Radiol Oncol 2024; 58(4): 544-555. doi: 10.2478/raon-2024-0046 544 research article Analysis of early diagnostic pathway for prostate cancer in Slovenia Mateja Kokalj Kokot1,2, Spela Mirosevic1, Nika Bric3, Davorina Petek1,4 1 Department of Family Medicine, Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia 2 Primary Healthcare Centre Grosuplje, Grosuplje, Slovenia 3 Sector for Oncology Epidemiology and Cancer Registry, Institute of Oncology Ljubljana, Ljubljana, Slovenia 4 Medical Centre Zdravje, Ljubljana, Slovenia Radiol Oncol 2024; 58(4): 544-555. Received 28 March 2024 Accepted 25 July 2024 Correspondence to: Assist. Mateja Kokalj Kokot, M.D., Department of Family Medicine, Faculty of Medicine, University of Ljubljana, Poljanski nasip 58, SI-1000 Ljubljana, Slovenia. E-mail: mateja.kokaljkokot@mf.uni-lj.si Disclosure: No potential conflicts of interest were disclosed. This is an open access article distributed under the terms of the CC-BY license (https://creativecommons.org/licenses/by/4.0/). Background. Prostate cancer (PCa) is a prevalent male malignancy globally. Prolonged diagnostic intervals are associated with poorer outcomes, emphasizing the need to optimize this process. This study aimed to evaluate the doctor and primary care interval, research their impact on patient survival and explore opportunities to improve PCa diagnostic pathway in primary care. Patients and methods. A retrospective cohort study using cancer patients’ anonymised primary care data and data of the Slovenian Cancer Registry. Results. The study found that the doctor interval had a median duration of 0 days (interquartile range ([IQR] 0−6) and primary care interval a median duration of 5 days (IQR 0−58). Longer intervals were observed in patients with more than two comorbidities, where general practitioners didn’t have access to laboratory diagnostic tests within their primary health care centre and when patients first presented with symptoms (reported symptoms at first presen- tation: dysuria, lower urinary tract symptoms [LUTS], abdominal pain). The analysis also revealed a statistically signifi- cant association between lower 5-year survival rate and the accessibility of laboratory and ultrasound diagnostics in primary healthcare centres and a shorter 5-year survival of symptomatic patients in comparison to patients who were identified by elevated levels of prostate specific antigen (PSA). Conclusions. This study shows that treating suspected PCa in primary care has a significant impact on 5-year survival. Several factors contribute to better survival, including easy access to laboratory and abdominal ultrasound in primary care centres. The study highlights the complex array of factors shaping PCa diagnosis, beyond individual clinicians’ skills, encompassing test and service availability. Key words: prostate cancer; doctor interval; primary care interval Introduction Prostate cancer is a common malignant tumour and has the highest incidence of all non-cutaneous cancers worldwide in males. It is also the fifth lead- ing cause of cancer death among men in 2020.1 In Slovenia the 2020 age-standardized (World standard population) incidence rate was 62.7.2 In Central and Eastern Europe it was 46.4 and 59.1 in Southern Europe.1 The ageing population un- doubtedly influences the increasing incidence of prostate cancer. However, the proportional growth of elderly individuals (aged 65 years and older) in Slovenia has not escalated sufficiently to solely ac- count for this rise in prostate cancer incidence. For example, in 1991, the proportion of elderly people in Slovenia was 11.2% and the crude incidence rate for prostate cancer was 26.3 per 100,000 individu- Radiol Oncol 2024; 58(4): 544-555. Kokalj Kokot M et al. / Analysis of early diagnostic pathway for prostate cancer in Slovenia 545 als. In 2001, the proportion of elderly increased to 14.3% and the crude incidence rate for prostate cancer was 74.5.3,4 The dramatic increase in the inci- dence of prostate cancer over the past two decades is not due to any newly identified risk factor, but rather to the increasing use of the prostate-specific antigen (PSA) test in healthy men and therefore detection of a large number of cancers that would have otherwise remained undetected for life. Data for recent years indicate that we have already reached the peak incidence of prostate cancer.2 The histological incidence far surpasses the proportion of individuals in whom the disease is manifested, and while there has always been a tendency not to misdiagnose those with prostate cancer, there is increasing focus on identifying patients who are treated by watchful waiting. Prostate cancer is characterised by a slow natural course of the dis- ease, with the majority of patients dying from oth- er, non-cancer related causes.5,6 Both the incidence and mortality rates of prostate cancer in Slovenia are above the European average. In 2020, Slovenia reported an age-standardized (World standard population) mortality rate of 14.9, while Central and Eastern Europe recorded a rate of 13.7, the highest among all European regions.1,2 Much effort has already been applied to opti- mize early detection in prostate cancer and some medical professional societies and organizations recommend prostate cancer screening or dis- cussing screenings with men of suitable age and life expectancy.7 Many countries in Europe have joined PRAISE-U project to establish screening programme for prostate cancer.8 In primary care- based health care systems, in which the general practitioner (GP) is the patient’s first contact and triages the patient’s further access to the system, most prostate cancer patients either present to a GP with symptoms (dysuria, lower urinary tract symptoms (LUTS), abdominal pain) or an elevated level of PSA is found in asymptomatic patients. Therefore, timely recognition of cancer-related complaints and adequate referral by the GP are and will remain essential to reduce time to diag- nosis until successful screening programmes are introduced. Even though the association between time intervals in the diagnostic pathway and clini- cal outcomes is complex and remains debated, evidence suggests an association between shorter times to diagnosis and more favourable outcomes in breast cancer, colorectal, head