The change in epidemiological characteristics and utilization of health services for the diagnosis and treatment of colorectal cancer: A hospital-based retrospective survey Authors: Wenjie Liang, Haiyan Lu, Kaiyong Huang, Hongyuan Huang, Li Yang Published: 2026 ABSTRACT Highlights•Epidemiologic trends in colorectal cancer, Guangxi (2005–2014).•Stage migration observed: Stage I–II up, Stage III down across years.•Diagnostic shift: CT/MRI and colonoscopy up; X-Ray/ultrasound down.•Pretherapy CEA and CA19-9 associated with metastasis and advanced stage.•Increasing prevalence of overweight and smoking among patients.Keywords: Colorectal cancer, Epidemiological characteristics, Health service utilization, Changing trends Introduction The Global Cancer Observatory estimated in 2020 that more than 1.9 million new Colorectal Cancer (CRC) cases and 935,000 deaths occurred in 2020, and CRC ranked third in terms of incidence but second in terms of mortality.1 From 1990 to 2017, the age-standardized incidence rates of CRC increased globally, with substantial regional and national heterogeneity2 geographic disparity of diagnostic and treatment interval, cancer stages, and mortality in CRC patients was also observed.3 Furthermore, the incidence and risk of early-onset CRC (adolescent and young adult) continuously rose.4, 5, 6, 7, 8, 9, 10 Therefore, global trends of CRC incidence and mortality were grouped into three patterns: an increase in both incidence and mortality in the most recent decade; an increase in incidence with a concomitant decrease in mortality; and a decrease in both incidence and mortality.11 The Global Cancer Observatory calculated that 550,628 new cases and 283,751 deaths due to CRC occurred in 2020 in China, and CRC ranked second in terms of incidence and fifth in terms of mortality in all cancers.12 From 1990 to 2017, the incidence and mortality rates of CRC in China demonstrated increasing trends,13 with substantial regional variation nationwide.13,14 In Guangxi Zhuang Autonomous Region (Guangxi), the age-standardized incidence and mortality of CRC were ranked fourth and third in all cancers in 2016, respectively14; moreover, this showed declining trends in urban and rural populations from 2013 to 2016, with a greater decrease in rural populations than that in urban populations.14, 15, 16, 17 Despite the observed downward trend in the incidence and mortality of CRC in Guangxi, the number of new cases still increases due to the large population size. There is a constant transition to western dietary and lifestyle practices with economic growth, whereas the long-term shift in the epidemiology of the demographic and clinical characteristics of CRC in Guangxi was unclear. Although demographic and clinical characteristics of CRC for Guangxi have been previously reported,18,19 these are limited to fewer characteristics to analyze, and neglect the temporal trend, making an understanding of changes with time difficult. This study aimed to demonstrate the shift in demographic and clinical characteristics of colorectal carcinoma and changes in healthcare utilization for CRC diagnosis and treatment in Guangxi from 2005 to 2014, providing evidence for future adjustments to prevention and control strategies. Materials and methods Study site and participants This study was performed in a tertiary A specialized cancer hospital in Nanning, the capital city of Guangxi. The inclusion criteria were as follows: 1) Patients who were diagnosed with primary CRC and accepted the main therapy in the study hospital; 2) The first clinical visit/admission for CRC diagnosis and/or treatment occurred between January 1, 2005, and December 31, 2014; and 3) The availability of comprehensive demographic and clinical information, particularly details on disease stages and pathological diagnoses. Patients who only visited the study hospital for diagnosis, post-treatment follow-up, or outpatient radiotherapy were excluded. All data were obtained from the electronic medical record system.20 Data acquisition and quality control In this study, at least 100 patients were selected by the sampling method yearly from 2005 to 2014. We examined the date of the first visit/admission of patients in the study hospital. We then selected the first patients diagnosed in May 2005, June 2006, July 2007, August 2008, September 2009, October 2010, November 2011, December 2012, March 2013, and April 2014, following which we included cases until 100 eligible patients were obtained. We used a simplified table to filter patients with eligible and noneligible CRC. Colorectal cancer staging was determined according to the American Joint Committee on Cancer (AJCC) TNM classification, 7th edition.20 A survey was conducted on patients with primary CRC via a structured case report form offered by the Chinese National Cancer Center (CNCC). The following data were collected: 1) Demographic characteristics including sex, age, height, weight, tobacco and alcohol consumption statuses; 2) Clinical characteristics, such as clinical stage, pathological diagnosis, tumor differentiation, and metastatic state; 3) Technologies used for diagnosis and treatment data and; 4) Results of tumor markers. Managers of our team attended a training course organized by the CNCC in Beijing. Several postgraduate students from the school of public health in Guangxi medical university were recruited, and then team managers trained