Thanks to computers, we can take and manage notes more easily and efficiently than ever before. However, the flip-side of this revolution in working practice is that small inputting errors can cause major problems when the data are processed and accessed.
GPs are now seeing the fruits of their labour in coding their quality and outcomes framework (QOF) data. Practices are receiving funding based on the performance data they have submitted to the Quality Management and Analysis System (QMAS). Audit of the data enables outcomes to be measured, encouraging practices to improve the quality of care they deliver.
Getting the most from the computer
The strength of the computer lies in its ability to process massive amounts of information. It can deliver guidelines customised locally or by the practice to the GP, thereby supporting decision making in the consultation. It also saves GPs from manually reproducing information. However, computers can only replicate the data that have been entered and as the information can be retrieved and used in many settings, there are many opportunities for any errors to proliferate.
Many clinicians follow their first instincts when they switch from paper to PC records, using free text as much as possible. They appreciate the ease with which they can retrieve their notes and their legibility.
However, using free text will prevent doctors from fully exploiting the potential of IT, because it is difficult to retrieve and audit.We can reap the full benefits of computerised data only when the coding is accurate.
The impact of coding errors
It is common to see coding errors that may seem insignificant to the casual observer but, when the data quality is assessed, are found to have major significance.
Data only become information when entered appropriately, into the correct position, in an electronic record. Incorrectly entered data are irrelevant and potentially dangerous. The following example illustrates this.
A new patient joins the practice. She is not an asthma sufferer, although her mother is. The practice nurse enters the code for ‘History of asthma’ into the record and adds, in free text, that the patient’s mother has asthma.This is incorrect because the history relates to the family member and not to the new patient, to whom the code should relate.
Instead, the practice nurse might have entered the code for ‘Asthma’ and added in free text that the patient’s mother had asthma. However, this is also incorrect.
Any search on either of these codes will wrongly identify the new patient as suffering from asthma. In fact, the correct way to code this patient’s notes is ‘Family history of asthma’. Free text could then be used to identify the patient’s mother as the asthma sufferer (see Figure 1, below).
|Figure 1: Screen shot showing examples of correct and incorrect coding for a fictitious patient with a family history of asthma|
The forms of coding described above would appear in the medical records as:
- H/O: asthma – mother has
- Asthma – mother has
- FH: asthma – mother has.
The difference appears small, but from a search and reporting perspective it is huge.
The importance of accurate data
Few receptionists, IT administrators, administrative assistants or clinicians understand these subtle but critical distinctions. Even after more than 15 years of using codes in general practice,there is still an enormous need for training in how to do this work properly and abandon bad habits.
However, good practice guidelines have been issued by the Department of Health, and practices that enter clinical data on their computers should comply with these (see Box 1, 2 and 3 below).
|Box 1: Ensuring that data are complete, accurate and relevant|
Good practice guidelines for general practice electronic patient records recommend that practices should consider:
To ensure complete, accurate and relevant data, practices should consider the following:
Practices should develop systems for:
|Box 2: Practice education and training needs|
Good practice guidelines recommend that practice staff should receive training in:
|Box 3: Sources of further information|
Until now, coding errors had significance only for the practice that entered the information. No lasting damage was done by a carelessly entered code.
During a quality visit to a practice, I was examining a set of notes on which codes for seven different types of malignancy and one for hypothyroidism had been entered, all on the same day.
On looking at the manual set of notes, which had clearly been ‘summarised’, the underlying errors became apparent.
Details of the patient’s family history of malignancies had been given ‘B2’ hierarchy codes – malignancy codes – instead of ‘family history’ hierarchy codes.
Instead of the date of diagnosis of hypothyroidism recorded on the manual records, the summariser had entered the date when the summary was made.This date was also entered as the date of diagnosis of the cancers.
The impact of errors like these will be felt in various ways now that data are starting to flow electronically from one part of the health service to another.
Capitalising on accurate data collection
During the coming year, the GP2GP project will start to roll out.This will enable practices to send electronic records direct to patients’new practices when they move to another area.The benefits to both patients and practices are immense.
Speedy transfer of notes is eagerly anticipated. Notes summarised by one practice can be sent to the patient’s new one, thus eliminating the need to repeat the time-consuming process of summarising.
QOF data can also be transferred by this means.Accurately coded data will then be paramount, because not only will patients’ notes be transferred from one practice to another, but the codes on which practices base their QOF points – and therefore their funding – will be transferred as well.
During the second half of 2006 the summary record will be generated, and general practice will be a major source of the information it contains.
The National Clinical Leads for General Practice within Connecting for Health are suggesting that practices be accredited for their standards of data entry before they supply data for the summary record.
We are arguing that the quality of the summaries the practices are signing off for their QOF points should meet those standards (see Box 3 above).
The ‘rubbish in, rubbish out’ adage of the IT world is true.The quality of the data we use depends greatly on the quality and accuracy of the information entered onto the computer in the first place.
GPs need to validate data and clinical information with their patients during the consultation and ensure that the data are correctly coded when entered onto the computer.