Objective: To describe early experience of replacing PSA with Stockholm3 for detection of prostate cancer in primary care. Design and methods: Longitudinal observations, comparing outcome measures before and after the implementation of Stockholm3. Setting: Stavanger region in Norway with about 370,000 inhabitants, 304 general practitioners (GPs) in 97 primary care clinics, and one hospital. Intervention: GPs were instructed to use Stockholm3 instead of PSA as standard procedure for diagnosis of prostate cancer. Main outcome measures: Proportion of GP clinics that had ordered a Stockholm3 test. Number of men referred to needle biopsy. Distribution of clinically significant prostate cancer (csPC) (Gleason Score !7) and clinically non-significant prostate cancer (cnsPC) (Gleason Score 6), in needle biopsies. Estimation of direct healthcare costs. Results: Stockholm3 was rapidly implemented as 91% (88/97) of the clinics started to use the test within 14 weeks. After including 4784 tested men, the percentage who would have been referred for prostate needle biopsy was 29.0% (1387/4784) if based on PSA level !3ng/ml, and 20.8% (995/4784) if based on Stockholm3 Risk Score (p < 0.000001). The proportion of positive biopsies with csPC increased from 42% (98/233) before to 65% (185/285) after the implementation. Correspondingly, the proportion of cnsPC decreased from 58% (135/233) before to 35% (100/285) after the implementation (p < 0.0017). Direct healthcare costs were estimated to be reduced by 23-28% per tested man. Conclusion: Replacing PSA with Stockholm3 for early detection of prostate cancer in primary care is feasible. Implementation of Stockholm3 resulted in reduced number of referrals for needle-biopsy and a higher proportion of clinically significant prostate cancer findings in performed biopsies. Direct healthcare costs decreased.

Bente Thorsen er spesialist i allmennmedisin og i samfunnsmedisin og er samhandlingsoverlege og PKO-leder på Oslo universitetssykehus. Forfatteren har fylt ut ICMJE-skjemaet og oppgir ingen interessekonflikter. KARIN FRYDENBERG Karin Frydenberg er spesialist i allmennmedisin, fastlege ved Skreia legesenter og PKO-leder på Sykehuset Innlandet. Forfatteren har fylt ut ICMJE-skjemaet og oppgir ingen interessekonflikter. KNUT-ARNE WENSAAS Knut-Arne Wensaas er spesialist i allmennmedisin, fastlege ved Kalfaret legesenter og seniorforsker ved Allmennmedisinsk forskningsenhet, Norwegian Research Centre, Bergen. Forfatteren har fylt ut ICMJE-skjemaet og oppgir ingen interessekonflikter. LINE CECILIE CHRISTIANSEN Line Cecilie Christiansen er spesialist i allmennmedisin og fastlege ved Stavanger medisinske senter. Forfatteren har fylt ut ICMJE-skjemaet og oppgir ingen interessekonflikter. BODIL AASVANG OLSEN Bodil Aasvang Olsen er spesialist i allmennmedisin og fastlege ved Tvedestrand legesenter. Forfatteren har fylt ut ICMJE-skjemaet og oppgir ingen interessekonflikter. KARSTEN KEHLET Karsten Kehlet er spesialist i allmennmedisin og fastlege ved Senjalegen, avdeling Silsand legekontor. Forfatteren har fylt ut ICMJE-skjemaet og oppgir ingen interessekonflikter. ELLEN ANITA FAGERBERG Ellen Anita Fagerberg er spesialist i allmennmedisin, fastlege ved Abildsø legekontor og praksiskonsulent på Oslo universitetssykehus. Forfatteren har fylt ut ICMJE-skjemaet og oppgir ingen interessekonflikter. KJELL OLAV B. SVENDSEN Kjell Olav B. Svendsen er fastlege ved Frogner helsesenter og lektor ved Universitetet i Oslo. Forfatteren har fylt ut ICMJE-skjemaet og oppgir ingen interessekonflikter. TOVE BORGEN Tove Borgen er spesialist i allmennmedisin, fastlege ved Frysja legesenter og praksiskonsulent på Diakonhjemmet Sykehus. Forfatteren har fylt ut ICMJE-skjemaet og oppgir ingen interessekonflikter.

Objective To investigate how GPs use the PSA test as a diagnostic tool in daily practice. Design Qualitative study using focus group interviews, the transcripts being analyzed by systemic text condensation. Subjects A total of 17 Norwegian GPs in three CME groups. Main outcome measures Exploring GPs’ attitudes to national guidelines and the practical use of the PSA test. Results Detecting prostate cancer in general practice is a common and important, but difficult diagnostic issue. Our participants experienced uncertainty regarding the test when to use it, how to interpret the results and when to refer to specialist health services. Conclusion The study revealed a general ambivalence to the use of PSA. Many patients present urological problems, and many are afraid of having cancer. PSA is commonly used, but sometimes generates problems rather than solving them. Implications The use of the PSA test should be based on a thorough clinical assessment and in close collaboration with the patient. Key points Many patients in general practice present urological problems, and many are afraid of having cancer. GPs have a general ambivalence to the use of PSA when to use it, how to interpret the results and when to refer to specialist health services. The use of PSA sometimes generates problems rather than solving them.