them until there was a need for an examination. One investigator gathered data from the electronic medical record system of the study hospital and recorded it on the paper-based case report form, while another rechecked. Data were double-entered into the CNCC’s purpose-built database (Epidata 3.1, Odense, Denmark) and then submitted to the CNCC, after which they were supplemented or modified when needed. The implementation of tumor biomarkers test The test was conducted daily by the laboratory department of the study hospital since the results were utilized by clinicians to guide them on diagnosis and treatment. We captured the serum level of Carbohydrate Antigen 19-9 (CA19-9) and Carcinoembryonic Antigen (CEA) and the corresponding normal reference values from the electronic medical record system; values that exceeded the corresponding upper limit of the normal threshold were considered positive, and those lower were considered negative. Statistical analysis According to the official government reform plan, health reforms in China can be stratified into two phases: Phase One (2009 to 2011) and Phase Two (2012 onward).21 Hence, the study period fell within three calendar periods: 2005–2008; 2009–2011; and 2012–2014. All data were analyzed using the Statistical Package for the Social Sciences version 20 statistical software (IBM Corp., Armonk, NY, USA). Demographic and clinical features were described using proportions; age at diagnosis was described by means; categorical variables were analyzed with either a chi-square test or Fisher exact test probability method when appropriate. The time trend of age at diagnosis from 2005 to 2014 was determined using Spearman correlation analysis. Temporal trends of proportion were determined using linear-by-linear association of the Chi-Square test. The influencing factors of the positive rate of CEA and CA19-9 were determined using the binary logistic regression analysis. Missing/unknown values were omitted from the statistical analyses. All tests of statistical significance were two-sided, and a p-value < 0.05 was considered statistically significant. Results Demographics and behavior characteristics Overall demographics and behavior characteristics Records of 1065 patients diagnosed with primary CRC from 2005 to 2014 were collected. The overall mean age at diagnosis was 57.33-years, and the total male-to-female ratio was 1.51. The majority of the patients were farmers and retirees (33.59% and 23.04%, respectively), and 44.49% had no insurance. The percentage of patients who consumed alcohol and tobacco was 9.18% and 6.68%, respectively. These data are provided in Table 1. Table 1Demographic and clinical characteristics of colorectal cancer in Guangxi during 2005‒2014.Table 1 dummy alt textCharacteristicsNumber of patients (%)p‡2005‒20142005‒20112009‒20112012‒2014Age at diagnosis (year), Mean ± SD57.33 ± 13.3956.45 ± 13.7158.74 ± 13.6057.03 ± 12.650.041Age at diagnosis (year), male, Mean ± SD57.44 ± 13.5656.41 ± 14.0259.35 ± 13.6957.02 ± 12.670.049Age at diagnosis (year), female, Mean ± SD57.14 ± 13.1456.50 ± 13.2357.91 ± 13.5057.03 ± 12.670.570Age group at diagnosis, Year0.075 <45220 (20.66)92 (22.17)66 (20.00)62 (19.38) 45‒59336 (31.55)130 (31.33)89 (26.97)117 (36.56) ≥60509 (47.79)193 (46.51)175 (53.03)141 (44.06)Sex0.475 Male641 (60.19)257 (61.93)190 (57.58)194 (60.62) Female424 (39.81)158 (38.07)140 (42.42)126 (39.38)Body mass index, kg/m2*0.008 <18.5179 (16.85)84 (20.24)54 (16.36)41 (12.81)b 18.5‒23.9724 (68.17)286 (68.92)219 (66.36)222 (69.38) ≥24.0159 (14.98)45 (10.84)57 (17.27)57 (17.81)a,bOccupation*0.001 Public servant108 (10.45)60 (15.04)29 (8.92)19 (6.15)a,b Manufacturing staff112 (10.84)50 (12.53)26 (8.00)36 (11.65) Farmer347 (33.59)107 (26.82)125 (38.46)115 (37.22)a,b Retiree238 (23.04)91 (22.81)84 (25.85)63 (20.39) Other228 (22.07)91 (22.81)61 (18.77)76 (24.60)Type of health insurance*<0.001 Urban employee basic medical insurance226 (21.48)61 (14.84)74 (22.63)91 (28.98)a,b New rural cooperative medical scheme254 (24.14)6 (1.46)113 (34.56)135 (42.99)a,b No insurance468 (44.49)314 (76.40)90 (27.52)64 (20.38)a,b Others104 (9.89)30 (7.30)50 (15.29)24 (7.64)Alcohol consumption status*<0.001 No960 (90.82)395 (95.18)286 (86.67)279 (89.42)a,b Yes97 (9.18)20 (4.82)44 (13.33)33 (10.58)a,bTobacco consumption status*<0.001 No884 (83.32)376 (90.82)259 (78.96)249 (78.06)a,b Yes177 (16.68)38 (9.18)69 (21.04)70 (21.94)a,bHistory of any CRC-related disease0.918 Yes55 (5.16)20 (4.82)18 (5.45)17 (5.31) No1010 (94.84)395 (95.18)312 (94.55)303 (94.69)History of tumors0.567 Yes15 (1.41)6 (1.45)3 (0.91)6 (1.88) No1050 (98.59)409 (98.55)327 (99.09)314 (98.13)Family history of tumors<0.001 Yes53 (4.98)17 (4.10)7 (2.12)29 (9.06)b,c No1012 (95.02)398 (95.90)323 (97.88)291 (90.94)b,cMetastasis at diagnosis0.179 No762 (73.27)288 (71.11)235 (72.31)239 (77.10) Yes278 (26.73)117 (28.89)90 (27.69)71 (22.90)Histologic type0.896 Adenocarcinoma988 (92.77)384 (92.53)309 (93.64)295 (92.19) Mucinous adenocarcinoma54 (5.07)20 (4.82)16 (4.85)18 (5.63) Signet-ring cell carcinoma6 (0.56)2 (0.48)2 (0.61)2 (0.63) Other types17 (1.60)9 (2.16)3 (0.91)5 (1.56)Tumor differentiation*<0.001 Well-differentiated50 (6.93)19 (9.60)7 (2.79)24 (8.82)a,c Moderately-differentiated548 (76.01)141 (71.21)186 (74.10)221 (81.25)b Poorly differentiated123 (17.06)38 (19.19)58 (23.11)27 (9.93)b,cStage at diagnosis*0.287 Stage Ⅰ45 (4.48)11 (2.72)16 (5.13)18 (6.25) Stage Ⅱ321 (31.97)122 (30.20)103 (33.01)96 (33.33) Stage Ⅲ346 (34.46)150 (37.13)103 (33.01)93 (32.29) Stage Ⅳ292 (29.08)121 (29.95)90 (28.85)81 (28.13)Tumor location*0.810 Right colon256 (24.06)101 (24.34)84 (25.45)71 (22.26) Left colon250 (23.50)98 (23.61)80 (24.24)72 (22.57) Rectum542 (50.94)209 (50.36)160 (48.48)173 (54.23) Colon and rectum16 (1.50)7 (1.69)6 (1.82)3 (0.94)CRC, Colorectal Cancer; SD, Standard Deviation.Notes:⁎The number of patients with not report included 3 patients with unknown body mass index, 32 patients with unknown occupation, 13 patients with unknown types of health insurance, 8 patients with unknown alcohol consumption status, 4 patients with unknown tobacco consumption status, 267 patients with unknown tumor differentiation, 61 patients with unknown stage at diagnosis, and 1 patient with unknown tumor location.‡Comparisons were performed among 2005‒2008, 2009‒2011, 2012‒2014.a2005‒2008 vs. 2009‒2011, p < 0.05.b2005‒2008 vs. 2012‒2014, p < 0.05.c2009‒2011 vs. 2012‒2014, p < 0.05.†CRC-related diseases included colorectal polyps, inflammatory bowel disease, and other benign colorectal conditions. Demographic and clinical characteristics of colorectal cancer in Guangxi during 2005‒2014. CRC, Colorectal Cancer; SD, Standard Deviation. Notes: The number of patients with not report included 3 patients with unknown body mass index, 32 patients with unknown occupation, 13 patients with unknown types of health insurance, 8 patients with unknown alcohol consumption status, 4 patients with unknown tobacco consumption status, 267 patients with unknown tumor differentiation, 61 patients with unknown stage at diagnosis, and 1 patient with unknown tumor location. Comparisons were performed among 2005‒2008, 2009‒2011, 2012‒2014. 2005‒2008 vs. 2009‒2011, p < 0.05. 2005‒2008 vs. 2012‒2014, p < 0.05. 2009‒2011 vs. 2012‒2014, p < 0.05. †CRC-related diseases included colorectal polyps, inflammatory bowel disease, and other benign colorectal conditions. Time trends in demographics and behavior characteristics The percentage of farmers increased from 25.81% to 33.59% between 2005 and 2014 (p-trend = 0.001). The percentage of patients with a new rural cooperative medical scheme and those with urban employee basic medical insurance increased from 2005 to 2014 (all p-trend < 0.001), while that of patients with no insurance decreased from 80.61% to 39.42% between 2005 and 2014 (p-trend < 0.001). The proportion of overweight or obese patients ranged from 4.95% to 23.76% between 2005 and 2014, with an upward trend (p-trend = 0.011). The percentage of patients with tobacco consumption increased from 4.00% to 25.45% between 2005 and 2014 (p-trend < 0.001). Quantitatively, the annual odds of being overweight/obese increased by 18% (OR = 1.18, 95% CI: 1.04–1.34), and the annual odds of tobacco consumption increased by 22% (OR = 1.22, 95% CI: 1.13–1.32). These results are illustrated in Fig. 1. Temporal trends in demographic, behavioral, clinical, and diagnostic characteristics of colorectal cancer patients in Guangxi, 2005–2014. (A) Time trends of patients in different occupations; (B) Time trends of patients covered by different medical insurances; (C) Time trends of patients with different body mass index categories (underweight, normal weight, overweight/obese); (D) Time trend of patients who are smokers or non-smokers; (E) Time trend of patients with different cancer stages (Stage I/II, Stage III, Stage IV); (F) Time trend of patients with different tumor differentiation levels (well/moderately differentiated, poorly differentiated); (G) Time trends of patients undergoing different diagnostic tests (X-Ray, CT, MRI, transabdominal ultrasound, colonoscopy); (H) Time trends of patients treated by different modalities (surgery alone, surgery plus chemotherapy, chemotherapy alone, others); (I) Time trend of the proportion of patients with positive serum CEA before treatment; (J) Time trend of the proportion of patients with positive serum CA19-9 before treatment. Abbreviations: CT, computed tomography; MRI, magnetic resonance imaging; CEA, carcinoembryonic antigen; CA19-9, carbohydrate antigen 19-9. Panels are labeled A–J in the top-left corner of each corresponding image. Clinical and pathological characteristics Overall clinical and pathological characteristics In total, 972 patients reported clinical symptoms. The main clinical symptoms were occult or overt rectal bleeding (27.88%), weight loss (27.37%), abdominal distension (8.95%), diarrhea (6.38%), increasing stool frequency (5.25%), constipation (4.01%), and 26.73% of patients were diagnosed with metastatic disease. The most common histologic type was adenocarcinoma (92.77%); 76.01% of the patients with adenocarcinoma were moderately differentiated, and 63.54% were diagnosed at an advanced stage (stages Ⅲ and Ⅳ). Further information is listed in Table 1. Time trends in clinical and pathological characteristics The percentage of patients diagnosed with stage Ⅰ and stage Ⅱ CRC increased from 13.13% to 39.33% between 2005 and 2014 (p-trend = 0.011), while that of those diagnosed with stage Ⅲ cancer decreased from 50.51% to 17.98% between 2005 and 2014 (p-trend = 0.027). The percentage of patients with poorly differentiated tumors ranged from 17.86% to 5.81%, showing a downward trend from 2005 to 2014 (p-trend = 0.001). This information is presented in Fig. 1. Medical service utilization Overall medical