Exploring shared decision-making in prostate cancer screening within primary health care settings in Portugal: a cross-sectional study Authors: Beatriz Oliveira, Laura Lopes, Susana de Oliveira Branco, Sílvia Oliveira, Afonso Granja, Francisca Gonçalves, Joaquim Pinto, Margarida Silva, Sofia Coelho, Avelino Fraga, Frederico Teves Published: 2026 ABSTRACT BackgroundThis study aims to assess how Primary Care Physicians (PCPs) in Portugal currently approach opportunistic prostate cancer (PCa) screening using Prostate Specific Antigen (PSA) testing, and to explore variations in both screening and shared decision-making (SDM) practices.MethodsAn electronic survey was distributed in 25 Health Centers in Portugal between June 2024 and June 2025. The questionnaire assessed respondents’ demographic and practice characteristics, PSA screening criteria and frequency, implementation of SDM, knowledge and use of decision aids, and urology referral patterns. Associations between provider characteristics and screening practices were analyzed using Pearson chi-square or Fisher’s exact tests.ResultsA total of 107 PCPs participated in the study, the majority being female (81.3%) and under 40 years old (72.9%). Those with less than 10 years of experience were more likely to adjust PSA testing frequency based on initial PSA levels, especially for men aged 50–59 (p = 0.015) and 60–74 (p = 0.004). Routine screening of younger men or those over 75 was uncommon unless specific clinical factors were present. National guidelines were the primary influence on clinical practice (75.7%), particularly among less experienced physicians (p = 0.027). In terms of SDM, most PCPs explained PSA testing (80.4%), but only 64.5% consistently discussed its benefits and risks or explored patient preferences, with those in the South region of Portugal more likely to do so (p = 0.012 and p = 0.002 respectively) than those in the North/Centre. Decision aid awareness and use were low; only 6.5% reported having used them, while 72% were unfamiliar with these tools. Most referrals to urology were triggered by abnormal transrectal ultrasound findings (72%) rather than elevated PSA alone (16.8%). Full support for population-based PCa screening was limited (12.2%), and male physicians were significantly more likely to endorse it (p = 0.007).ConclusionsPCa screening practices among Portuguese PCPs remain a complex area of primary care, characterized by variability in clinical approaches, inconsistent guideline adherence, and limited application of SDM. Despite general awareness, notable gaps persist in the use of decision aids and evidence-based referral criteria. These results highlight the need for clearer national guidance and greater support for individualized, patient-centered screening strategies.Trial registrationnot applicable.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s12875-026-03409-z. Background Prostate cancer (PCa) is among the most frequently diagnosed cancers in men worldwide [1]. Rising incidence rates may reflect changes in population aging, overall population growth, or shifts in age-specific risk In Portugal, as in many countries, incidence varies regionally and appears closely linked to the use of prostate-specific antigen (PSA) screening [2]. Since 1998, national incidence rates have risen by an average of 1.8% per year—except in the northern region, where they have declined by 3.6% annually since 2005 [3]. Despite the widespread use of the prostate-specific antigen (PSA) test for PCa early detection/screening, major randomized trials and systematic reviews show that, although PSA screening increases PCa detection, it does not consistently translate into reduced disease-specific mortality [4, 5]. Alongside uncertain mortality benefit, PSA screening carries risks of overdiagnosis and overtreatment, exposing many men to unnecessary biopsies and treatments with potential harms [6]. As a result, recommendations from organizations such as the American Urological Association (AUA), European Association of Urology (EAU), and national guidelines from the Portuguese Directorate-General of Health / Direção Geral da Saúde (DGS) vary in terms of target populations, screening intervals, and consideration of life expectancy. The AUA recommends shared decision-making for men aged 55–69 or > 70 years with 10–15 years life expectancy; the EAU proposes risk-adapted screening starting at age 50, and the DGS emphasizes screening starting at age 50 with no life expectancy limit [6–8]. In Portugal, this PCa screening is undertaken by Primary Care Physicians (PCPs). These conflicting results and guidelines reflect ongoing uncertainty and the need for patient-centered approaches. To tackle this complex matter, one common recommendation found in all these recommendations is Shared Decision-making (SDM). SDM is an interactive process where clinicians and patients work together to make healthcare choices. This method involves sharing information, discussing preferences, and jointly deciding on the best course of action. SDM ensures that patients have a central role in decision-making, promoting their autonomy and