Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Decision support for ordering blood tests in primary care


General practitioners order blood tests on one of every 25 patients they see. They will usually order not one test, but perhaps seven or eight different tests each time. Since they see a lot of patients, that means they order a lot of blood tests. Three things follow. First that this costs a lot. Second, that GPs have a lot of information to digest. Third, since normal ranges are often defined as the middle 95% of results in a normal population, 5% will be "abnormal" even when there is no disease. So GPs end up chasing lots of normally abnormal results with nothing to show.

The picture is even more complicated because all the guidelines and guidance being produced - about 5 kg of it, about a metre high, in each GP's office. And now the GPs are being audited on these guidelines. So can technology help, and how do we know it helps? A new randomised trial from Holland [1] shows the way.


The setting was the Delft region of Holland with 60 general practitioners in 44 practices with a median of about 3,400 patients per practice. The practices had replaced paper-based patient records with electronic records. Practices were randomised by computer-generated random numbers to receive one of two versions of a computer decision support system for blood test ordering.

One version initially displayed a reduced list of tests. It offered GPs an initial set of 15 tests covering most of the clinical situations seen in primary care. Other tests could also be ordered. Test ordering could be customised by adding or deleting tests for individual patients.

The alternative system was similar, but based on the 54 guidelines from the Dutch College of General Practitioners. These focus on symptoms commonly seen in primary care, or diseases commonly seen in primary care. GPs could select guidelines to see that tests being ordered (or not ordered) were relevant for a particular clinical situation.

Paper forms could still be used, though computer ordering was preferred. The introduction of the computer decision aid systems was accompanied by a three month phase-in period after an orientation presentation. Information on ordering was then followed for a year, with the main outcome being the number of order forms and tests ordered.


The main results are shown in Table 1. There was no difference between the baseline characteristics of practices or GPs. The different computer systems made no difference to the number of patients for whom blood tests were ordered. Computer generated blood test numbers was lower with the guideline assistance. The number of tests per request was down by an average of 20%, from 6.9 ± 1.6 (SD) to 5.5 ± 0.9. This was significantly lower.

Table 1: Main results

  Restricted Guidelines
Number of practices 21 23
Number of GPs 29 31
Number of patients 77,336 78,461
Number of test order forms 12,742 12,668
Computer ordered percent 88 71
Average number of tests 6.9 5.5
Requests per patient per year 0.16 0.16
Tests per patient per year 1.14 0.89

Not all tests were equally affected. Of the 20 most commonly requested tests that made up 80% of tests ordered, significant reductions occurred only in some (Table 2). But these were large reductions, of about half for some liver enzymes and about a third for some haematological tests.

Table 2: Results for individual tests

Test Percent reduction
Potassium 50
AST 49
Gamma-GT 46
Free thyroxine 43
ALT 38
MCV 33
Creatinine 32
Erythrocyte count 30
ESR 28
Haematocrit 26
Differential leucocyte 26
Leucocyte count 23
Hb 18


The implications of this are really impressive. Dutch GPs without computer-generated guideline support ordered 1.14 tests per patient per year. With guideline support this reduced to 0.89 tests per patient per year. The implication for a laboratory serving a million people would be that guideline-assisted test ordering would reduce the workload by 250,000 tests a year, or about 5,000 tests a week. For particular tests, not all of them inexpensive, the impact would be significant.

Moreover, because this behaviour would be based on guidelines, presumably built around best evidence, the saving would have no negative impact on the health of the population. Indeed, perhaps as many as 12,500 falsely abnormal test results would be avoided, as well as additional unnecessary diagnostic procedures, hospital visits, professional concern and personal worry.

Improved quality does not necessarily come at increased cost. Doing simple things well works extremely well. A metre of paper guidelines doesn't help. On the computer, available when we need them, they seem to.


Check out the correspondence about this story here


  1. MA van Wijk et al. Assessment of decision support for blood test ordering in primary care. A randomized trial. Annals of Internal Medicine 2001 134: 274-281.
previous or next story in this issue