service utilization According to the utilization of CRC-related diagnostic techniques from 2005 to 2014, chest X-Ray was the most commonly used technology (78.59%), followed by transabdominal ultrasound (76.43%) and Computed Tomography (CT) scanning (62.44%). The proportion of patients who conducted the CA19-9 and CEA tests were 71.68% and 97.08%, respectively. The proportion of patients undergoing pretherapeutic serum CEA and CA19-9 testing increased during 2005–2008 and then stabilized during 2009–2014. According to treatment regimens, surgery combined with chemotherapy was the most common treatment option (38.83%). Further details are provided in Table 2. Table 2The usage of service for diagnosis and treatment of colorectal cancer in Guangxi during 2005‒2014.Table 2 dummy alt textCharacteristicsNumber of patients, n (%)p*2005‒20142005‒20082009‒20112012‒2014Utilization of diagnostic technique X-Ray837 (78.59)369 (88.92)266 (80.61)202 (63.13)<0.001a,b,c CT665 (62.44)126 (30.36)263 (79.70)276 (86.25)<0.001a,b MRI107 (10.06)0.00 (0.00)20 (6.06)87 (27.19)<0.001a,b,c Transabdominal ultrasound814 (76.43)362 (87.23)223 (67.58)229 (71.56)<0.001a,b Colonoscopy411 (38.63)90 (21.74)134 (40.61)187 (58.44)<0.001a,b,cUtilization of cancer biomarkers test Carbohydrate antigen19-9 test762 (71.68)130 (33.33)315 (95.45)317 (99.69)<0.001a,b,c Cancer embryo antigen test1032 (97.08)396 (95.42)319 (96.67)317 (99.69)0.003b,cReceipt of treatment<0.001 Surgery alone355 (33.46)119 (28.81)138 (41.95)98 (30.72)a,c Surgery and chemotherapy412 (38.83)190 (46.00)96 (29.18)126 (39.50)a,c Chemotherapy alone254 (23.94)93 (22.52)83 (25.23)78 (24.45) Others40 (3.77)11 (2.66)12 (3.65)17 (5.33)CT, Computed Tomography; MRI, Magnetic Resonance Imaging.Notes: * Comparisons were performed among 2005‒2008, 2009‒2011, 2012‒2014.a2005‒2008 vs. 2009‒2011, p < 0.05.b2005‒2008 vs. 2012‒2014, p < 0.05.c2009‒2011 vs. 2012‒2014, p < 0.05. The usage of service for diagnosis and treatment of colorectal cancer in Guangxi during 2005‒2014. CT, Computed Tomography; MRI, Magnetic Resonance Imaging. Notes: * Comparisons were performed among 2005‒2008, 2009‒2011, 2012‒2014. 2005‒2008 vs. 2009‒2011, p < 0.05. 2005‒2008 vs. 2012‒2014, p < 0.05. 2009‒2011 vs. 2012‒2014, p < 0.05. Time trends in medical service utilization The rate of utilization of chest X-Ray and transabdominal ultrasound decreased from 91.00% to 64.55% and 84.00% to 70.00% between 2005 and 2014 (p-trend < 0.001), (p-trend < 0.001), respectively. In contrast, the rate of utilization of CT, magnetic resonance imaging (MRI), and colonoscopy increased from 12.00% to 90.00%, 0.00% to 25.45%, and 18.00% to 73.64% between 2005 and 2014, respectively, (p-trend < 0.001), (p-trend < 0.001) (p-trend < 0.001). The percentage of surgery combined with chemotherapy decreased from 42.86% to 26.36% between 2005 and 2014 (p-trend = 0.032). This information is displayed in Fig. 1. CA19-9 and CEA The positive rate of CA19-9 and CEA The positive rates of serum CEA and CA19-9 were computed in patients who did not receive treatment in other hospitals, and the serum levels of CEA and CA19-9 before treatment were detected in the study hospital. There were notable differences in the positive rates of serum CEA and CA19-9 before treatment in patients with or without metastasis and disease stages at diagnosis in the three calendar periods. Additional information is presented in Table 3, Table 4. Table 3The positive rate of CEA by demographic and clinical characteristics of colorectal cancer in Guangxi during 2005‒2014.Table 3 dummy alt textCharacteristics2005‒20142005‒20082009‒20112012‒2014N° of patientsPositive rate, n (%)p-valueN° of patientsPositive rate, n (%)p-valueN° of patientsPositive rate, n (%)p-valueN° of patientsPositive rate, n (%)p-valueTotal663308 (46.46)224138 (61.61)21174 (35.07)22896 (42.11)Sex0.7000.2400.0090.896 Male397182 (45.84)14191 (64.54)11732 (27.35)13959 (42.45) Female266126 (47.37)8347 (56.63)9442 (44.68)8937 (41.57)Age group at diagnosis, year0.5200.1030.6090.296 <4512251 (41.8)4220 (47.62)3911 (28.21)4120 (48.78) 45‒5920396 (47.29)7349 (67.12)4617 (36.96)8430 (35.71) ≥60338161 (47.63)10969 (63.30)12646 (36.51)10346 (44.66)Body mass index, kg/m20.0960.6870.2870.460 <18.58545 (52.94)2919 (65.52)2812 (42.86)2814 (50.00) 18.5‒23.9470222 (47.23)173104 (60.12)13750 (36.50)16068 (42.50) ≥24.010841 (37.96)2215 (68.18)4612 (26.09)4014 (35.00) Metastasis at diagnosis<0.0010.009<0.001<0.001 No531218 (41.05)178103 (57.87)16949 (28.99)18466 (35.87) Yes12085 (70.83)4032 (80.00)3924 (61.54)4129 (70.73)Tobacco consumption status0.2840.8230.2030.481 No543257 (47.33)201123 (61.19)16260 (37.04)18074 (41.11) Yes11749 (41.88)2214 (63.64)4813 (27.08)4722 (46.81)Alcohol consumption status0.2771.0000.3630.121 No587275 (46.85)209125 (59.81)0.03918266 (36.26)19684 (42.86) Yes7028 (40.)1513 (86.67)298 (27.59)267 (26.92)Stage at diagnosis<0.0010.071<0.001<0.001 Stage Ⅰ and Ⅱ26594 (35.47)7944 (55.70)9124 (26.37)9526 (27.37) Stage Ⅲ231118 (51.08)9560 (63.16)7125 (35.21)6533 (50.77) Stage Ⅳ12887 (67.97)4333 (76.74)3824 (63.16)4730 (63.83)Tumor location0.0220.1130.0040.108 Right colon14358 (40.56)4531 (68.89)5013 (26.00)4814 (29.17) Left colon14782 (55.78)5437 (68.52)5228 (53.85)4117 (41.46) Rectum363162 (44.63)12267 (54.92)10632 (30.19)13563 (46.67)Histologic type0.9940.1180.9110.235Non-adenocarcinoma4320 (46.51)188 (44.44)93 (33.33)169 (56.25) Adenocarcinoma620288 (46.45)206130 (63.11)20271 (35.15)21287 (41.04)Degree of differentiation0.1710.6750.6250.120 Well and moderately398173 (43.47)7651 (67.11)13749 (35.77)18573 (39.46) Poorly7740 (51.95)1813 (72.22)4016 (40.00)1911 (57.89)CEA, Cancer Embryo Antigen; N°, Number. The positive rate of CEA by demographic and clinical characteristics of colorectal cancer in Guangxi during 2005‒2014. CEA, Cancer Embryo Antigen; N°, Number. Table 4The positive rate of CA19-9 by demographic and clinical characteristics of colorectal cancer in Guangxi during 2005‒2014.Table 4 dummy alt textCharacteristics2005‒20142005‒20082009‒20112012‒2014N° of patientsPositive rate, n (%)p-valueN° of patientsPositive rate, n (%)p-valueN° of patientsPositive rate, n (%)p-valueN° of patientsPositive rate, n (%)p-valueTotal48289 (18.46)NA5210 (16.13)NA20538 (18.54)21541 (19.07)NASex0.0440.1230.9090.056 Male28444 (15.49)414 (9.76)11521 (18.26)12819 (14.84) Female19845 (22.73)216 (28.57)9017 (18.89)8722 (25.29)Age group at diagnosis, year0.7491.0000.7550.831 <458416 (19.05)122 (16.67)347 (20.59)387 (18.42) 45‒5914624 (16.44)183 (16.67)477 (14.89)8114 (17.28) ≥6025249 (19.44)325 (15.63)12424 (19.35)9620 (20.83)Body mass index, kg/m20.5480.4010.9590.381 <18.5608 (13.33)81 (12.50)285 (17.86)242 (8.33) 18.5‒23.932763 (19.27)459 (20.00)13224 (18.18)15030 (20.00) ≥24.09518 (18.95)00 (0.00)459 (20.00)419 (21.95)Metastasis at diagnosis<0.0010.038<0.001<0.001 No39546 (11.65)485 (10.42)16619 (11.45)18122 (12.15) Yes8241 (50.00)145 (35.71)3718 (48.65)3118 (58.06)Tobacco consumption status0.0720.5860.4680.140 No37975 (19.79)5510 (18.18)15429 (18.83)17036 (21.18) Yes10012 (12.00)70 (0.00)497 (14.29)445 (11.36)Alcohol consumption status0.0331.0000.5330.010 No41682 (19.71)549 (16.67)17734 (19.21)18539 (21.08) Yes605 (8.33)81 (12.5)284 (14.29)260 (0.00)Stage at diagnosis<0.0010.051<0.001<0.001 Stage Ⅰ and Ⅱ21020 (9.52)262 (7.69)909 (10.00)949 (9.57) Stage Ⅲ15523 (14.84)233 (13.04)688 (11.76)6412 (18.75) Stage Ⅳ8542 (49.41)135 (38.46)3618 (50.00)7519 (52.78)Tumor location0.0570.0150.4270.714 Left colon10327 (26.21)166 (37.5)4912 (24.49)389 (23.68) Right colon11016 (14.55)160 (0.00)498 (16.33)458 (17.78) Rectum26244 (16.79)294 (13.79)10517 (16.19)12823 (17.97)Histologic type0.9721.0000.2340.240 Non‒adenocarcinoma224 (18.18)10 (0.00)73 (42.86)141 (7.14) Adenocarcinoma46085 (18.48)6110 (16.39)19835 (17.68)20140 (19.9)Degree of differentiation0.1400.5910.7250.064 Well and moderately34062 (18.24)294 (13.79)13525 (18.52)17633 (18.75) Poorly6517 (26.15)82 (25.00)388 (21.05)197 (36.84)CA19-9, Carbohydrate Antigen19-9; N°, Number. The positive rate of CA19-9 by demographic and clinical characteristics of colorectal cancer in Guangxi during 2005‒2014. CA19-9, Carbohydrate Antigen19-9; N°, Number. The positive rate of CEA combined with CA19-9 was different in the presence or absence of metastasis and disease stage at diagnosis; detailed data are provided in Table 5. The binary logistic regression models showed that metastasis and the disease stage at diagnosis can affect the serum CEA level, and that metastasis and sex can influence the expression of serum CA19-9 (Table 6). Notably, the OR for metastasis in the CEA model had a wide 95% CI (1.02–17.53), suggesting imprecision likely due to the relatively small number of metastatic cases with complete data and correlation among predictors; therefore, this estimate should be interpreted cautiously. Table 5The positive rate of CEA combined with CA19-9 by demographic and clinical characteristics of colorectal cancer.Table 5 dummy alt textCharacteristics2005‒20142005‒20082009‒20112012‒2014N° of patientsPositive rate, n (%)p-valueN° of patientsPositive rate, n (%)p-valueN° of patientsPositive rate, n (%)p-valueN° of patientsPositive rate, n (%)p-valueTotal516256 (49.61)7358 (79.45)21391 (42.72)230107 (46.52)Sex0.0340.6540.0010.972 Male304139 (45.72)4535 (77.78)11939 (32.77)14065 (46.43) Female212117 (55.19)2823 (82.14)9452 (55.32)9042 (46.67)Age group at diagnosis, year0.7080.1300.7290.617 <459645 (46.88)159 (60.00)3915 (38.46)4221 (50.) 45‒5915575 (48.39)2320 (86.96)4719 (40.43)8536 (42.35) ≥60265136 (51.32)3529 (82.86)12757 (44.88)10350 (48.54)Body mass index, kg/m20.1060.1770.4170.745 <18.56638 (57.58)109 (90.00)2815 (53.57)2814 (50.) 18.5‒23.9353178 (50.42)5444 (81.48)2815 (53.57)16176 (47.2) ≥24.09740 (41.24)95 (55.56)13858 (42.03)4117 (41.46) Metastasis at diagnosis<0.0011.000<0.001<0.001 No411177 (43.07)5745 (78.95)16961 (36.09)18571 (38.38) Yes9977 (77.78)1613 (81.25)4129 (70.73)4235 (83.33)Tobacco consumption status0.1860.3560.0610.733 No410209 (50.98)6651 (77.27)16274 (45.68)18284 (46.15) Yes10345 (43.69)77 (100.00)4915 (30.61)4723 (48.94)Alcohol consumption status0.0590.2890.1710.043 No447227 (50.78)6550 (76.92)18482 (44.57)19895 (47.98) Yes6324 (38.1)88 (100.00)299 (31.03)267 (26.92)Stage at diagnosis<0.0011.000<0.001<0.001 Stage Ⅰ and Ⅱ21582 (38.14)2823 (82.14)9131 (34.07)9628 (29.17) Stage Ⅲ16486 (52.44)2823 (82.14)7128 (39.44)6535 (53.85) Stage Ⅳ10378 (75.73)1512 (80.00)4029 (72.5)4837 (77.08)Tumor location0.0430.8070.0020.139 Left colon11365 (57.52)1914 (73.68)5233 (63.46)4218 (42.86) Right colon11748 (41.03)1714 (82.35)5217 (32.69)4817 (35.42) Rectum277138 (49.82)3528 (80.00)10640 (37.74)13670 (51.47)Histologic type0.5260.3710.6520.419 Non-adenocarcinoma2715 (55.56)21 (50.00)95 (55.56)169 (56.25) Adenocarcinoma489241 (49.28)7157 (80.28)20486 (42.16)21498 (45.79)Degree