engagement. For this to be effective, healthcare providers must deliver transparent and relevant information about all options—including risks and benefits—and encourage patients to reflect on these choices in relation to their personal values and priorities. In the context of prostate cancer screening, these discussions address particularly sensitive issues, as opting for PSA testing can entail potential harms—including increased anxiety, unnecessary prostate biopsies, overdiagnosis, and overtreatment, which may result in adverse effects such as erectile dysfunction and urinary incontinence [9, 10]. The challenge is further compounded by the significant responsibility placed on physicians, who must tailor their communication strategies to each patient’s unique background and needs. Educational efforts and providing sufficient information to make an informed choice are essential first steps to SDM; without them, patient choice will mainly rely on PCPs. Therefore, a very important aspect of SDM from a provider perspective is to provide information, and decision aids can be useful at this stage. Decision aids are evidence-based tools—such as pamphlets, videos, or web-based resources—designed to support patients facing specific health treatment or screening decisions. Unlike general educational materials, decision aids explicitly outline the available options, present balanced and clear information about the potential benefits, risks, and uncertainties associated with each choice, and guide individuals to consider how these options align with their personal values and preferences [11]. Decision aids are explicitly recommended in AUA Guidelines [7, 12], as their efficacy in increasing patient knowledge prior to PCa screening decisions is documented in four systematic reviews [13–16]. However, previous studies have suggested inconsistencies in screening practices, with variations influenced by physician experience, clinical setting, and familiarity with national and international guidelines [17–21]. Additionally, despite growing emphasis on SDM as a best practice in PSA screening, its actual implementation in primary care remains inconsistent [14, 22, 23]. The use of structured decision aids—which can enhance patient understanding and align decisions with individual values—also appears limited [24]. In Portugal, even though a culturally-adapted decision aid has already been published [25], there is a proeminent lack of research on these topics. This study aims to characterize current PCa screening practices among Portuguese PCPs, with a specific focus on PSA testing criteria and frequencies, SDM implementation, awareness and use of decision aids, and referral patterns to urology. We aimed to identify any difference in screening practice between PCPs by their gender, years of experience, area of practice, and setting of practice (urban or intermediate). Methods Study design and participants A cross-sectional study design was used. This study was approved by the Ethics Committee of ULS Santo António | ICBAS, and by the Office of Research Projects of the Clinical Academic Centre with the reference number 2024.102 (089-DEFI/091 CE). Eligible participants were medical doctors currently practicing as PCPs in Portugal. Participation was voluntary and anonymous, and informed consent was obtained before the commencement of the survey. No incentives were provided for any of the respondents. 40 Health Units (HU) in Portugal were contacted via e-mail to invite their working PCPs to participate in the study. From these, 25 responded positively via their coordinator – eight in Porto, two in Aveiro, four in Coimbra, ten in Lisbon, and one in the Algarve. The survey was distributed from the HC coordinator via email to the corresponding HC’s PCPs. The survey was open for responses from June 2024 to June 2025. Although the number of PCPs employed at the contacted HCs was known (estimated at n = 288), it was not possible to determine how many individual physicians received or opened the email or survey link. Therefore, a formal response rate could not be calculated. Definition of SDM and outcome variables The survey distributed in this study was adapted from a questionnaire originally developed by Ranasinghe et al. [26] in the USA to evaluate perceptions and attitudes toward PSA screening and treatment in average-risk men—defined as those with no significant comorbidities, no family history of prostate cancer, and a life expectancy of more than 10 years. It was initially piloted among a small group of general practitioners (n = 10) in Australia. The survey question items included the characteristics of the respondents, such as gender, age group, years of experience in clinical practice, and location of practice (urban, intermediate, or rural, as defined by Eurostat’s Methodological Manual on Territorial Typologies [27]). The questionnaire assessed general knowledge on PSA screening, knowledge and opinion on decision-aids, experience with SDM, and opinion on population-based PCa screening. SDM experience assessment was done through questionnaire items that were adapted from the SDM Model by Pekala et