The use of prostate-specific antigen (PSA) tests in general practice in Denmark has been rising since their introduction in 1997 [1]. A PSA tests may be used as a diagnostic test or even a screening instrument or as a control test for follow-up on incident prostate cancer (PC) cases. The largest increase in the use of PSA testing has been seen in non-symptomatic men as part of opportunistic or ‘grey area’ screening. This use remains not recommended internationally as well as in Denmark [2], [3]
We created a web-based CDSS and pushed its web-address to the intervention group which consisted of 114 practices with 247 GPs in the Central Denmark Region who were using the EMR software MedWin™.
Whenever the GP entered the keywords “PSA” or “PROSTATE” into the medical chart contents, the EMR replaced the word with a hyperlink to the CDSS. The hyperlink format was in an accepted format, coloured and underlined to keep its purpose apparent. The control group consisted of all other practices in
General practice ordered a total of 47,545 PSA tests for 30,701 men in the study period. Table 1 presents baseline characteristics of general practices by group.
Table 2, Table 3, Table 4 present the age-standardised rate difference (RD) estimates for the three compared periods. Furthermore, the tables present the DIDs between the intervention and control groups; practices with a hosted EMR solution and practices with a local solution within the intervention group and practices with a hosted
The intervention as composed in our design had no measurable effect on the GPs’ PSA testing.
Including 357 practices and PSA tests from 30,000 men strengthens this study and its statistical precision. Although we did not use a randomised trial design, the practices’ choice of EMR system may be considered random, which also strengthens our study. In addition, all data originate from clinical databases which ensure a high validity.
The study was planned as a health services research study of how
Providing the GPs’ EMR with an automatic replacement of words like PSA or prostate with a hyperlink to a CDSS with information on whether to use PSA and how to interpret the test result had no measurable effect on the GPs’ PSA testing behaviour.
PV, FB, FO all conceived the study, participated in its design and helped draft the manuscript. TOM conceived the study, collected data, did the statistical analyses and wrote the manuscript. All authors read and approved the final manuscript.
The authors declare that they have no competing interests. Summary points What was already known on the topic? Prostate-specific antigen (PSA)-testing in General Practice can be ambiguous. General practitioners may benefit from guidance on best clinical practice when PSA-testing. Danish general practitioners have full access to the Internet from their clinics and they all use electronic medical record systems. What this study added to our knowledge? The study demonstrated that it was possible to grant
We would like to thank The Danish Council for Independent Research – Medical Sciences, TrygFonden, and the Central Denmark Region for financial support for the project. We also thank the software development team at EG Data Inform for coding and distributing the software development in MedWin™.
