記住我
ORIGINAL ARTICLE
Year : 2022 | Volume
: 33
| Issue : 2 | Page : 56-62
The impact of socioeconomic inequality on urological cancer: A nationwide population-based study in Taiwan
Yi-Hsuan Wu1, Hung-Lung Ke2, Hung-Pin Tu3, Ching-Chia Li2, Wen-Jeng Wu2, Wei-Ming Li2
1 Department of Urology, Kaohsiung Medical University Hospital; Department of Urology, Kaohsiung Municipal Hsiao-Kang Hospital, Kaohsiung, Taiwan
2 Department of Urology, Kaohsiung Medical University Hospital; Department of Urology, Kaohsiung Medical University, Kaohsiung, Taiwan
3 Department of Public Health and Environmental Medicine, School of Medicine, College of Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan
Correspondence Address:
Wei-Ming Li
100, Shih-Chuan 1st Road Sanmin District, Kaohsiung 807
Taiwan
Source of Support: None, Conflict of Interest: None
DOI: 10.4103/UROS.UROS_60_21
Purpose: Socioeconomic inequality may contribute to different risk factors for cancers. This study aims to analyze the socioeconomic patterns of urological cancer incidence and mortality in Taiwan. Materials and Methods: Using data from the National Health Insurance, we designed a retrospective longitudinal cohort study of 3686 subjects who were newly diagnosed with bladder cancer (BC), kidney cancer (KC), and upper urinary tract cancer (UTUC) between 2000 and 2010. We analyzed patients' characteristics and mortality among the three cancers. Results: The average age of KC diagnosis was the youngest among the cancers. Moreover, KC tends to occur in patients with higher-income occupations who reside in urban areas. Both BC and UTUC were much more prevalent in patients with less socioeconomic means and those living in rural areas. Varied comorbidities showed different distributions among urological cancers. Although the extent was most prominent in KC, both overall mortality and cancer-specific mortality of the three cancers increased every year during the follow-up period. Conclusion: Our results demonstrate different patient characteristics and mortality among BC, KC, and UTUC in Taiwan.
Keywords: Income, mortality, occupation, urological cancer
The incidence of major cancers in different countries is unequally distributed across socioeconomic groups.[1],[2] A higher incidence is found mostly in patients with lower socioeconomic status. Each cancer has its natural course and shows varying levels of aggressiveness. Despite the characteristics of cancer itself, unfavorable socioeconomic status predicts poor survival across a wide range of cancers.[3],[4] These findings may be associated with educational level, occupational exposure, income, and lifestyle-related risk factors. However, limited investigations focusing on cancer in the urinary system are noted.
Urological cancers affect organs in both genders and impose an economic burden on the patient and the health-care system. Bladder cancer (BC), upper urinary tract cancer (UTUC), and kidney cancer (KC), which mostly refer to renal cell carcinoma (RCC), are common.[5],[6] Extrapolating from the latest GLOBOCAN data, BC is the seventh most commonly diagnosed cancer and accounts for 4.5% of new cancer cases in men worldwide.[7] Its incidence has risen steadily, especially in developed countries. Although the urothelial cell lining from the renal pelvis and ureter to the urinary bladder is similar, UTUC is much rarer and accounts for 5%–10% of all urothelial cancers.[8] Moreover, pyelocaliceal cancers are twice as common as ureteral cancer.[9] Unlike urothelial cancer, RCC originates from the proximal renal tubular epithelium. It constitutes nearly 90% of primary malignant renal tumors, and the highest incidence occurs in western countries.[10] It also accounts for 5% and 3% of all oncological diagnoses in men and women, respectively.[10]
Although they share similar characteristics, urothelial carcinomas of the bladder and upper tract are distinct diseases. Tobacco consumption and occupational exposure to aromatic amines are important risk factors for both.[8],[11] Drinking water with arsenic compounds and taking aristolochic acid in Chinese herbs are also recognized causes of urothelial carcinoma of the bladder and upper tract. A possible association is noted between these carcinogenic exposures and the highest UTUC incidence on the southwest coast of Taiwan.[8] However, overt differences exist in BC and UTUC (e.g., gender incidence, response to surgery, intravesical therapy, systemic chemotherapy, and even survival outcomes). They may result from anatomical, biological, or molecular variations. Much is still unknown in this field.[9] Contrary to urothelial cancer, KC is not considered an occupational disease. KC mainly affects older and male populations. It is also associated with obesity, physical inactivity, reduced vegetable consumption, and some comorbidities, which generally occur in industrialized countries.[12]
It is well known that a high incidence of UTUC exists in Taiwan. It probably results from exposure to arsenic and aristolochic acid in Chinese herbs.[8] This research provided us the opportunity to compare cancers in the urinary tract. Apart from known carcinogens, the association between urological cancers and personal factors from comorbidities or socioeconomic factors was considered. The Taiwan National Health Insurance (NHI) Research Database provides not only disease information but also associated premium levels, which are calculated according to employment status and dependent household members. This study aimed to investigate the differences among urological cancers, accounting for comorbidities and socioeconomic parameters.