al. [24], based on a three-step model (choice talk, option talk, and decision talk). That includes patient education, acknowledgment of the need for a decision, discussion of options and risks, and elicitation of preferences [28]. A detailed description is provided in Fig. 1, with the full questionnaire in Additional File 1. Fig. 1Outcome variables corresponding to specific items within the questionnaire. The last component of the questionnaire reflects the model of full shared decision-making for PSA screening, adapted from ref [28]. The process begins with choice talk, where information about screening is provided to the patient. This is followed by option talk, involving a detailed discussion of benefits and risks and the inclusion of decision support, such as decision aids and time for patient reflection and questions between consultations. The decision talk stage enables patients to assess the benefits and harms of screening in light of their individual priorities and preferences. Ultimately, the process concludes with a collaborative decision-making step, in which the patient and clinician reach a decision together. PSA, prostate-specific antigen Outcome variables corresponding to specific items within the questionnaire. The last component of the questionnaire reflects the model of full shared decision-making for PSA screening, adapted from ref [28]. The process begins with choice talk, where information about screening is provided to the patient. This is followed by option talk, involving a detailed discussion of benefits and risks and the inclusion of decision support, such as decision aids and time for patient reflection and questions between consultations. The decision talk stage enables patients to assess the benefits and harms of screening in light of their individual priorities and preferences. Ultimately, the process concludes with a collaborative decision-making step, in which the patient and clinician reach a decision together. PSA, prostate-specific antigen Statistical analysis Descriptive statistics were calculated for all variables, namely frequencies and percentages for categorical variables. Associations between categorical variables (e.g., gender, practice setting, region) and binary outcomes were assessed using the Chi-square test of independence or Fisher’s exact test when expected cell counts were below 5. For outcomes with more than two categories (e.g., frequency of PSA testing), Chi-square tests were also used to assess associations with demographic variables. Percentages reported in the text refer to the entire sample, as the survey required a response to every question to be submitted. All tests were two-sided with a significance level set at < 0.05. Analysis was performed with R version 4.4.3 (2025-02-28). Results Sample characteristics All physicians who completed the survey were included in the analysis, resulting in a final convenience sample of 107 respondents. The majority were female (81.3%), under 40 years of age (72.9%), and had less than five years of clinical experience (47.7%). The distribution of respondents across practice settings and regions was relatively even, as presented in Table 1. Table 1Characteristics of respondentsCharacteristics of respondentsNumber of respondents (n/%)Gender Male20 (18.7) Female87 (81.3)Region of practice North35 (32.7) Centre13 (12.1) South59 (55.1)Setting of practice* Rural4 (3.7) Intermediate/rural50 (46.7) Urban53 (49.5)Age range ≤ 4078 (72.9) 41 to 4921 (19.6) 50 to 594 (3.7) 60 to 694 (3.7)Years in practice < 551 (47.7) 5 to 1025 (23.4) 10 to 2024 (22.4) 20 to 304 (3.7) > 303 (2.8)* Due to small sample size, PCPs working in the intermediate and rural sectors were combined and labelled as intermediate setting in the analysis. For statistical purposes, the “years in practice” category was simplified to two groups: “less than 10 years/ <10”, and “10 years or more/≥ 10 ” Table 1Characteristics of respondentsCharacteristics of respondentsNumber of respondents (n/%)Gender Male20 (18.7) Female87 (81.3)Region of practice North35 (32.7) Centre13 (12.1) South59 (55.1)Setting of practice* Rural4 (3.7) Intermediate/rural50 (46.7) Urban53 (49.5)Age range ≤ 4078 (72.9) 41 to 4921 (19.6) 50 to 594 (3.7) 60 to 694 (3.7)Years in practice < 551 (47.7) 5 to 1025 (23.4) 10 to 2024 (22.4) 20 to 304 (3.7) > 303 (2.8)* Due to small sample size, PCPs working in the intermediate and rural sectors were combined and labelled as intermediate setting in the analysis. For statistical purposes, the “years in practice” category was simplified to two groups: “less than 10 years/ <10”, and “10 years or more/≥ 10 ” Characteristics of respondents * Due to small sample size, PCPs working in the intermediate and rural sectors were combined and labelled as intermediate setting in the analysis. For statistical purposes, the “years in practice” category was simplified to two groups: “less than 10 years/ <10”, and “10 years or more/≥ 10 ” Current practice of PCa screening For men aged 40–49 years without prior history, PCPs with fewer years of experience (p = 0.034) and those practicing in urban settings (p = 0.019) were less likely to perform