and neck, tes- ticular cancer and melanoma.9,10 Optimising the diagnostic pathway from first presentation to di- agnosis and start of treatment, usually interpreted as shortening the diagnostic phase, has therefore been a main objective of health care organisations involved in cancer care worldwide. For some coun- tries in Europe, the duration of several of these intervals has been charted.11-17 For other countries, such as Slovenia, the duration of these intervals is unknown. International comparison of the dura- tion of diagnostic intervals in different health care systems and cultural environments is important to identify system-, disease- and patient- related fac- tors that contribute to an unnecessarily prolonged patient journey. Exploring the duration of the di- agnostic pathway in Slovenia and how primary care contributes to it generates relevant informa- tion on international differences in the duration of the diagnostic pathway. This provides the op- portunity to distinguish underlying mechanisms of delay, including system-, disease- and patient- related delay. The aims of the study were (i) to evaluate the duration and specifics of management during the doctor and primary care intervals in Slovenia, (ii) to investigate the potential association between the durations of the intervals, sociodemographic and organizational variables and (iii) to assess the association between these variables, intervals and the 5-year survival rates of prostate cancer patients in Slovenia. Patients and methods Study design, data source and patient selection A retrospective cohort study was performed us- ing data from the Slovenian Cancer Registry and primary care data collected from selected GPs and family doctors with whom these patients were reg- istered at the time of the study. Cancer Registry of Republic of Slovenia provides reliable and detailed information on Slovene cancer patients since 1950. This study was part of the research project of the Institute of Oncology Ljubljana, in collaboration with the Department of Family Medicine, Faculty of Medicine, University of Ljubljana, and the Clinical Department of Urology, University Medical Centre Ljubljana, entitled Integrated analysis of the early management of patients with urological cancers, assessing delays in referral, diagnosis and first treatment. The study was reviewed and approved by the Commission of the Republic of Slovenia for Medical Ethics (0120-233/2019/4). We included all newly diagnosed prostate can- cer patients during the year 2014 in the Slovenian Radiol Oncol 2024; 58(4): 544-555. Kokalj Kokot M et al. / Analysis of early diagnostic pathway for prostate cancer in Slovenia546 Cancer Registry database. Consecutively, 1431 pa- tients with all stages of prostate cancer were in- cluded. We then linked the Cancer Registry data to the National Insurance Company Registry in order to identify each patient’s GP and send them a questionnaire. The exclusion criterion was if the diagnosis of prostate cancer was made at the au- topsy (29 patients). A flowchart of patients’ data inclusion in our study is presented in Figure 1. We asked the GPs to collect information from the pa- tient’s records and complete the attached question- naire on management of the initial symptoms of the disease, the diagnostics performed and referral decisions, using the dates and information given in the patients’ medical record and discharge let- ters from hospitals. We also asked them about the accessibility to diagnostics (laboratory, abdominal ultrasound) in their primary healthcare centre. Non-responders received a reminder after eight months. Data collection Doctor interval and primary care interval The intervals were defined according to the Aarhus statement.18,19 Doctor interval was defined as the period of time from first consultation in primary care to beginning of the first investigation in pri- mary care. Date of first presentation was defined as the first contact with the GP (in person or telemedi- cine) with possible prostate cancer related signs and/or symptoms. The questionnaire allowed the GP to choose from six different reasons why the patient chose to see a doctor, with multiple choices also possible. Possible symptoms and signs report- ed by the patient at first presentation were dysu- ria, LUTS and lower abdominal pain. Additional reasons for visiting the doctor included a family history of cancer, an elevated PSA level detected during routine check-ups (preventive examination, opportunistic screening, regular annual follow-up for chronic diseases, occupational health and safe- ty check-up), and doctors could also choose other. Those were the cases where prostate cancer was not first suspected in the GP’s practice but else- where, e.g. during hospitalisation for another ill- ness, as reported by the GPs in the questionnaire notes. GPs themselves set the most accurate date estimate for the first presentation after reviewing the patient’s records. They also noted if and when they performed a digital rectal exam, referred the patient to the abdo- men ultrasound and/or checked the patient blood (complete blood count [CBC] and/or C-reactive protein [CRP] and/or PSA) and/or urine. All of the stated actions counted as first investigation in pri- mary care. We calculated the length of the doctor interval based on the date of the first presentation and the date of the earliest diagnostic procedure. Primary care interval was defined as the period of time from first presentation in primary care to referral to urologist. Date of referral was defined as the day the referral letter to the urologist was written, which was also stated by the GP. If the date of first presentation was the same or later than the date of the earliest diagnostic proce- dure, the duration of doctor interval was coded as 0 days. Similarly, if the date of first presentation was the same as the date of the referral to urolo- gist, the duration of primary care interval was cod- ed as 0 days. If the length of any interval was equal to or more than one year, it was coded as 365 days. Characteristics The decision to collect data for certain characteris- tics and to include them in our analyses was based on previously reported diagnostic procedures and predictors in the literature20–25, on clinical