Materials and MethodsData source and ethical approval
Taiwan's NHI system covers over 99% of the population. Using the Taiwan Cancer Registry and NHI Research Database (LGTD2000) of 2 million people, patient data, including personal characteristics and all medical records from 2000 to 2010, were retrieved. Patient information, including birthday, diagnosis date, gender, residency, employment status, income inferred from insurance fees, cause of mortality, and mortality date was retrieved. The diagnoses were coded according to the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM). All information from the NHI research database was decoded to protect patient privacy. Informed consent was not required because of the use of coded information following the National Health Research Institutes regulations. This study was approved by the Institutional Review Board of Kaohsiung Medical University Hospital (KMUH-IRB-EXEMPT-20140016).
Study design and enrolled subjects
This study used a longitudinal cohort of 1 million insured individuals. One million subjects were traced and included those who were diagnosed with urological cancer between 2000 and 2010 by maximal 10-year follow-up as the study cohort. The current study ultimately identified 3686 subjects with newly diagnosed urological cancer, coded as ICD-9-CM (code 188 for BC, ICD-9-CM code 189.0 for KC, ICD-9-CM code 189.1, and ICD 189.2 for UTUC). The exclusion criteria were as follows: (1) age <18 years and (2) diagnosis of BC, KC, or UTUC before the index date. The subject was enrolled by the first diagnosis if a subject obtains two or more of these urological cancers. The medical records of enrolled subjects were examined and their baseline characteristics, comorbidities, insurance premiums, geographical residence area, and occupation were collected.
Measures and definitions
Subjects with newly diagnosed urological cancer and comorbidities in the study cohort were identified using ICD-9-CM diagnostic codes and diagnoses made by a clinical physician. In Taiwan, board-certified urologists made cancer diagnoses and staging according to clinical, laboratory, and imaging findings, and they investigated the pathology last. Moreover, the database of the current study had concatenated with the Taiwan Cancer Registry. Therefore, the cancer diagnosis is accurate. In addition, the Charlson Comorbidity Index (CCI), which has been used to evaluate the coexisting medical condition of the subjects, was applied. Comorbidities were also examined with diagnostic codes, including myocardial infarction (ICD-9-CM codes 410 and 412), congestive heart failure (ICD-9-CM code 428), peripheral vascular disease (ICD-9-CM codes 443.9, 441, 785.4, and V43.4), cerebrovascular disease (ICD-9-CM codes 430–438), dementia (ICD-9-CM code 290), chronic obstructive pulmonary disease (ICD-9-CM codes 490–505), rheumatologic disease (ICD-9-CM codes 710.0, 710.1, 710.4, 714.0–714.2, 714.81, and 725), peptic ulcer disease (ICD-9-CM codes 531–534), liver disease (ICD-9-CM codes 571.2, 571.4–571.6, 456.0–456.21, and 572.2–572.8), diabetes mellitus (ICD-9-CM codes 250.0–250.7), chronic kidney disease (ICD-9-CM codes 582, 583–583.7, 585, 586, and 588), any malignancy (ICD-9-CM code 140–172, 174–195.8, and 200–208), and metastatic solid tumor (ICD-9-CM codes 196–199.1). The CCI scores were determined based on the defined algorithms of the ICD-9 codes that have been previously validated.[13]
Statistical analysis
All enrolled subjects were followed from the diagnosis date until mortality or the end of follow-up. Statistical analyses were performed using SASv9.3 (SAS Institute Inc., Cary, NC, USA). The average age at diagnosis was presented as the mean ± the standard deviation (SD). Percentages were calculated for categorical variables. The Chi-square test or Fisher's exact test was used to analyze the differences between categorical variables. Baseline characteristics (e.g., age, gender, insurance premiums, residency, occupation, and selected comorbidities) were considered risk factors for the diagnosis of urological cancer and were included in the analysis. Cox proportional hazards regression models were applied to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for the incidence rate of overall mortality and cancer-specific mortality. It was expressed as person-year because the time each participant spends in the cohort study before developing the outcome of interest needs to be calculated. The person-years at risk in an incidence rate is the sum of each individual's time at risk. The annual change in overall mortality and cancer-specific mortality per 100 person-years from 2000 to 2010 was also calculated. Two-tailed P ≤ 0.05 was considered statistically significant.
ResultsFrom the NHI Research Database (LGTD2000), subjects newly diagnosed with BC, KC, and UTUC, excluding those who were aged <18 years or diagnosed before the index date were retrospectively collected. Finally, 2044, 706, and 936 BC, KC, and UTUC were enrolled, respectively [Figure 1].