PSA testing. No significant differences were observed by gender or region. Among men > 50 years, PCPs in intermediate settings were more likely to request PSA testing (p = 0.049), with no differences by gender, experience, or region. No significant differences were found across all variables for men > 40 years with BRCA2 mutations or men > 45 years with a positive family history. For men > 75 years with > 10 years life expectancy, less experienced PCPs (< 10 years) were more likely to request PSA testing (p = 0.003) (Table 2). Table 2Frequencies of criteria for requesting the PSA test by subgroupsn (%)NeverNot alwaysAlwaysP valueIn men aged 40 to 50 years without prior historyGender Female53 (60.9)33 (37.9)1 (1.1)0.552 Male10 (50.0)10 (50.0)0 (0.0)Years of experience < 1050 (65.8)25 (32.9)1 (1.3) 0.034 > 1013 (41.9)18 (58.1)0 (0.0)Setting Urban37 (69.8)15 (28.3)1 (1.9) 0.019 Intermediate/rural26 (48.1)28 (51.9)0 (0.0)Region North and Centre27 (56.3)21 (43.8)0 (0.0)0.756 South36 (61.0)22 (37.3)1 (1.7)In men aged more than 50 yearsGender Female8 (9.2)57 (65.5)22 (25.3)0.192 Male2 (10.0)9 (45.0)9 (45.0)Years of experience < 106 (8.7)51 (73.9)12 (17.4)0.097 > 104 (13.8)15 (51.7)10 (34.5)Setting Urban7 (13.2)36 (67.9)10 (18.9) 0.049 Intermediate/rural3 (5.6)30 (55.6)21 (38.9)Region North and Centre2 (4.2)28 (58.3)18 (37.5)0.089 South8 (13.6)38 (64.4)13 (22.0)In men aged more than 40 years with BRCA2 mutationGender Female2 (2.3)22 (25.3)63 (72.4)0.852 Male0 (0.0)4 (22.2)16 (88.9)Years of experience < 101 (1.3)19 (25.0)56 (73.7)0.785 > 101 (3.2)7 (22.6)23 (74.2)Setting Urban1 (1.9)16 (30.2)36 (67.9)0.369 Intermediate/rural1 (1.9)10 (18.5)43 (79.6)Region North and Centre0 (0.0)16 (33.3)32 (66.7)0.076 South2 (3.4)10 (16.9)47 (79.7)In men aged more than 45 years with a family history of PCaGender Female0 (0.0)28 (32.2)59 (67.8)0.602 Male0 (0.0)5 (27.8)15 (83.3)Years of experience < 100 (0.0)23 (30.3)53 (69.7)0.822 > 100 (0.0)10 (32.3)21 (67.7)Setting Urban0 (0.0)18 (34.0)35 (66.0)0.535 Intermediate/rural0 (0.0)15 (27.8)39 (72.2)Region North and Centre0 (0.0)13 (27.1)35 (72.9)0.530 South0 (0.0)20 (33.9)39 (66.1)In men aged more than 75 years and with a life expectancy of more than 10 yearsGender Female9 (10.3)59 (67.8)19 (21.8)1.000 Male2 (11.1)14 (77.8)4 (22.2)Years of experience < 103 (3.9)55 (72.4)18 (23.7) 0.003 > 108 (25.8)18 (58.1)5 (16.1)Setting Urban7 (13.2)36 (67.9)10 (18.9)0.545 Intermediate/rural4 (7.4)37 (68.5)13 (24.1)Region North and Centre5 (10.4)29 (60.4)14 (29.2)0.206 South6 (10.2)44 (74.6)9 (15.3) Table 2Frequencies of criteria for requesting the PSA test by subgroupsn (%)NeverNot alwaysAlwaysP valueIn men aged 40 to 50 years without prior historyGender Female53 (60.9)33 (37.9)1 (1.1)0.552 Male10 (50.0)10 (50.0)0 (0.0)Years of experience < 1050 (65.8)25 (32.9)1 (1.3) 0.034 > 1013 (41.9)18 (58.1)0 (0.0)Setting Urban37 (69.8)15 (28.3)1 (1.9) 0.019 Intermediate/rural26 (48.1)28 (51.9)0 (0.0)Region North and Centre27 (56.3)21 (43.8)0 (0.0)0.756 South36 (61.0)22 (37.3)1 (1.7)In men aged more than 50 yearsGender Female8 (9.2)57 (65.5)22 (25.3)0.192 Male2 (10.0)9 (45.0)9 (45.0)Years of experience < 106 (8.7)51 (73.9)12 (17.4)0.097 > 104 (13.8)15 (51.7)10 (34.5)Setting Urban7 (13.2)36 (67.9)10 (18.9) 0.049 Intermediate/rural3 (5.6)30 (55.6)21 (38.9)Region North and Centre2 (4.2)28 (58.3)18 (37.5)0.089 South8 (13.6)38 (64.4)13 (22.0)In men aged more than 40 years with BRCA2 mutationGender Female2 (2.3)22 (25.3)63 (72.4)0.852 Male0 (0.0)4 (22.2)16 (88.9)Years of experience < 101 (1.3)19 (25.0)56 (73.7)0.785 > 101 (3.2)7 (22.6)23 (74.2)Setting Urban1 (1.9)16 (30.2)36 (67.9)0.369 Intermediate/rural1 (1.9)10 (18.5)43 (79.6)Region North and Centre0 (0.0)16 (33.3)32 (66.7)0.076 South2 (3.4)10 (16.9)47 (79.7)In men aged more than 45 years with a family history of PCaGender Female0 (0.0)28 (32.2)59 (67.8)0.602 Male0 (0.0)5 (27.8)15 (83.3)Years of experience < 100 (0.0)23 (30.3)53 (69.7)0.822 > 100 (0.0)10 (32.3)21 (67.7)Setting Urban0 (0.0)18 (34.0)35 (66.0)0.535 Intermediate/rural0 (0.0)15 (27.8)39 (72.2)Region North and Centre0 (0.0)13 (27.1)35 (72.9)0.530 South0 (0.0)20 (33.9)39 (66.1)In men aged more than 75 years and with a life expectancy of more than 10 yearsGender Female9 (10.3)59 (67.8)19 (21.8)1.000 Male2 (11.1)14 (77.8)4 (22.2)Years of experience < 103 (3.9)55 (72.4)18 (23.7) 0.003 > 108 (25.8)18 (58.1)5 (16.1)Setting Urban7 (13.2)36 (67.9)10 (18.9)0.545 Intermediate/rural4 (7.4)37 (68.5)13 (24.1)Region North and Centre5 (10.4)29 (60.4)14 (29.2)0.206 South6 (10.2)44 (74.6)9 (15.3) Frequencies of criteria for requesting the PSA test by subgroups Regarding PSA testing frequency, most PCPs did not recommend testing for men aged 40–49 or did so based on initial PSA levels. PCPs with < 10 years of experience were more likely to tailor testing frequency to initial PSA levels, particularly for men aged 50–59 (p = 0.015) and 60–74 (p = 0.004). For men > 75 years, the majority either did not recommend PSA testing or based it on PSA values (Table 3). 