relevance of patients and disease characteristics, and on availability of data in the Cancer registry and pri- mary care data. The questionnaire was approved by the entire research team. We collected data on age, level of education, stage of cancer, comorbidi- ties, symptoms and signs at first presentation, vital status 5 years after diagnosis and location of their GP’s primary health care centre. The Slovenian Cancer Registry Database uses a simplified definition of stages at diagnosis for sol- id tumours, classifying them into localized, spread (regional) and metastatic stage of disease. The sim- plified stage definition generally follows the TNM classification. Localized stage includes all cancers where the tumour has been classified as T1 and T2. In these cases, neither regional lymph node in- volvement nor distant metastases are found (N0, M0). The spread stage includes tumours classified as T3 and T4 and/or with regional lymph node metastases (N1), without presence of metastases in distant lymph nodes or organs (M0). A disease with metastases in distant lymph nodes or organs is classified as a metastatic stage (M1).2 Statistical analysis The data on the patients’ demographics, cancer stage, symptoms at first presentation, comorbidi- Radiol Oncol 2024; 58(4): 544-555. Kokalj Kokot M et al. / Analysis of early diagnostic pathway for prostate cancer in Slovenia 547 ties, diagnostics executed at the primary health care centre, region of GP’s practice and accessibil- ity to diagnostic tests reported in questionnaires were collected and managed in Microsoft Access Database 2007-2016 Version. Additional data pro- cessing was performed in Excel (Microsoft Office Professional Plus 2019). Descriptive statistics, in- cluding means, median, standard deviations and frequencies, were calculated to provide an over- view of the data distribution. To examine the rela- tionships between variables, we used Independent Sample T-test and chi-square tests for categorical variables. In case of the presence of variables with non-normal distributions, we also used the Mann- Whitney U-test. To identify potential predictors for the outcome variable, we performed a multiple logistic regression analysis that included the rele- vant sociodemographic, clinical and organizational variables. To account for missing data that occurred ran- domly and without a recognizable pattern, we conducted multiple imputation and analysis using Version 29.0 of the IBM SPSS Statistics. This tech- nique allows the estimation of missing values by creating multiple plausible imputation datasets, thereby maintaining statistical power, and mini- mizing bias. All statistical analyses were performed using IBM SPSS Statistics with the significance level set at p < 0.05. Results Patients’ characteristics Among the cohort of 1431 patients registered with a diagnosis of prostate cancer in the Cancer Registry in 2014, we successfully obtained primary care data for a total of 814 patients, as illustrated in Figure 1. The included patients’ average age was 69 years, with a standard deviation of 8.5 years. Detailed so- ciodemographic and clinical characteristics of the prostate cancer patients included in this study are presented in Table 1. Doctor and primary care intervals’ characteristics The median duration of the doctor interval, in terms of days, was found to be 0, with an interquar- tile range spanning from 0 to 6 days. The median duration of the primary care interval was 5 days, with an interquartile range extending from 0 to 58 days. The average number of visits to the doctor during the primary care interval was 1.68 (SD 0.9), range 1−5. Table 2 provides also the mean of the in- tervals. For the analysis of the doctor and primary care intervals, we excluded the Zasavska region due to a limited sample size, with only three patients included in the study, which did not provide suf- ficient statistical power for meaningful analysis. Consequently, our analysis was based on a dataset consisting of 811 patients. Table 3 shows the diag- nostic procedures performed by the GP and their access to laboratory and abdominal ultrasound within the primary health care centre, overall and by region. Table 4 presents the median and interquartile range data for the duration of the doctor and pri- mary care intervals, computed based on patient and presentation characteristics. Statistically significant longer doctor intervals were observed in patients who had localized vs. metastatic cancer stage, presence of more than two comorbidities, where GPs’ didn’t have access to laboratory diagnostic tests (CBC, CRP, urine, PSA) within their primary health care centre and when patients first presented with symptoms (reported symptoms at first presentation: dysuria, LUTS, ab- dominal pain). The primary care interval was also statistically significantly longer in patients who had more than two comorbidities, where GPs’ didn’t have access to laboratory tests within their primary health care centre and when patients first presented with symptoms. In addition, there was a marked differ- ence between the Osrednjeslovenska and Savinjska regions, with the former having the longest and the latter the shortest primary care interval. To further elucidate the relationship between the predictor variables and the primary care in- terval, we employed a logistic regression model (Table 5). We set the limit at a primary level inter- val duration of 14 days. Omnibus tests of the model coefficients yielded a chi-square statistic of 53.642 at 11 degrees of freedom, resulting in a p-value of less than 0.001. This indicates robust overall sig- nificance of the prediction model, confirming that the independent variables significantly contrib- ute to the variability in the primary care interval. Nevertheless, the Nagelkerke R² coefficient indi- cates that only 9.4% of the variance of the depend- ent variable is explained by the model, reflecting the complexity of health service utilization behav- iour. The model’s -2 log-likelihood of 1018.153 indi- cates a satisfactory fit to the empirical data. Radiol Oncol 2024; 58(4): 544-555. Kokalj Kokot M et al. / Analysis of early diagnostic pathway for prostate cancer in Slovenia548 5-year survival analysis In the group of 814 prostate cancer patients, 610 (74.9%) were alive five years after diagnosis. Our analysis showed statistically significant differences in 5-year survival depending on the accessibility of laboratory and ultrasound diag- nostics in primary health care centres (Table 6). In addition, we observed lower 5-year survival in symptomatic patients. There were statistically significant differences between the duration of primary care interval and 5-year survival (204 dead patients, mean primary care interval 90.18 and 610 alive patients with mean interval 59.79, p-value = 0.007). Discussion This study provides an overview of early man- agement of patients with prostate cancer in fam- ily medicine in Slovenia. Our aim was to evaluate the treatment timelines and specificities within FIGURE 1. Flowchart of patient selection. GP = general practitioner TABLE 1. Sociodemographic and clinical characteristics of prostate cancer patients (N = 814) Characteristics Sample of prostate cancer patients, n (%) Age (mean ± SD), range: 44−97 69.0 ± 8.5 Age, groups < 65 65−75 > 75 265 (32.5) 357 (43.9) 192 (23.6) Cancer stage Localized Spread Metastatic 572 (70.3) 191 (23.4) 51 (6.3) Education Primary/Elementary education Secondary/High school education Higher education 280 (34.4) 386 (47.4) 148 (18.2) Comorbidities No. of comorbidities (mean ± SD), range: 0−6 None 1−2 > 2 1.2 ± 1.2 290 (35.6) 410 (50.4) 114 (14) Number of patients by region of the primary health care centre’s location Pomurska Podravska Koroška Savinjska Zasavska Spodnjeposavska JV Slovenija Osrednjeslovenska Gorenjska Notranjsko-kraška Goriška Obalno-kraška 106 (13) 114 (14) 31 (3.8) 85 (10.4) 3 (0.4) 25 (3.1) 53 (6.5) 204 (25.1) 69 (8.5) 27 (3.3) 56 (6.9) 41 (5.0) Symptoms and signs at first presentation (multiple options possible) Dysuria LUTS Pain in the abdomen Family history of cancer Elevated PSA Other 111 (13.6) 307 (37.7) 63 (7.7) 17 (2.1) 246 (30.2) 130 (16) Alive 5-years after diagnosis 610 (74.9) JV Slovenija = South-East Slovenia; Koroška = Carinthia; LUTS = lower urinary tract symptoms; Obalno-kraška = Coastal-Karst; Osrednjeslovenska = Central Slovenia; PSA = prostate specific antigen TABLE 2. Duration of doctor and primary care interval mean (± SD) range Doctor interval, days 37.5 (92.8) 0−365 Primary care interval, days 67.4 (123.8) 0−365 Radiol Oncol 2024; 58(4): 544-555. Kokalj Kokot M et al. / Analysis of early diagnostic pathway for prostate cancer in Slovenia 549 TABLE 3. Diagnostic procedures performed by the general practitioner (GP) and their access to laboratory and abdominal ultrasound within the primary health care centre, overall and by region (N = 811) Diagnostic procedures performed by GPs Sample of prostate cancer patients, n (%) CBC, CRP 497 (61.1) PSA 664 (81.6) Urine 471 (57.9) US 434 (53.3) DRE 255 (31.3) Diagnostic procedures performed by GPs by region CBC, CRP PSA urine US DRE Pomurska (N = 106) 48 (45.3) 77 (72.6) 52 (49.1) 39 (36.8) 34 (32.1) Podravska (N = 114) 77 (67.5) 95 (83.3) 60 (52.6) 68 (59.6) 23 (20.2) Koroška (N = 31) 15 (48.4) 24 (77.4) 13 (41.9) 4 (12.9) 11 (35.5) Savinjska (N = 85) 55 (64.7) 64 (75.3) 61 (71.8) 50 (58.8) 26 (30.6) Spodnjeposavska (N = 25) 14 (56) 20 (80) 16 (64) 11 (44) 13 (52) JV Slovenija (N = 53) 35 (66) 39 (73.6) 25 (47.2) 19 (35.8) 11 (20.8) Osrednjeslovenska (N = 204) 135 (66.2) 180 (88.2) 131 (64.2) 140 (68.6) 46 (22.5) Gorenjska (N = 69) 48 (69.6) 63 (91.3) 44 (63.8) 39 (56.5) 22 (31.9) Notranjsko-kraška (N = 27) 21 (77.8) 23 (85.2) 19 (70.4) 20 (74.1) 15 (55.6) Goriška (N = 56) 28 (50) 45 (80.4) 29 (51.8) 22 (39.3) 33 (58.9) Obalno-kraška (N = 41) 19 (46.3) 32 (78) 19 (46.3) 20 (48.8) 19 (46.3) Accessibility to diagnostic tests in the primary health care centre Sample of prostate cancer patients, n (%) Laboratory (CBC, CRP, urine) 734 (90.2) Laboratory (PSA) 653 (80.2) US 306 (37.6) Accessibility to diagnostic tests in the primary health care centre by region CBC, CRP, urine PSA US Pomurska (N = 106) 96 (90.6) 79 (74.5) 34 (32.1) Podravska (N = 114) 98 (86) 89 (78.1) 42 (36.8) Koroška (N = 31) 28 (90.3) 26 (83.9) 22 (71) Savinjska (N = 85) 78 (91.8) 63 (74.1) 36 (42.4) Spodnjeposavska (N = 25) 23 (92) 16 (64) 10 (40) JV Slovenija (N = 53) 46 (86.8) 42 (79.2) 15 (28.3) Osrednjeslovenska (N = 204) 187 (91.7) 173 (84.8) 74 (36.3) Gorenjska (N = 69) 63 (91.3) 60 (87) 31 (44.9) Notranjsko-kraška (N = 27) 24 (88.9) 20 (74.1) 7 (25.9) Goriška (N = 56) 53 (94.6) 52 (92.9) 23 (41.1) Obalno-kraška (N = 41) 36 (87.8) 31 (75.6) 12 (29.3) CBC = complete blood count; CRP = C-reactive protein; DRE = digital rectal exam; GP = general practitioner; JV Slovenija = South-East Slovenia; Koroška = Carinthia; Obalno-kraška = Coastal-Karst; Osrednjeslovenska = Central Slovenia PSA = prostate specific antigen; US = abdominal ultrasound the doctor and primary care interval. The median length of the doctor interval was very short at 0 days, the primary care interval exhibited a me- dian duration of 5 days. Our study demonstrated statistically significant correlation of cancer stage, comorbidities, and the accessibility of laboratory tests within primary health care centres and sur- vival of prostate cancer patients. The presence of laboratory tests within primary health care centres emerged as a significant determinant of the pri- mary care interval’s duration and 5-year survival. The exclusion of 617 patients’ of 1431 patients total from our analysis was necessitated by vari- ous reasons outlined in Figure 1. It is worth not- Radiol Oncol 2024; 58(4): 544-555. Kokalj Kokot M et al. / Analysis of early diagnostic pathway for prostate cancer in Slovenia550 TABLE 4. The distribution and association of selected characteristics with doctor and primary care interval. (N = 811) N % Doctor Interval Primary care interval Median Interquartile range P Median Interquartile range P Age groups 0.131 0.382 < 65 264 32.55 0 0–4 5 0–65 65–75 355 43.77 0 0–5 4 0–51 > 75 192 23.67 0 0–20 7 0–73 Education 0.437 0.204 Primary/Elementary education 278 34.28 0 0–6 7 0–38 Secondary/High school education 385 47.47 0 0–5 3 0–53 Higher education 148 18.25 0 0–18 7 0–127 Region of the location of primary health care centre 0.863 0.021 Pomurska 106 13.07 0 0–18 3 0–62 Podravska 114 14.06 0 0–20 8 0–50 Koroška 31 3.82 0 0–20 0 0–42 Savinjska 85 10.48 0 0–3 0 0–28 Spodnjeposavska 