The three groups were significantly different in terms of diagnosis age, cancer-specific mortality, age group distribution, gender, residency, and occupation [Table 1]. KC patients were the youngest when their diagnosis was confirmed, followed by UTUC and BC (diagnosis age ± SD: KC, 62.6 ± 13.9; UTUC, 67.4 ± 12.0; and BC, 68.5 ± 12.8). The highest proportion of cancer-specific mortalities was for KC (30.6%), followed by UTUC (26.8%) and BC (23.9%). These patients were mostly diagnosed in their 60s. Yet, more than half of the patients with KC were detected earlier (56.3%). BC and KC showed male predominance and UTUC female predominance. KC occurred significantly in North Taiwan (56.1%), especially in Taipei City (41.4%). In contrast, both BC and UTUC were much more prevalent in southern Taiwan, including the Kaoping region, which has a developed city in the South region. The Eastern population was a relatively rural and agricultural area, without overdevelopment. The cancer patients in that area accounted for the least proportion in all three groups. In terms of occupation, BC and UTUC tend to be low income (e.g., farmers, fishermen, or unemployed; 57.9% of BC and 55% of UTUC). KC occurred in those with stable well-paid jobs, even as the head of a company (51.3%).
The three cancers revealed different associations in various coexisting medical conditions [Table 2]. Many patients with urothelial cell carcinoma had suffered simultaneously with peripheral vascular disease, chronic pulmonary disease, and peptic ulcer disease. BC tends to occur in patients with cerebrovascular disease. The results revealed an exclusive association between UTUC and renal disease. After standardizing age, UTUC patients had the most comorbidities and KC the least [Figure 2].
Figure 2: The schematic diagram for distribution of different Charlson Comorbidity Index among three cancers
Table 2: The differences of comorbidity of bladder cancer, kidney cancer, and upper urinary tract cancerThe results indicated higher overall and cancer-specific mortalities in both UTUC and KC subjects [Table 3]. During the follow-up period, the incidence rates of overall mortalities were 10.62, 11.20, and 13.05/100 person-years in the BC, KC, and UTUC groups, respectively. Individuals with KC and UTUC were more likely to expire than BC subjects (adjusted HR of the KC group, 1.39; 95% CI, 1.22–1.59; P < 0.0001; adjusted HR of the UTUC group, 1.30; 95% CI, 1.16–1.46; P < 0.0001). Regarding cancer-specific mortality, the incidence rates were 50.42, 70.52, and 71.31/100 person-years in the BC, KC, and UTUC groups, respectively. Cancer-specific KC mortality was as much as one and a half times than that of BC (adjusted HR of KC group, 1.53; 95% CI, 1.3–1.81; P < 0.0001). Similarly, cancer mortalities in the UTUC patients were much more common than in BC patients (adjusted HR of UTUC group, 1.39; 95% CI, 1.19–1.62; P < 0.0001). In addition, overall mortalities and cancer-specific mortality from these cancers increased from 2000 to 2010. Growth in the KC group was notable [Figure 3].
Figure 3: (a) Overall mortality and (b) cancer-specific mortality of three cancers from 2000 to 2010
Table 3: Incidence and hazard ratios for overall deaths and cancer-specific deaths of three urological cancers Discussion This was the first nationwide population-based study on urological cancers that considered the socioeconomic parameters of diagnosed patients. The incidence of KC was found to be increased in individuals with better socioeconomic status and those living in prosperous regions. In contrast, urothelial cell carcinoma attacks patients with low socioeconomic means living in unprosperous areas.
Advanced age and gender are overtly related to the risk of BC, UTUC, and KC. The average age of BC and UTUC diagnosis was between 70 and 90 years.[8],[14] Men were diagnosed three to four times more than women.[8],[14] Although the incidence and mortality of RCC showed demographic variation, it occurred more often in the older population with 1.5 male predominance.[12] The age-standardized incidence rate per 10,000 is 35.0 in those older than 75 years. The peak incidence of RCC was 60–70 years.[12] Compared with the results of the current study, the age distribution of the diagnosis of these three cancers is generally consistent with previous epidemiological studies. However, the age at diagnosis of BC and UTUC was slightly younger, and the male UTUC predominance was overwhelmingly reversed. The reason for this reversed finding may be explained by aristolochic acid exposure. Women tend to be associated with a higher risk of developing urothelial cell carcinoma.[15],[16]
According to the current research results, urothelial cancer patients mostly lived in the southern region of Taiwan where industry- and agriculture-based cities exist. Urothelial cancer was thought to be more associated with environmental factors.[8],[11],[17] However, the higher incidence of urothelial cancer in these areas may be explained by occupational carcinogen exposure. In addition, environmental factors (e.g., arsenic compound water contamination and aristolochic-based Chinese traditional herbal medicines) were negligible in the southern coastal area of Taiwan. However, demographic factors in RCC are distinct. North America, Western Europe, and Australia/New Zealand had the highest RCC incidences.[12] The RCC incidence was much lower in undeveloped or developing regions or countries.[18] Similarly, RCC occurred more often in those who lived in the northern region of Taiwan, which is relatively prosperous. RCC occurred more often among individuals with well-paid jobs or higher socioeconomic means. The higher RCC incidence in these thriving areas was speculated to relate to lifestyle factors (e.g., obesity, physical inactivity, smoking, or metabolic syndrome). Heck et al. raised the opposite observation about increasing RCC risk among mechanical engineers and agricultural laborers.[19] Therefore, further epidemiological investigations are required.