86% of PCPs reported that 5-alpha-reductase inhibitors affect PSA levels, while 9.3% were uncertain and only 4.7% believed these drugs have no effect. Years of clinical practice were significantly associated with this belief (p = 0.035), with more experienced physicians more likely to correctly identify the impact of 5-alpha-reductase inhibitors on PSA values. Table 3Frequency of PSA testingn (%)Every 2 yearsEvery 8 yearsDepends on initial PSAWould not recommendNot sureP valuePSA based screening for men aged 40 to 49Gender Female6 (6.9)1 (1.1)19 (21.8)52 (59.8)9 (10.3)0.741 Male1 (5.0)0 (0.0)7 (35.0)10 (50.0)2 (10.0)Years of experience < 105 (6.6)1 (1.3)21 (27.6)42 (55.3)7 (9.2)0.704 > 102 (6.5)0 (0.0)5 (16.1)20 (64.5)4 (12.9)Setting Urban4 (7.7)1 (1.9)8 (15.4)34 (65.4)5 (9.6)0.206 Intermediate3 (5.5)0 (0.0)18 (32.7)28 (50.9)6 (10.9)Region North and Centre3 (6.3)0 (0.0)15 (31.3)26 (54.2)4 (8.3)0.574 South4 (6.8)1 (1.7)11 (18.6)36 (61.0)7 (11.9)PSA based screening for men aged 50 to 59Gender Female37 (42.5)1 (1.1)34 (39.1)7 (8.0)8 (9.2)0.858 Male10 (50.0)0 (0.0)6 (30.0)2 (10.0)2 (10.0)Years of experience < 1030 (39.5)1 (1.3)35 (46.1)4 (5.3)6 (7.9) 0.015 > 1017 (54.8)0 (0.0)5 (16.1)5 (16.1)4 (12.9)Setting Urban25 (47.2)1 (1.9)16 (30.2)5 (9.4)6 (11.3)0.504 Intermediate/rural22 (40.7)0 (0.0)24 (44.4)4 (7.4)4 (7.4)Region North and Centre19 (39.6)1 (2.1)24 (50.0)2 (4.2)2 (4.2) 0.050 South28 (48.3)0 (0.0)16 (27.6)7 (12.1)7 (12.1)PSA based screening for men aged 60 to 74Gender Female37 (42.5)0 (0.0)35 (40.2)7 (8.0)8 (9.2)0.978 Male8 (40.0)0 (0.0)8 (40.0)2 (10.0)2 (10.0)Years of experience < 1028 (36.8)0 (0.0)38 (50.0)4 (5.3)6 (7.9) 0.004 > 1017 (54.8)0 (0.0)5 (16.1)5 (16.1)4 (12.9)Setting Urban24 (40.0)0 (0.0)25 (41.7)5 (8.3)6 (10.0)0.956 Intermediate/rural21 (44.7)0 (0.0)18 (38.3)4 (8.5)4 (8.5)Region North and Centre19 (39.6)0 (0.0)25 (52.1)2 (4.2)2 (4.2)0.058 South26 (44.1)0 (0.0)18 (30.5)7 (11.9)8 (13.6)PSA based screening for men aged more than 75 yearsGender Female10 (11.5)1 (1.1)34 (39.1)33 (37.9)9 (10.3)0.674 Male1 (5.0)0 (0.0)6 (30.0)11 (55.0)2 (10.0)Years of experience < 108 (10.5)1 (1.3)32 (42.1)28 (36.8)7 (9.2)0.474 > 103 (9.7)0 (0.0)8 (25.8)16 (51.6)4 (12.9)Setting Urban8 (15.4)1 (1.9)17 (32.7)20 (38.5)6 (11.5)0.533 Intermediate/rural3 (6.0)0 (0.0)18 (36.0)24 (48.0)5 (10.0)Region North and Centre3 (6.3)1 (2.1)21 (43.8)20 (41.7)3 (6.3)0.304 South8 (13.6)0 (0.0)19 (32.2)24 (40.7)8 (13.6) Table 3Frequency of PSA testingn (%)Every 2 yearsEvery 8 yearsDepends on initial PSAWould not recommendNot sureP valuePSA based screening for men aged 40 to 49Gender Female6 (6.9)1 (1.1)19 (21.8)52 (59.8)9 (10.3)0.741 Male1 (5.0)0 (0.0)7 (35.0)10 (50.0)2 (10.0)Years of experience < 105 (6.6)1 (1.3)21 (27.6)42 (55.3)7 (9.2)0.704 > 102 (6.5)0 (0.0)5 (16.1)20 (64.5)4 (12.9)Setting Urban4 (7.7)1 (1.9)8 (15.4)34 (65.4)5 (9.6)0.206 Intermediate3 (5.5)0 (0.0)18 (32.7)28 (50.9)6 (10.9)Region North and Centre3 (6.3)0 (0.0)15 (31.3)26 (54.2)4 (8.3)0.574 South4 (6.8)1 (1.7)11 (18.6)36 (61.0)7 (11.9)PSA based screening for men aged 50 to 59Gender Female37 (42.5)1 (1.1)34 (39.1)7 (8.0)8 (9.2)0.858 Male10 (50.0)0 (0.0)6 (30.0)2 (10.0)2 (10.0)Years of experience < 1030 (39.5)1 (1.3)35 (46.1)4 (5.3)6 (7.9) 0.015 > 1017 (54.8)0 (0.0)5 (16.1)5 (16.1)4 (12.9)Setting Urban25 (47.2)1 (1.9)16 (30.2)5 (9.4)6 (11.3)0.504 Intermediate/rural22 (40.7)0 (0.0)24 (44.4)4 (7.4)4 (7.4)Region North and Centre19 (39.6)1 (2.1)24 (50.0)2 (4.2)2 (4.2) 0.050 South28 (48.3)0 (0.0)16 (27.6)7 (12.1)7 (12.1)PSA based screening for men aged 60 to 74Gender Female37 (42.5)0 (0.0)35 (40.2)7 (8.0)8 (9.2)0.978 Male8 (40.0)0 (0.0)8 (40.0)2 (10.0)2 (10.0)Years of experience < 1028 (36.8)0 (0.0)38 (50.0)4 (5.3)6 (7.9) 0.004 > 1017 (54.8)0 (0.0)5 (16.1)5 (16.1)4 (12.9)Setting Urban24 (40.0)0 (0.0)25 (41.7)5 (8.3)6 (10.0)0.956 Intermediate/rural21 (44.7)0 (0.0)18 (38.3)4 (8.5)4 (8.5)Region North and Centre19 (39.6)0 (0.0)25 (52.1)2 (4.2)2 (4.2)0.058 South26 (44.1)0 (0.0)18 (30.5)7 (11.9)8 (13.6)PSA based screening for men aged more than 75 yearsGender Female10 (11.5)1 (1.1)34 (39.1)33 (37.9)9 (10.3)0.674 Male1 (5.0)0 (0.0)6 (30.0)11 (55.0)2 (10.0)Years of experience < 108 (10.5)1 (1.3)32 (42.1)28 (36.8)7 (9.2)0.474 > 103 (9.7)0 (0.0)8 (25.8)16 (51.6)4 (12.9)Setting Urban8 (15.4)1 (1.9)17 (32.7)20 (38.5)6 (11.5)0.533 Intermediate/rural3 (6.0)0 (0.0)18 (36.0)24 (48.0)5 (10.0)Region North and Centre3 (6.3)1 (2.1)21 (43.8)20 (41.7)3 (6.3)0.304 South8 (13.6)0 (0.0)19 (32.2)24 (40.7)8 (13.6) Frequency of PSA testing Test types for PCa detection and risk stratification Most PCPs considered mpMRI to be very effective (59.7%) in detecting clinically significant prostate cancer and potentially reducing mortality, followed by PSA testing, which was viewed as somewhat effective by the majority (67.3%) (Fig. 2). Perceptions of the effectiveness of various tests for risk stratification before prostate biopsy are presented in Fig. 3. Fig. 2Beliefs in current tests in reducing prostate cancer mortality. mpMRI, multiparametric Magnetic Resonance Imaging. DRE, Digital rectal exam. PSA, prostate-specific antigen Beliefs in current tests in reducing prostate cancer mortality. mpMRI, multiparametric Magnetic Resonance Imaging. DRE, Digital rectal exam. PSA, prostate-specific antigen Fig. 