25 3.08 0 0–10 3 0–91 JV Slovenija 53 6.54 0 0–7 1 0–62 Osrednjeslovenska 204 25.15 0 0–5 10 0–88 Gorenjska 69 8.51 0 0–2 4 0–54 Notranjo-kraška 27 3.33 0 0–7 6 0–127 Goriška 56 6.90 0 0–5 4 0–34 Obalno–kraška 41 5.06 0 0–20 4 0–48 Cancer stage 0.037 0.058 Localized 569 70.16 0 0–8 6 0–65 Spread 191 23.55 0 0–1 1 0–28 Metastatic 51 6.29 0 0–38 3 0–336 Comorbidities 0.001 0.026 None 288 35.51 0 0–31 7 0–101 1–2 409 50.43 0 0–4 5 5–44 > 2 114 14.06 0 0–1 0 0–33 GP access to laboratory (CBC,CRP,urine) < 0.001 < 0.001 Yes 732 90.26 0 0–4 3 0–37 No 79 9.74 40 0–330 365 0–365 GP access to laboratory (PSA) < 0.001 < 0.001 Yes 651 80.27 0 0–2 3 0–37 No 160 19.73 0 0–177 21 0–365 GP access to US 0.781 0.124 Yes 306 37.73 0 0–5 4 0–31 No 505 62.27 0 0–7 5 0–78 Symptomatic patient < 0.001 0.001 Yes 399 49.20 0 0–12 5 0–42 No 219 27.00 0 0–0 0 0–18 CBC = complete blood count; CRP = C-reactive protein; DRE = digital rectal exam; GP = general practitioner; JV Slovenija = South-East Slovenia; Koroška = Carinthia; Obalno-kraška = Coastal-Karst; Osrednjeslovenska = Central Slovenia; PSA = prostate specific antigen; US = abdominal ultrasound Radiol Oncol 2024; 58(4): 544-555. Kokalj Kokot M et al. / Analysis of early diagnostic pathway for prostate cancer in Slovenia 551 than for routine screening of asymptomatic in- dividuals. Interestingly, we found that patients with stage spread cancer had the shortest doctor interval, while patients with metastatic cancer had the longest. This may be due to non-specific or absent symptoms and signs in localised pros- tate cancer. Symptoms and signs become more fre- quent and pronounced in spread stage of cancer and those caused by metastatic cancer could also be wrongly attributed to other patient’s comor- bidities. Considering health system factors, inter- national comparisons suggests that the problem of early cancer diagnosis is ubiquitous across con- temporary health systems, including high-income countries, though the same underlying problem is manifested differently depending on health service organisation, healthcare professional cul- tures, and the public understanding of cancer.10 This is the first study using Slovenian data to show a correlation between not having easy access to laboratory diagnostics (CBC, CRP, urine, PSA) TABLE 5. Logistic regression model on predicting primary care interval Variables in the Equation B Wald Exp(B) (95% CI) Age -0.004 0.144 0.996 (0.979, 1.015) Education primary vs secondary -0.359 4.223 0.698 (0.496, 0.984)* Education primary vs higher 0.145 0.452 1.156 (0.757, 1.765) Cancer stage localized vs spread -0.336 3.315 0.714 (0.497, 1.026) Cancer stage localized vs metastatic -0.316 0.955 0.729 (0.387, 1.374) No comorbidities -0.070 1.051 0.932 (0.815, 1.066) First symptom - dysuria -0.175 0.592 0.840 (0.538, 1.310) First symptom – LUTS -0.084 0.228 0.919 (0.650, 1.300) First symptom – abdominal pain 0.132 0.222 1.141 (0.659, 1.977) First symptom – family history of cancer 0.257 0.249 1.293 (0.471, 0.913) First symptom – elevated PSA -0.486 5.804 0.615 (0.414, 0.913)* Accessible US diagnostics -0.160 0.940 0.852 (0.616, 1.178) Accessible laboratory diagnostics (PSA) -0.196 0.660 0.822 (0.512, 1.319) Accessible laboratory diagnostics (CBC,CRP,urine) -1.139 11.375 0.320 (0.165, 0.620)*** Constant 1.483 4.119 4.405 * CBC = complete blood count; CRP = C-reactive protein; LUTS = lower urinary tract symptoms; PSA = prostate specific antigen; US = abdominal ultrasound *p < 0.05, **p < 0.01, ***p < 0.001 ing that the proportion of cases excluded due to missing interval data was similar in our study to previous studies.11 Evidence about the length of the doctor inter- val in patients with different cancers is sparse. We found a study of Denmark’s cancer patients26, where median duration of doctor interval for pros- tate cancer patients was 0 days and interquartile range (IQR) 0−6 days, which is the same as in our study. One of the reasons for a short doctor inter- val is that in primary-level diagnostic procedures, doctors sometimes use laboratory and ultrasound tests already performed prior to this first visit (e.g. blood and urine taken from the patient a month earlier, or an abdominal ultrasound performed three months earlier - doctor’s interval is negative, marked as 0 days in the analysis). In this way they omit unnecessary duplication of tests and expedite further management of suspected prostate cancer. There were no statistical differences in the du- ration of doctor interval between different age groups, levels of education, region of primary health care centre and accessibility to abdominal ultrasound in the primary health care centre. The duration was statistically significant longer when patients had no comorbidities, when they first pre- sented with symptoms in comparison to elevated PSA value and where there was not access to labo- ratory diagnostics (CBC, CRP, urine, PSA) in pri- mary health care centre. We interpret these results by assuming that multimorbid patients see their doctor more often, for different reasons, and thus are more likely to report or be asked about differ- ent symptoms and signs, as opposed to patients without comorbidities. In patients with elevated PSA levels in the blood, the guidelines recommend that prostate cancer should be excluded, so the de- cision to refer to a urologist was easy and quick. Dysuria, LUTS and abdominal pain are typical symptoms of a number of different diseases and are not in themselves of great predictive value for a cancer diagnosis.27–29 Men experiencing urinary problems were more inclined to seek medical at- tention compared to asymptomatic men, leading to more frequent PSA testing and consequently earlier detection of prostate cancer. Our study fo- cused on diagnosing prostate cancer cases in 2014 rather than estimating the overall prevalence that year. Therefore, it’s probable that symptomatic cases outnumber asymptomatic ones. In Slovenian primary health care centers, PSA testing is more commonly conducted for symptomatic men or as part of annual check-ups for those undergoing therapy for benign prostatic hyperplasia, rather Radiol Oncol 2024; 58(4): 544-555. Kokalj Kokot M et al. / Analysis of early diagnostic pathway for prostate cancer in Slovenia552 in primary health care