of differentiation0.2771.0000.7460.118 Well and moderately360170 (47.22)3527 (77.14)13860 (43.48)18783 (44.39) Poorly6837 (54.41)86 (75.00)4119 (46.34)1912 (63.16)CEA, Cancer Embryo Antigen; CA19-9, Carbohydrate Antigen19-9; N°, Number.p-values of “1.000″ indicate exact values from Fisher's exact test (due to small expected cell counts). The positive rate of CEA combined with CA19-9 by demographic and clinical characteristics of colorectal cancer. CEA, Cancer Embryo Antigen; CA19-9, Carbohydrate Antigen19-9; N°, Number. p-values of “1.000″ indicate exact values from Fisher's exact test (due to small expected cell counts). Table 6Binary logistic regression analysis of tumor biomarkers and demographic and clinical characteristics for colorectal cancer.Table 6 dummy alt textParameterβp-valueOR (95% CI)CEA Intercept−0.66<0.0010.52 Metastasis at diagnosis (Yes vs. No)1.440.0474.23 (1.02‒17.53) Stage at diagnosis (Ⅲ vs. ⅠandⅡ)0.640.0011.90 (1.31‒2.75) Stage at diagnosis (Ⅳ vs. ⅠandⅡ)−0.010.9880.99 (0.24‒4.06)CA19-9 Intercept−2.41<0.0010.09 Sex (Female vs. Male)0.690.0122.00 (1.16‒3.45) Metastasis at diagnosis (Yes vs. No)2.08<0.0017.99 (4.51‒14.14)CEA combined with CA19-9 Intercept−0.71<0.0010.49 Sex (Female vs. Male)0.500.0121.66 (1.12‒2.45) Stage at diagnosis (Ⅲ vs. ⅠandⅡ)0.550.0101.74 (1.14‒2.66) Stage at diagnosis (Ⅳ vs. ⅠandⅡ)1.73<0.0015.62 (3.22‒9.83)CEA, Cancer Embryo Antigen; CA19-9, Carbohydrate Antigen19-9. Binary logistic regression analysis of tumor biomarkers and demographic and clinical characteristics for colorectal cancer. CEA, Cancer Embryo Antigen; CA19-9, Carbohydrate Antigen19-9. The correlation between the positive rate of CEA, CA19-9, and disease stage at diagnosis To confirm that serum levels of CEA and CA19-9 before treatment were associated with staging, we further assessed the correlations between the positive rate of biomarkers and stages at diagnosis from 2009 to 2014. The positive rate of CEA was positively correlated with stages at diagnosis (r = 0.31, 95% Confidence Interval [95% CI]: 0.18–0.34, p < 0.001). Similarly, the positive rate of CA19-9 was also positively correlated to stages at diagnosis (r = 0.35, 95% CI: 0.21–0.47, p < 0.001), and the positive rate of CEA combined with CA19-9 was positively correlated with stages at diagnosis (r = 0.38, 95% CI: 0.25–0.51, p < 0.001). Time trend in the detection and positive rates of CA19-9 and CEA expressions No obvious shift of the positive rate of serum CA19-9 expression before treatment was observed between 2005 and 2014 (p-trend = 0.278). With regard to positivity, the positive rate of serum CEA before treatment showed a downward trend from 2005 to 2014, (p for trend = 0.003); stabilizing after 2009 (p-trend = 0.096). Additional information is presented in Fig. 1. Discussion As part of the cancer screening program in urban China, this study is a significant high-quality, long-term retrospective epidemiology survey of CRC in Guangxi. We observed that demographics, behavior characteristics, and clinical features such as occupation, medical insurance status, body mass index, tobacco consumption, stages at diagnosis, and tumor differentiation had shifted over 10-years. The utilization of diagnosis techniques also changed remarkably, whereas treatment regimens showed no marked shift. Furthermore, the rate of tumor marker tests continually increased. However, the positive rate was kept stable and was associated with metastasis and the disease stage. In this study, we collected data on demographic and clinical characteristics and usage of medical services over 10-years for the diagnosis and treatments of CRC to comprehensively reveal the long-term dynamic temporal tendency. However, this was not similar to other studies.18,19 In the present study, the observed mean age at diagnosis was 57.33-years, which remained constant over 10-years; this is slightly above that (56-years) observed in a previous study that analyzed patients diagnosed with CRC between 1993 and 2007 in Guangxi,18 and was slightly below that (59-years) reported in another study that analyzed patients diagnosed with CRC between 2003 and 2011 in Guangxi.19 The average age at diagnosis in our study was below the average age at diagnosis reported in China (59.3-years)20 and the global average age at diagnosis (68.8-years).22 The male-to-female ratio (1.57) observed in this study was similar to that reported in two studies that analyzed patients with CRC in Guangxi (1.46 and 1.61).18,19 Furthermore, the ratio was higher than that (1.34) reported in a multicenter retrospective survey20 and that (1.37) reported in the 2017 China cancer registration annual report.23 These results indicate that CRC patients in Guangxi are diagnosed at a younger age and have a higher male-to-female ratio compared with national averages. This study revealed that the proportion of patients with CRC in Guangxi who were underweight showed a downward trend while the percentage of patients who were overweight or obese showed an upward trend from 2005 to 2014. This change observed in the patients who were overweight or obese conformed with a national multicenter survey.20 Obesity has been confirmed as a risk factor for CRC.24, 25, 26 For several decades, there have been remarkable changes in the dietary structure of the Chinese population, characterized by the replacement of the classical diets by more westernized ones (cereals, low-fat mixed