Artigos

Introduction
Cancer is the second cause of death after cardiovascular diseases in the world (1), thus representing 14% of total world mortality. Risk factors classically associated with gastric cancer are dietary factors, improper food preservation, and Helicobacter pylori infection (2). In turn, gastric cancer mortality has ranked fourth among all cancer classes since 2015 (3), being the leading cause of cancer death in men (4). More than half (4.4 million) of the eight million cancer deaths occurred in people under 70 years old (5). Of these, 44% occurred in upper-middle-income countries (UMICs), such as Colombia, while only eight percent occurred in low-income countries (LICs) (6, 7). Inequalities in cancer survival rates between and within countries differ for reasons such as variations in education, access to specialized care, timely treatment, and the status of health insurance (9,10).
In Colombia, mortality from malignant neoplasms was the third leading cause of death in men and the second in women between 2003 and 2012. It represented 16.6% (157,017 deaths) of all deaths, with gastric cancer being the leading cause of death. Death among all cancer groups represented 14.1% of deaths in both sexes and a standardized mortality rate of 14.2/100.000 inhabitants (8).
Despite the efforts directed by the World Health Organization (WHO) to address the social causes of poor health and avoidable health inequalities (11), persistent inequalities in the distribution of wealth and access continue to be an obstacle to the improvement of the health conditions of the communities (12).
Social inequalities manifest themselves in different categories, such as social class, educational level, gender, age, ethnicity, disability, and geographic location (13,14). Educational inequalities may contribute to socioeconomic inequality, and it is a matter of main interest worldwide due to the strong association that exists between education and health (8). This relationship is driven by differences in lifestyle, health behaviors, and accessibility to timely cancer care (15) from an individual and social perspective (16).
Health inequalities have been documented mainly in Western countries (17), leading to an increasing desire at the international level to formulate policies under the Social Determinants of Health (SDH) approach as the main action to reducing inequalities (11). In a previous study in Colombia, we found that the main contributor to educational inequalities in cancer mortality among men was gastric cancer (49%). It was the second-highest contributor among women after cervical cancer (18). As Colombia is a country with great socioeconomic inequalities, there are marked differences in life expectancy and mortality from different causes (19), and as it has formally adopted a health model under the framework of the SDH, there is a growing need for an analysis of health inequalities (20). In response to the above, the SDH model proposed by the WHO highlights the importance of thoroughly analyzing socioeconomic inequalities (21) to reorient effective actions and interventions at a global level in public health aimed at reducing cancer mortality (22). We assume the analysis of educational inequalities to provide useful input for public programs and policies (12).
In this article, an analysis of trends in educational inequalities in gastric cancer mortality by sex and age group in Colombia from 1998 to 2015 was conducted.

Materials and methods

Research Design
An ecological time series study was conducted using Colombian vital statistics and official population estimations.

Geographical context
Colombia is an Upper Middle-Income Country (UMIC) located in South America (23). The population of the country increased during the studied period (1998–2015) from 38,088,000 to 46,314,000 (24).
Colombian geography is highly variable, including two coastal regions, the north extremity of the Andes range, and plains and forests with enormous biodiversity and natural resources (25), but also with regions with very limited geographical access where the state has very little control, and, consequently, large areas of the country are dominated by various armed criminal groups funded by drug trafficking and other illegal activities (26).
The quality and accessibility of education in Colombia are also low, with private education being significantly superior to state education (27). In the studied period, the share of the population with primary or lower educational levels steadily decreased, while participation in both secondary and tertiary education increased, with a much higher increase in those with secondary education (28).
Finally, most of the total mortality in Colombia during the recent decades is driven by chronic diseases, mainly cardiovascular diseases, and secondly by Injuries due to external causes, predominantly homicide, suicide and traffic accident (1).

Variables of interest
For the analysis, data from four variables were considered: year, sex, five-year age, and educational level, which was the socioeconomic status (SES) explanatory variable of this study.
Age was grouped into three categories: young adults (25–44 years), middle-aged adults (45–64 years of age), and older adults (65 years of age and over).
Educational levels were classified into three categories (according to the last grade level coursed): primary (up to primary), secondary (high school), and tertiary (higher education: technical, technological, and university).