centres, longer duration of doctor and primary care interval and worse 5-year survival. Considering the impact of the length of the intervals on 5-year survival, we assume that the longer intervals are indicative of the broader health care situation (the influence of the patient, the doctor and the local health care system charac- teristics), which as a whole influenced the survival of this patient. The duration of the primary care interval has been investigated in several studies.11,13,14,17,30,31 The median duration of the primary care interval in our study was 5 days, with an interquartile range (IQR) extending from 5 to 58 days which is shorter than in Helsper et al study, where the median was 14 days (IQR 3-153).15 Lyratzopoulos et al11 reported the mean length of primary care interval in pros- tate cancer patients 31 days (25th Centile 2 days, 90th Centile 74 days) whereas in our study the mean was 67.4 days with SD 123.8. The duration was, similar to doctor interval, statistically significant longer when patients had no comorbidities, when they first presented with symptoms in comparison to elevated PSA value and where there was no direct access to laboratory diagnostics (CBC, CRP, urine, PSA) in primary health care centres. Surprisingly, we observed a significant difference between the Osrednjeslovenska and Savinjska regions, with the latter showing a shorter primary care interval. This was the only parameter that showed differ- ence in the Slovenian regions. Osrednjeslovenska region is the most densely populated, the largest in terms of number of inhabitants and the second largest in terms of area. In 2014 the Savinjska re- gion had less than half as many inhabitants as the Osrednjeslovenska region and had the lowest pro- portion of the population aged over 80 years in the country (4.3%). Osrednjeslovenska region had the highest gross domestic product (GDP) per capita in the country (25.329 €/resident), 11.7% registered unemployment rate and 11.4% of people at risk of poverty. Savinjska region had a regional GDP of 16.455 €/resident, 13.9% registered unemployment rate and 15% of people at risk of poverty.32 Lower socio-economic factors are usually associated with longer intervals.12,33,34 In 2014 Savinjska and Osrednjeslovenska region had the same age-stand- ardized prostate cancer incidence rate (99.2 vs. 100.1). Savinjska region had a higher standardized death rate (574.8 vs. 463.0) and age-standardized death rate due to neoplasms (231.4 vs. 189.4) than Osrednjeslovenska region.35 Taking all these data into account, it is difficult to explain the shorter primary care interval in the Savinjska region. TABLE 6. 5-year survival in relation to sociodemographic and organizational variables Variables Dead n (%) Alive n (%) P value Age groups < 0.001 < 65 32 (12) 233 (88) 65−75 72 (20.2) 285 (79.8) > 75 100 (52) 92 (48) Education < 0.001 Primary/Elementary education 91 (32.5) 189 (67.5) Secondary/High school education 84 (21.8) 302 (78.2) Higher education 29 (19.6) 119 (80.4) Region of the location of primary health care centre NS Pomurska 32 (30.2) 74 (69.8) Podravska 34 (29.8) 80 (70.2) Koroška 9 (29) 22 (71) Savinjska 18 (21.2) 67 (78.8) Spodnjeposavska 5 (20) 20 (80) JV Slovenija 15 (28.3) 38 (71.7) Osrednjeslovenska 44 (21.6) 160 (78.4) Gorenjska 17 (24.6) 52 (75.4) Notranjo-kraška 6 (22.2) 21 (77.8) Goriška 14 (25) 42 (75) Obalno-kraška 10 (24.4) 31 (75.6) Cancer stage < 0.001 Localized 124 (21.7) 448 (78.3) Spread 36 (18.8) 155 (81.2) Metastatic 44 (86.3) 7 (13.7) Comorbidities < 0.001 None 72 (24.8) 218 (75.2) 1−2 86 (21) 324 (79) >2 46 (40.4) 68 (59.6) GP access to laboratory (CBC,CRP,urine) < 0.001 Yes 159 (21.7) 575 (78.3) No 45 (56.2) 35 (43.8) GP access to laboratory (PSA) < 0.001 Yes 136 (20.8) 517 (79.2) No 68 (42.2) 93 (57.8) GP access to US < 0.05 Yes 62 (20.3) 244 (79.7) No 142 (28) 366 (72) Symptomatic patient < 0.001 Yes 103 (25.8) 296 (74.2) No 29 (13.2) 190 (86.8) CBC = complete blood count; CRP = C-reactive protein; DRE = digital rectal exam; GP = general practitioner; JV Slovenija = South-East Slovenia; Koroška = Carinthia; NS = not statistically significant; Obalno-kraška = Coastal-Karst; Osrednjeslovenska = Central Slovenia; PSA = prostate specific antigen; US = abdominal ultrasound Radiol Oncol 2024; 58(4): 544-555. Kokalj Kokot M et al. / Analysis of early diagnostic pathway for prostate cancer in Slovenia 553 In this study we also aimed to explore poten- tial associations between the duration of primary care interval and the 5-year survival rates among prostate cancer patients in Slovenia. As expected we found worse 5-year survival in older patients, with higher cancer stages, lower education and more comorbidities.25 Survival was not influenced by the region where the GP worked. We found a statistically significant difference in 5-year sur- vival in patients with lower access to laboratory tests and abdominal ultrasound by their GP, and in symptomatic patients compared to those with a first contact due to a detected elevated PSA level. Other research has also confirmed that the quality and speed of the diagnostic process is influenced by a number of factors beyond the diagnostic skills of individual clinicians, such as the tests and ser- vices available to them, time constraints to consul- tation duration and the quality of doctor-patient communication.10 The use of the Slovenian Cancer Registry da- tabase enabled a comprehensive analysis as a substantial number of patients diagnosed with prostate cancer in 2014 were included. This large cohort provides robust statistical power and im- proves the generalizability of the study results. By using registry information, all patients diagnosed with prostate cancer in 2014 were identified and included in the study, minimizing potential selec- tion bias and ensuring a representative sample. The limitation was the 58% response rate. As part of the study, GPs were encouraged to review both paper and electronic patient records, including discharge letters, to ensure comprehensive and ac- curate reporting. This approach aimed to reduce recall bias and increase the validity of the results. The retrospective nature of the study, relying on established diagnoses and historical data, brings inherent limitations. In particular, the study