dishes, and decreased vegetable intake, while the intake of animal and processed foods, sugars, sweetened beverages, and ultra-processed foods high in energy, fat, sugar, and salt increased),27,28 leading to the prevalence of undernutrition and overnutrition (overweight or obesity) in China.29 This implies a gradual improvement in the standard of living and lifestyle changes in Guangxi over 10-years. However, it also reinforces the need to strengthen public awareness of the benefit of healthy diets and appropriate physical activity to prevent CRC. In this study, the percentage of patients with CRC who were public servants demonstrated a declining trend, whereas that of farmers showed an increasing trend in Guangxi from 2005 to 2014. The change in the percentage of public servants with CRC may be attributed to improved awareness of the relationship between physical inactivity and the development of the disease, while the shift in the percentage of farmers may be due to the sedentary lifestyle in rural areas and the large number of farmers in China. In addition, our study revealed that the proportion of patients with CRC without medical insurance showed a marked decline in Guangxi from 2005 to 2014. This is due to China’s healthcare reform, which has enhanced health insurance coverage to over 90% since 2010.21 Furthermore, China has made substantial progress in improving equal access to care and enhancing financial protection from 2009 to 2019, especially for people of a lower socioeconomic status.30, 31, 32 In this study, an increase was observed in the percentage of tobacco consumption of patients with CRC over time. CRC was designated to be causally related to smoking by the International Agency for Research on Cancer Monographs.33,34 Although smoking prevalence in China decreased from 30.1% in 2000 to 25.6% in 2016 and will continue to reduce to 20%, which is the goal of Healthy China 2030,35,36 following the shift pattern of the prevalence of smoking and cancer deaths in the USA, smoking-related cancer deaths are expected to decrease in 40-years.37 Therefore, there is an expected increase in the percentage of smoking patients with CRC in Guangxi in future studies. Furthermore, the proportion of patients diagnosed with stage Ⅰ and Ⅱ CRC showed an upward trend, while the percentage diagnosed with stage Ⅲ declined from 2005 to 2014. This pattern may be mainly due to healthcare reform, which increased hospital admission and outpatient visit rates,30 and to improved socioeconomic status, which enhanced health awareness and promoted earlier cancer detection. The relatively younger mean age at diagnosis in Guangxi may reflect regional demographic and lifestyle characteristics, including a larger rural population, earlier exposure to risk factors such as tobacco and dietary transitions, and lower life expectancy compared with national and global levels, all of which may contribute to earlier onset of CRC. The utilization of CRC-related diagnostic techniques in Guangxi varied remarkably from 2005 to 2014, indicating gradual improvement in CRC diagnosis and treatment. This trend was consistent with the overall change observed in China.20 Since accurate diagnosis is beneficial for conducting optimal treatment regimens, the application of CT and colonoscopy has gradually increased in CRC diagnosis before the first protocol for the diagnosis and treatment of CRC was made in China.38 In the latest Chinese protocol for the diagnosis and treatment of CRC (2020 edition), CT and MRI were ranked the first and second of all recommended examination methods, respectively. The present survey found that the rate of the pretherapeutic serum CEA test and CA19-9 test in CRC patients grew first during 2005–2008 and then became stable during 2009–2014. This shift may be due to the recommendations of the domestic and foreign protocols about diagnosis and treatment for CRC. The American Society of Clinical Oncology and the European Group on Tumor Markers recommended that preoperative CEA testing should be used for staging/treatment planning and providing prognostic information for CRC patients39,40; nevertheless, the former did not recommend CA 19-9 for diagnosis, staging, surveillance, or monitoring treatment of patients with colorectal cancer,39 whereas the latter suggested that preoperative levels of CA 19-9 may provide independent prognostic information for patients with CRC. The first-edition Chinese protocol of diagnosis and treatment of colorectal cancer was established in 2010 and revised three times. The pretherapeutic serum CEA and CA19-9 testing were recommended in all four editions of the Chinese protocol of diagnosis and treatment of colorectal cancer.38,41, 42, 43 Our study discovered that a combination of the pretherapeutic serum CEA with CA 19-9 can detect more patients with metastatic disease and those diagnosed at advanced stage. This indicated that combination of the pretherapeutic serum CEA with CA19-9 may be helpful for identify asymptomatic patients with metastatic disease or advanced stage. The total positive rate of the pretherapeutic serum levels of CEA and CA19-9 of CRC patients in Guangxi was similar to that reported by a study conducted in Jilin province in China.44 In this study, we observed that the positive rate of the pretherapeutic serum CA19-9 and CEA levels in CRC patients with metastatic disease at diagnosis was significantly higher than that