3Risk stratification. PSA-d, prostate-specific antigen-density. T/F PSA Ratio, total/free PSA ratio. mpMRI, multiparametric Magnetic Resonance Imaging Risk stratification. PSA-d, prostate-specific antigen-density. T/F PSA Ratio, total/free PSA ratio. mpMRI, multiparametric Magnetic Resonance Imaging Influential organisations The most influential guidelines informing respondents’ current clinical practice were the national DGS Guidelines (75.7%), followed by the EAU Guidelines (49.5%), AAFP Guidelines (34.6%), and AUA Guidelines (28%). A significant association was found between years of experience and the influence of the national DGS Guidelines (p = 0.027), with physicians with fewer years of experience more likely to be familiar with and adhere to these national recommendations. Fig. 4Perceived experience with SDM. Each topic reflects the various steps inherent to the SDM model outlined in Fig. 1. The final topic outlines the need for a clinical registry when the discussion about PSA screening takes place, as outlined by the DGS National Guidelines [8]. SDM, shared decision-making. PSA, prostate-specific antigen. DGS, Direção-geral da Saúde Perceived experience with SDM. Each topic reflects the various steps inherent to the SDM model outlined in Fig. 1. The final topic outlines the need for a clinical registry when the discussion about PSA screening takes place, as outlined by the DGS National Guidelines [8]. SDM, shared decision-making. PSA, prostate-specific antigen. DGS, Direção-geral da Saúde Opinion on population-based PSA screening A total of 11.2% of respondents fully agreed with the existence of a population-based prostate cancer (PCa) screening program, while 45.8% somewhat agreed, 36.4% somewhat disagreed, and 6.5% fully disagreed. Support for such screening was significantly associated with gender (p = 0.007), with male physicians more likely to express full agreement. SDM experience When discussing opportunistic PCa screening, most PCPs reported consistently explaining PSA testing (80.4%), its benefits and risks (64.5%), and exploring patient preferences (64.5%). PCPs in the South region were more likely to report providing an explanation of PSA testing (p = 0.012) and discuss benefits/risks (p = 0.002) than those in the North/Centre. A reflection period between consultations was not routinely offered by 40.2%, though it was more common in the South (p = 0.044). Only 37.4% documented these discussions in the clinical registry. Further details are shown in Fig. 4 and Additional File 2: Table A. Most respondents (72%) were unaware of decision aids. While 21.5% were familiar with them and considered them useful, they had not used them in practice; only 6.5% reported actual use. Awareness and use were not significantly associated with gender (p = 0.685), years of practice (p = 0.475), practice setting (p = 0.863), or region (p = 0.207). Referral to a urologist Most PCPs referred patients to a urologist when a prostatic nodule was detected on TRUS (72.0%) or when an abnormal DRE was present (72%). Less commonly cited referral criteria included uncertainty in DRE (46.7%), F/T PSA ratio < 15% (28%), PSA > 3.5 ng/mL (16.8%), and patient request for a second opinion (16.8%) (Fig. 5). PCPs in intermediate settings were more likely to refer based on TRUS findings than those in urban settings (p = 0.02). No other significant differences were found across categories (Additional File 2: Table B). Overall, 53.3% of PCPs considered fewer ([1–2]) referral criteria. Referral decisions based on specific ([1–2]) vs. extended (≥ 3) criteria were not significantly associated with gender (p = 0.745), years of practice (p = 0.136), practice setting (p = 0.494), or region (p = 0.867). Before referral, most PCPs requested F/T PSA ratio (97.2%), TRUS (86%), and blood tests (57.9%), while only 2.8% ordered mpMRI. Fig. 5Criteria for referral to a urologist. The results are organized from the least to the most chosen criteria for referral. PSA, prostate-specific antigen. F/T PSA, free/total PSA ratio. DRE, digital rectal exam. TRUS, transrectal ultrasound Criteria for referral to a urologist. The results are organized from the least to the most chosen criteria for referral. PSA, prostate-specific antigen. F/T PSA, free/total PSA ratio. DRE, digital rectal exam. TRUS, transrectal ultrasound Discussion PCa screening remains highly controversial in urology [4]. A 2013 Cochrane review of five large, randomized trials (over 721,000 men) [5], updated in 2018 [29], found that while screening increases the diagnosis of prostate cancer (IR: 1.23), it does not improve prostate cancer-specific survival (IR: 0.96). Differences between studies—including trial size, age groups, and compliance—contribute to conflicting results and make interpretation challenging. This inconsistency is echoed in guideline recommendations. The AUA and AAFP advise opportunistic screening for men aged 55–69 [7, 30], while the EAU and DGS recommend starting at age 50 [6, 8]. Moreover, only some guidelines consider life expectancy [6, 7, 30], and recommend varying screening intervals (for patients not at risk, the DGS recommends