Rather than investigate the association between these cancers and each disease, the proportion of comorbidities in cancers of the urinary tract was compared. Although many comorbidities, including hypertension,[20],[21] liver disease,[20],[22] and chronic kidney disease,[20],[23],[24] are independently associated with RCC, a higher proportion of these diseases in RCC patients was not found. Instead, urothelial cancer tends to be related to vascular illnesses (e.g., peripheral vascular and cerebrovascular disease). They are common comorbidities of chronic kidney disease, which increases the risk of urothelial cancer.[24],[25],[26] The underlying biological explanation may be multifactorial, including pro-inflammatory status, vascular sclerosis, or a compromised immune system.[27] Interestingly, a higher proportion of peptic ulcer disease and chronic pulmonary disease was noted in urothelial cancer. Michaud et al. found that, after controlling for potential confounders, men who reported a history of gastric ulcers had a higher risk of BC (relative risk = 1.55, 95% CI = 1.03–2.33). They surmised that poor antioxidant absorption caused poor resistance to bladder carcinogens and long-term nonsteroidal inflammatory drug use was an underlying problem.[28] In terms of the association between BC and chronic pulmonary disease, tobacco consumption may be a reasonable explanation because it is related to several types of cancer. Few studies have proved the direct association, but chronic hypoxia caused by lung disease has been found to deteriorate the expression of clinicopathology in BC.[29]
Apart from the aforementioned comorbidities, overall and cancer-specific KC mortalities were the highest among the three urological cancers, which dramatically increased between 2000 and 2010. KC's biological features were speculated to be more aggressive than the other cancers. Regarding stage and size, the surveillance, epidemiology, and end results (SEER) database documented a rise in mortality rate despite improvements in detection imaging and health care.[30] With the understanding of RCC, refinement of prognostic models, and the application of new drugs, the SEER database revealed that the overall RCC mortality has been decreasing since 2001, but a visible decline in cancer-specific mortality was delayed by 10 years.[31] Research on molecular pathways should be a priority.
As a nationwide population-based study with a long observation period, the strength of this research is its large number of cases, reliable information, and parameter categories from the national database. Besides age, gender, and comorbidities, the results also remind physicians of the importance of recording patient's epidemiological and demographic information (e.g., residency and occupation). However, this study had several limitations. First, data were retrieved from a database rather than a questionnaire-based study or medical chart review. Lifestyle-related factors could not be evaluated, condition and medical treatment of the coexisting disease could not be assessed, and specific environmental concerns where patients lived or worked could not be identified. Second, subjects were enrolled and categorized into different cancer groups without prior related malignancies. In addition, the subject was enrolled by the first diagnosis if the subjects were consecutively diagnosed with two or more of these urological cancers. Thus, those who had metachronous urological cancers may be underdetected. The results represent a potential bias. Third, the patients' education level was not analyzed separately from their socioeconomic conditions. Fourth, the living area information from the NHI research database showed bias because the address where people lived may be different from the address of the group insurance applicant. Although the insured address and living area were the same or nearby in most people in Taiwan, residency information may be more precise when corrected by records for the usual outpatient clinic or hospitalization. Last, for accentuating the regional sociodemography, the region was chosen as the parameter rather than urbanization level, which lead to potential bias. Despite the existing limitations, the current study provided a large population with disproportionate impacts on urological cancers based on residency, occupation, and income. A more specific investigation is necessary.
ConclusionKC tends to occur in those with a higher socioeconomic status who live in prosperous areas. In contrast, a higher incidence of urothelial cell carcinoma was found in those with a lower socioeconomic status who lived in resource-poor areas. The influence of occupation, income, and residency on major urological cancers was confirmed and should be taken seriously. These potential parameters should be investigated in detail in future studies.
Financial support and sponsorship
Nil.
Conflicts of interest
Prof. Wen Jeng Wu, an editorial board member at Urological Science, had no role in the peer review process of or decision to publish this article. The other authors declared no conflicts of interest in writing this paper.
References
留言 (0)