fo- cused on the doctor and primary care interval rather than the patient interval, which could af- fect the accuracy of reported symptoms and tim- ing. Despite efforts to minimize recall errors, in- complete information in some questionnaires may have led to information bias, particularly with re- gard to the duration of doctor and primary care intervals. This bias could underestimate the actual intervals as it is based on fragment-ed or incom- plete patient records. The change of GP chosen by patients between diagnosis and data collection, coupled with the variability of medical record systems, posed a challenge in accessing complete and consistent data. Paper records were frequently relied upon, which were often incomplete and dif- ficult to decipher, while electronic records experi- enced compatibility issues, further complicating data retrieval and potentially leading to missing or inaccurate information. The presence of missing data leads to uncertainty about the actual perfor- mance or recording of diagnostic procedures, such as digital rectal examinations, which could affect the completeness and reliability of results. The same applies to shared decision-making. Clinical practice guidelines on prostate cancer screening using the PSA test have clearly recommended that clinicians practice shared decision making - a process involving clinician-patient discussion of the pros, cons, and uncertainties of screening. However, studies have shown that most men have never engaged in shared decision-making conver- sations with a healthcare provider about PSA test- ing.36,37 In our study, shared decision-making was not recorded in the GP’s medical records, but we cannot say with certainty that it was not carried out either. This study provides a comprehensive exami- nation of early management of prostate cancer patients within the primary healthcare system in Slovenia and offers valuable insights into diagnos- tic timelines and their impact on patient outcomes. The significant correlations found between the duration of intervals, 5-year survival rates and ac- cessibility of laboratory diagnostics emphasize the crucial role of timely and comprehensive diagnos- tic testing in improving prognosis. From a clinical perspective, these findings underscore the impor- tance of ensuring adequate resources and infra- structure for diagnostic testing in primary health care centres to minimize delays in diagnosis and optimize patient care. In addition, the observed re- gional differences in diagnostic intervals highlight the need for targeted interventions to address in- equalities in access to healthcare and improve di- agnostic efficiency in different regions. From a re- search perspective, these findings provide a basis for further investigation of the underlying factors influencing diagnostic processes and the effective- ness of interventions aimed at reducing diagnostic delays. Future research efforts should prioritize the validation of these findings through longitu- dinal studies and comparative analyses, while exploring new strategies to increase diagnostic ef- ficiency and improve patient outcomes in prostate cancer. This study also emphasizes the need for further investigation of the factors that influence an individual’s decision to seek primary health care services, particularly those beyond the scope of the variables included in the current model. Radiol Oncol 2024; 58(4): 544-555. Kokalj Kokot M et al. / Analysis of early diagnostic pathway for prostate cancer in Slovenia554 Acknowledgments We are grateful to all our colleagues who partici- pated in the study and all members of the V3-1713 research group. This research was funded by the Ministry of Education, Science and Sport of the Republic of Slovenia. Grant V3-1713. Principal investigator: Vesna Zadnik. References 1. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mor- tality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2021; 71: 209-49. doi: 10.3322/caac.21660 2. Zadnik V, Gašljević G, Hočevar M, Jarm K, Pompe-Kirn V, Strojan P, et al, editors. Cancer in Slovenia 2020 [Internet]. Ljubljana: Institute of Oncology Ljubljana, Epidemiology and Cancer Registry, Slovenian Cancer Registry; 2023. Available from: http://www.onko-i.si/rrs/ 3. Zaletel M, Vardič D, Hladnik M, editors. [Health statistical yearbook of Slovenia 2017]. [Slovenian]. [internet]. Ljubljana: NIJZ; 2019. Available from: https://nijz.si/publikacije/zdravstveni-statisticni-letopis-slovenije-2017/ 4. Zadnik V, Primic Zakelj M, Lokar K, Jarm K, Ivanus U, Zagar T. Cancer burden in Slovenia with the time trends analysis. Radiol Oncol 2017; 51: 47-55. doi: 10.1515/raon-2017-0008 5. Hamdy FC, Donovan JL, Lane JA, Mason M, Metcalfe C, Holding P, et al. 10-year outcomes after monitoring, surgery, or radiotherapy for local- ized prostate cancer. N Engl J Med 2016; 375: 1415-24. doi: 10.1056/ NEJMoa1606220 6. Wilt TJ, Brawer MK, Jones KM, Barry MJ, Aronson WJ, Fox S, et al. Radical prostatectomy versus observation for localized prostate cancer. N Engl J Med 2012; 367: 203-13. doi: 10.1056/NEJMoa1113162 7. Jain MA, Leslie SW, Sapra A. Prostate cancer screening. Treasure Island (FL): StatPearls Publishing; 2023. 8. PRAISE-U – Uroweb. Smart early dectection of prostate cancer. [Internet]. European Association of Urology [cited 2024 Jan 7]. Available at: https:// uroweb.org/praise-u 9. Neal RD, Tharmanathan P, France B, Din NU, Cotton S, Fallon-Ferguson J, et al. Is increased time to diagnosis and treatment in symptomatic cancer associated with poorer outcomes? Systematic review. Br J Cancer 2015; 112(Suppl 1): S92-107. doi: 10.1038/bjc.2015.48 10. Nicholson BD, Lyratzopoulos G. Progress and priorities in reducing the time to cancer diagnosis. Br J Cancer 2023; 128: 468. doi: 10.1038/s41416-022- 02045-5 11. Lyratzopoulos G, Saunders CL, Abel GA, McPhail S, Neal RD, Wardle J, et al. The relative length of the patient and the primary care interval in patients with 28 common and rarer cancers. Br J Cancer 2015; 112(Suppl 1): S35-40. doi: 10.1038/bjc.2015.40 12. Petrova D, Špacírová Z, Fernández-Martínez NF, Ching-López A, Garrido D, Rodríguez-Barranco M, et al. The patient, diagnostic, and treatment inter- vals in adult patients with cancer from high- and lower-income countries: A systematic review and