repeating the PSA test every two years, while the EAU recommends it for every eight years). Our results show that PCPs reference all these major guidelines, particularly the national DGS (75.7%) and the EAU Guidelines (49.5%), leading to confusion and potentially significant variation in clinical practice when it comes to PSA screening criteria and frequency. Although most PCPs acknowledged the importance of PSA and other modalities in PCa screening, their practices varied across patient scenarios and were influenced by years of experience. Notably, less experienced physicians in urban areas tended to be more conservative about ordering PSA testing for younger men (40–49 years), and more likely to tailor testing frequency based on initial PSA values—a pattern more aligned with current guidelines [6–8]. Among men > 50 years, PCPs in intermediate (vs. urban) settings showed somewhat greater propensity for PSA screening. For men at higher risk because of family history or BRCA2 mutation, all respondents tended to consider PSA testing at an adequate age. For older men (> 75 years) with 10 years of life expectancy, those with less experience were more likely to recommend screening—perhaps reflecting differences in medical training or in guideline familiarity. Volk et al. (2020) noted that “routine screeners” were more likely to automatically order PSA tests based on professional experience, patient expectation, and habitual practice patterns, while newer physicians were somewhat less likely to screen by default [31, 32]. However, Marcella et al. (2007) demonstrated that during medical training, students generally are very optimistic about the benefits of screening for PCa but with increased knowledge, a more conservative view of screening is adopted [33]. In our study, although less experienced physicians also demonstrated a higher propensity to recommend PSA testing in certain scenarios, their decisions appeared to be more closely aligned with guideline-based risk stratification rather than routine or expectation-driven screening, suggesting a difference in the underlying rationale rather than in overall screening frequency. Support for population-based PSA screening was moderate, with only 11.2% fully agreeing with its implementation—mirroring the ongoing debate in the literature and guideline ambiguity. This aligns with a recent European survey, where most GPs expressed unfavorable opinions about broad PSA screening, with half stating they would not recommend the test to relatives, often citing perceived risks outweighing benefits [17]. Interestingly, male PCPs were more supportive of population-based screening than their female counterparts. Some studies also suggest that physician gender may impact attitudes and practices, with male physicians generally showing slightly more support for, and higher rates of, PSA testing than female physicians [34, 35]. In our study, we chose to evaluate SDM using the specific components outlined by Pekala et al. [24], as their comprehensive definition addresses several key aspects often omitted in broader or less precise frameworks. Few studies assess whether full SDM occurs in practice. Most of the literature defines SDM merely as explaining the PSA test and discussing benefits and risks, often without specifying those pros and cons [36–43]. Rarely is it examined whether patients were given time to reflect between consultations or whether their personal preferences and priorities were actively discussed [44–46]. Our results show that while most PCPs reported consistently explaining PSA testing and exploring patient preferences, only two-thirds discussed potential benefits and harms, and fewer than half provided structured reflection periods between consultations. Documentation of SDM in clinical records was even less frequent (37.4%), even though it is part of DGS audit criteria [8]. Regional differences were apparent: PCPs in Lisbon and the South of Portugal reported more comprehensive communication and greater use of reflection time than those from the North. These statistically significant differences may be partly related to regional variation in sociodemographic profiles and health literacy described in national surveys, which show marked inequalities in education, ageing and health status across Portuguese regions and highlight generally low—but heterogeneous—health literacy levels in the population [47]. Primary care organization and workload may also have a role. These findings are consistent with studies in other countries reporting inconsistent SDM implementation for PCa screening, despite explicit recommendations in clinical guidelines [6, 7, 12], including the national recommendations [8]. For instance, Stevens et al. (2024), in a qualitative analysis of audio-recorded patient-provider interactions, reported that a minority of primary care interactions for PSA screening included discussing both benefits and harms and eliciting patient preferences. Most discussions focus on reasons to screen, with few patient-centered components [22]. A narrative review found that the rate of SDM in primary care for prostate cancer screening