of those without metastatic disease. We also observed that these rates increased with clinical stages. The relationship between the clinical stage and the tumor makers in this study was in accordance with previous studies.44, 45, 46 This indicated that CRC patients with an elevated pretherapeutic serum CA19-9 or CEA level should be further investigated to detect asymptomatic metastatic disease and advanced stage early. This study had several limitations. First, the survey was conducted only in a tertiary A special tumor hospital in the capital city (Nanning) of Guangxi, which may have caused selection bias. Second, the specific test method and the test kit of tumor markers were not available because the study data were retrospectively collected several years after testing. Therefore, considering the possibility of using different test methods and test kits in different years, we only calculated the positive rate of tumor markers in place of the level of tumor markers. Third, the sample size was relatively small; as part of the cancer screening program in urban China, this study used a sampling method, which gathered only about 1000 primary CRC cases over 10-years, which may also have reduced the precision of some regression estimates (e.g., metastasis in the CEA model). Fourth, our sampling method selected patients from a single month per year, which may not fully represent seasonal variations in CRC diagnosis or healthcare seeking behavior. Lastly, the completeness and accuracy of patient information based on the electronic medical record system of the study hospital may be biased. Despite the stated limitations, the findings of this study have significant and practical implications. Time trends observed in our study provide a critical perspective of understanding the change in the demographic and clinical characteristics of CRC and the change in the utilization of technology and service in the diagnosis and treatment of CRC in Guangxi. Furthermore, these results indicated the effect of prevention strategies, control, and management of CRC in the past and provided a significant baseline for assessing the measures of prevention and control of CRC in Guangxi. Conclusion In conclusion, the increasing percentage of patients who are overweight/obese and smoke, as well as the increasing population of farmers, pose challenges for CRC prevention, control, and treatment in Guangxi. The utilization of CT, MRI, and colonoscopy for diagnosing CRC has significantly increased, and the serum level of CEA before therapy can provide information for diagnosing CRC with asymptomatic metastatic disease and advanced stage. Abbreviations CRC, Colorectal Cancer; CanSPUC, Cancer Screening Program in Urban China; CRF, Case Report Form; CNCC, Chinese National Cancer Centre. Ethics approval and consent to participate The survey was approved by the Institutional Review Board of the cancer hospital at the Chinese Academy of Medical Sciences as a component of a broader screening program (program 15-071/998, since the retrospective and noninterventional design of the survey, informed consent from participants were waived). All experimented and procedures were performed in accordance with the Declaration of Helsinki. Consent for publication Not applicable. Data availability Since the National Health and Family Planning Commission (currently National Health Commission) of China and the National Cancer Center of China have formulated security measures. The data are not publicly available due to national policy restrictions; however, de-identified data supporting the findings of this study are available from the corresponding author upon reasonable request for validation purposes, subject to institutional and national regulations. Authors’ contributions Yang Li designed and supervised the work and contributed to the revision of the intellectual content. Liang Wenjie was responsible for the specific implementation of the project, collected and analyzed the data, and drafted the manuscript. Lu Haiyan mainly contributed to the analysis and interpretation of the data. Huang Kaiyong mainly analyzed the data and made some comments when revising the manuscript. Hongyuan Huang mainly analyzed data. Funding Supported by the Guangxi Sub-project of the “National Major Public Health Service Project”; Funded by the Guangxi Natural Science Foundation Youth Fund Project (n° 2022GXNSFBA035641). Declaration of competing interest The authors declare no conflicts of interest. Acknowledgments We are grateful to the Health Economic Evaluation Working Group of the Cancer Screening Program in Urban China from the Chinese National Cancer Center for all their support; we also thank the study hospital, its related staff, and investigators. Edited by José Maria Soares Junior Footnotes Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.clinsp.2026.101075. Supplementary materials References Associated Data Supplementary Materials Data Availability Statement Since the National Health and Family Planning Commission (currently National Health Commission) of China and the National Cancer Center of China have formulated security measures. The data are not publicly available due to national policy restrictions; however, de-identified data supporting the findings of this study are available from the corresponding author upon reasonable request for validation purposes, subject to institutional and national regulations.