meta-analysis. PLoS Med 2022; 19: e1004110. doi: 10.1371/journal.pmed.1004110 13. Bosch X, Montori-Palacin E, Martínez-Ferrer R, Aldea A, Moreno P, López- Soto A. Time intervals in the care pathway to cancer diagnosis during the COVID-19 pandemic: a large retrospective study from a high-volume center. Int J Cancer 2023; 152: 384-95. doi: 10.1002/ijc.34260 14. Jessen NH, Jensen H, Helsper CW, Falborg AZ, Glerup H, Gronbaek H, et al. Cancer suspicion, referral to cancer patient pathway and primary care interval: a survey and register study exploring 10 different types of abdomi- nal cancer. Fam Pract 2021; 38: 589-97. doi: https://pubmed.ncbi.nlm.nih. gov/33904928/ 15. Helsper C, van Erp N, Peeters P, de Wit N. Time to diagnosis and treatment for cancer patients in the Netherlands: room for improvement? Eur J Cancer 2017; 87: 113-21. doi: 10.1016/j.ejca.2017.10.003 16. Nguyen DD, Haeuser L, Paciotti M, Reitblat C, Cellini J, Lipsitz SR, et al. Systematic review of time to definitive treatment for intermediate risk and high risk prostate cancer: are delays associated with worse outcomes? J Urol 2021; 205: 1263-74. doi: 10.1097/JU.0000000000001601 17. van Erp NF, Helsper CW, Olyhoek SM, Janssen RRT, Winsveen A, Peeters PHM, et al. Potential for reducing time to referral for colorectal cancer patients in primary care. Ann Fam Med 2019; 17: 419-27. doi: 10.1370/ afm.2446 18. Weller D, Vedsted P, Rubin G, Walter FM, Emery J, Scott S, et al. The Aarhus statement: improving design and reporting of studies on early cancer diag- nosis. Br J Cancer 2012; 106: 1262-7. doi: 10.1038/bjc.2012.68 19. Coxon D, Campbell C, Walter FM, Scott SE, Neal RD, Vedsted P, et al. The Aarhus statement on cancer diagnostic research: turning recommendations into new survey instruments. BMC Health Serv Res 2018; 18: 677. doi: 10.1186/s12913-018-3476-0 20. Cranfield BM, Koo MM, Abel GA, Swann R, McPhail S, Rubin GP, et al. Primary care blood tests before cancer diagnosis: National Cancer Diagnosis Audit data. Br J Gen Pract 2023; 73: E95-103. doi: 10.3399/BJGP.2022.0265 21. Cranfield BM, Abel GA, Swann R, Moore SF, McPhail S, Rubin GP, et al. Pre- referral primary care blood tests and symptom presentation before cancer diagnosis: National Cancer Diagnosis Audit Data. Cancers 2023; 15: 3587. doi: 10.3390/cancers15143587 22. Harris M, Brekke M, Dinant GJ, Esteva M, Hoffman R, Marzo-Castillejo M, et al. Primary care practitioners’ diagnostic action when the patient may have cancer: an exploratory vignette study in 20 European countries. BMJ Open 2020; 10: e035678. doi: 10.1136/bmjopen-2019-035678 23. Sekhoacha M, Riet K, Motloung P, Gumenku L, Adegoke A, Mashele S. Prostate cancer review: genetics, diagnosis, treatment options, and alternative approaches. Molecules 2022; 27: 5730. doi: 10.3390/mol- ecules27175730 24. Wilkinson AN. Cancer diagnosis in primary care: Six steps to reducing the diagnostic interval. Can Fam Physician 2021; 67: 265-8. doi: 10.46747/ cfp.6704265 25. Koo MM, Swann R, McPhail S, Abel GA, Renzi C, Rubin GP, et al. Morbidity and measures of the diagnostic process in primary care for patients subse- quently diagnosed with cancer. Fam Pract 2022; 39: 623-32. doi: 10.1093/ fampra/cmab139 26. Hansen RP, Vedsted P, Sokolowski I, Søndergaard J, Olesen F. Time inter- vals from first symptom to treatment of cancer: a cohort study of 2,212 newly diagnosed cancer patients. BMC Health Serv Res 2011; 11: 284. doi: 10.1186/1472-6963-11-284 27. Gnanapragasam VJ, Greenberg D, Burnet N. Urinary symptoms and prostate cancer-the misconception that may be preventing earlier presentation and better survival outcomes. BMC Med 2022; 20: 264. doi: 10.1186/s12916- 022-02453-7 28. Holtedahl K, Borgquist L, Donker GA, Buntinx F, Weller D, Campbell C, et al. Symptoms and signs of urogenital cancer in primary care. BMC Primary Care 2023; 24: 1-15. doi: 10.1186/s12875-023-02063-z 29. Scheel BI, Holtedahl K. Symptoms, signs, and tests: the general practitioner’s comprehensive approach towards a cancer diagnosis. Scand J Prim Health Care 2015; 33: 170-7. doi: 10.3109/02813432.2015.1067512 30. Swann R, Lyratzopoulos G, Rubin G, Pickworth E, McPhail S. The fre- quency, nature and impact of GP-assessed avoidable delays in a population- based cohort of cancer patients. Cancer Epidemiol 2020; 64: 101617. doi: 10.1016/j.canep.2019.101617 31. Lim AWW, Mesher D, Gentry-Maharaj A, Balogun N, Widschwendter M, Jacobs I, et al. Time to diagnosis of Type I or II invasive epithelial ovar- ian cancers: a multicentre observational study using patient questionnaire and primary care records. BJOG 2016; 123: 1012-20. doi: 10.1111/1471- 0528.13447 32. Bajželj M. [Statistical portrait of Slovenian regions 2026]. [Slovenian]. [internet]. Ljubljana: Statistical Office of the Republic of Slovenia; 2016. [cited 2024 Feb 6]. Available at: https://www.stat.si/dokument/8941/ regije-v-stevilkah.pdf Radiol Oncol 2024; 58(4): 544-555. Kokalj Kokot M et al. / Analysis of early diagnostic pathway for prostate cancer in Slovenia 555 33. Petrova D, Garrido D, Špacírová Z, Fernández-Martínez NF, Ivanova G, Rodríguez-Barranco M, et al. Duration of the patient interval in breast cancer and factors associated with longer delays in low-and middle-income countries: a systematic review with meta-analysis. Psychooncology 2023; 32: 13-24. doi: 10.1002/pon.6064 34. Huepenbecker SP, Sun CC, Fu S, Zhao H, Primm K, Giordano SH, et al. Factors impacting the time to ovarian cancer diagnosis based on classic symptom presentation in the United States. Cancer 2021; 127: 4151-60. doi: 10.1002/ cncr.33829 35. Zaletel M, Vardič D, Hladnik M, editors. [Health statistical yearbook of Slovenija 2014]. [Slovenian]. [internet]. Ljubljana: National Institute of Public Health; 2016. Available at: https://nijz.si/publikacije/zdravstveni-statisticni- letopis-2014/ 36. Han PKJ, Kobrin S, Breen N, Joseph DA, Li J, Frosch DL, et al. National evidence on the use of shared decision making in prostate-specific antigen screening. Ann Fam Med 2013; 11: 306-14. doi: 10.1370/afm.1539 37. Bhojani N, Miller LE, Zorn KC, Chughtai B, Elterman DS, Bhattacharyya S, et al. Prevalence and determinants of shared decision-making for PSA test- ing in the United States. Prostate Cancer Prostatic Dis 2024. doi: 10.1038/ s41391-024-00843-x