varies widely, between 11% and 98% depending on the study and setting, but is much lower (often under 25%) when using strict definitions and patient reports rather than physician self-report [24]. Time constraints, lack of structured tools, and variability in SDM training likely contribute to these gaps [16, 23, 24, 48, 49]. Decision aid use in this sample was very low—only 6.5% had used such tools, though a fifth knew about them and considered them valuable. Yet, over 70% remained unaware of decision aids altogether. This aligns with international studies and systematic reviews of poor integration of decision support tools in clinical practice [13, 16, 49–53], despite evidence that they can improve patient knowledge and satisfaction [13–16]. The AUA Guidelines even mention and recommend their use [7, 12] based in the results of a 2017 Cochrane systematic review and meta-analysis of 105 studies that showed that people who view decision aids feel more knowledgeable, better informed, and clearer about their values [11]. Several studies suggest that, even when clinicians are aware of decision aids, perceived time constraints during consultations, competing priorities, and lack of workflow integration are major barriers to their routine use [16, 41]. Other reported obstacles include limited access to tools in the local language, insufficient training on how to incorporate them into clinical encounters, and uncertainty about their added value compared with usual counselling [25, 54]. These factors are likely to contribute to the very low uptake observed in our sample, even among PCPs who recognize the potential benefits of decision aids. Finally, criteria used by the respondents for referral to a urologist vary considerably. One notable finding is that the most cited independent criteria was the detection of a prostatic nodule in TRUS. TRUS is not reliable at detecting PCa [55]. Specificity of suspicious nodules on TRUS for prostate cancer is relatively low. While one study reported a specificity of 88% for suspicious nodules, the sensitivity was only 31%, meaning many cancers are missed, and not all detected nodules are malignant [56]. Other research highlights that the accuracy of greyscale TRUS for detecting prostate cancer is low, as benign and malignant lesions can have overlapping sonographic features [57]. Prostatic nodules and hypoechoic lesions seen on TRUS may represent non-cancerous conditions such as benign prostatic hyperplasia [58], and the diagnostic yield of additional biopsies performed on hypoechoic lesions is negligible [59]. Knowledge gaps regarding the criteria for urology referral underscore the importance of targeted educational initiatives to improve PCa screening practices among Portuguese PCPs. Our findings indicate clear opportunities for improvement in the alignment of PCa screening practices with evidence-based guidelines and patient preferences. Given the convenience sample, unknown response rate, and over-representation of younger and female PCPs, these findings should be regarded as exploratory and hypothesis-generating, warranting confirmation in larger and more representative studies. The limited adoption of SDM and decision aids points to the need for their dissemination in Portuguese primary care, particularly given that a culturally adapted decision aid already exists [25]. Greater dissemination of updated guidelines and education on the value of individualized risk-based screening is also warranted. Future research should explore the underlying reasons for practice variation (including barriers and facilitators to SDM), evaluate strategies to improve uptake of decision aids, and examine the impact of enhanced SDM on patient outcomes. Limitations This study is not without limitations. First, its cross-sectional design limits any inference about causality or temporal trends. Second, because it was not possible to determine how many PCPs accessed the survey link, we could not calculate an accurate response rate, raising the possibility of response bias. The convenience sampling method and recruitment of self-selected participants from responding HCs may affect the generalizability of findings to all Portuguese PCPs, especially given the predominance of younger and female respondents in our sample. Finally, responses were self-reported, which introduces risks of recall bias and social desirability bias—particularly for practices such as SDM and guideline adherence. Future research using longitudinal study designs and randomized sampling could help address these limitations and provide deeper insight into evolving PSA screening practices. Conclusions PCa screening practices among Portuguese PCPs remain a complex area of primary care, characterized by variability in clinical approaches, inconsistent guideline adherence, and limited application of SDM. Despite general awareness, notable gaps persist in the use of decision aids and evidence-based referral criteria. These findings underscore the need for clearer national guidelines and enhanced support for individualized, patient-centered screening. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1