and correct errors and is really a good idea as it has been shown to improve patient understanding and engagement as well as record accuracy.
You can read about the work in the Open Notes Movement and so on here:
and here:
To me this is an idea whose time has come and which can also improve information quality.
David.