Study population (deaths)
Deaths by malignant stomach tumors (C16 according to tenth revision of the International Classification of Diseases - ICD 10), were introduced in the study. The main inclusion criterion was that the age of death had been at least 25 years of age since it has been observed that Colombians, in general, have completed their full educational cycle by that age (29).
The data used were taken from national mortality records between 1998 and 2015, obtained from official mortality records and population projections from the National Administrative Department of Statistics (referred to by its Spanish acronym DANE).

Population data
The yearly population data were obtained from the DANE population projections by year, sex, and 5-year age, based on censuses and surveys. Yearly educational levels by these variables were calculated in three stages:
1. The proportion of individuals according to educational level (distributed according to the five-year age group, sex, and year) was obtained from the National Demographic and Health Survey (DHS) corresponding to the years 1995, 2000, 2005, and 2010 [REF].
2. The annual population size for each DHS educational level was obtained by multiplying the percentage of individuals in each educational category by the population count of the annual projections of the DANE census.
3. Smoothed interpolations were performed on the remaining years to obtain annual population counts using the PASEX demographic software.

Analyses
The information about the sex and age variables was found to be available in almost 100% of cases, while for educational level, 17% (men) and 16% (women) were missing counts of death by stomach cancer. The aforementioned limitation could lead to a possible overestimation of the educational inequalities due to the potentially higher rates of missing education for lower-educated individuals. However, we imputed missing values for educational level using a multiple imputation process through the IVEWARE (30) package, based on a broad set of demographic variables: age, sex, marital status, region, and urban/rural residence, thereby limiting the potential impact of this source of bias.
The Age Standardized Mortality Rates expressed per 100,000 person-years (ASMR) were estimated by sex, educational level, and year, using the direct method and the WHO 1997 standard (31).
The relative inequalities in mortality were evaluated by calculating the Relative Risk (RR) by educational group, with the highest educational level as a reference, and the Relative Inequality Index (RII) with independent Poisson regression models. The number of deaths was used as the dependent variable, and the natural logarithm of the person-year as the offset variable (32). Age and educational levels were used as independent variables through a regression of mortality at the midpoint of the cumulative distribution of education, taking into account the population size of each group and comparing the risk of the educational level with the lowest versus the highest risk (33). The figures, calculated using the Poisson regression model, are presented with confidence intervals (?=0.05). To evaluate changes over time in the ASMR, the annual percentage change (APC) was calculated based on a Poisson model that incorporated an interaction between educational level and year (34). The verification of the statistical significance of the changes was performed using the Joinpoint regression analysis, using the Monte Carlo permutation method to identify the best reference point (inflection point), where the rate of increase or decrease changes significantly (35).
Statistical analyses were performed in SAS® version 9.4. The level of significance used was p<0.05.

Ethical statement:
In accordance with resolution 8430 of 1993 issued by the Colombian Ministry of Health, this study is considered of “No risk.”

Results
During the study period, 80,520 deaths from gastric cancer were recorded in people over 25 years of age, of which 60% (n=46,570) corresponded to men and the remaining 40% (n=31,851) to women (Table 1).
According to age-standardized mortality rates (ASMR per 100,000 person-years), Table 1 shows that men were the most affected group, with 26.4/100,000 versus 15.5/100,000 for women. When discriminating by age group, it is observed that the group with the most representative values was older men (ASMR 110.6/100,000), with older women following far behind (ASMR 68.4/100,000). Generally, the highest ASMRs were concentrated in the oldest age groups throughout the study period.
When evaluating the information related to educational level (Table 1), higher ASMRs were evidenced in the lower educational levels (men 30.3/100,000; women 17.4/100,000). This behavior was observed in all age groups.
In general, the ASMR (Figure 1) shows a downward trend in both sexes. In the period 1998–2012, men went from 33.5/100,000 to 22.0/100,000; from here on, the ASMRs increased slightly. Regarding women (1998–2011), the ASMRs went from 19.7/100,000 to 12.0/100,000. Unlike men, the rates remained constant until the end of the period, an effect reflected in the Joinpoint, capturing these two points of inflection.
In the analysis of ASMR trends by age groups (Figure 2), there is a clear decrease in the rates for both sexes, in the middle-aged adult groups, and particularly in older adults, with a significant decrease in men for 2004–2011 of 4.9% APC and in women of 7.7% APC.
Regarding the behavior of the ASMR according to the educational level (Figure 3), it is evident that the tertiary level for both sexes presented greater decreases compared to the other two levels (until 2010), being significant for both sexes. A larger gap can be observed for the entire period between the ASMRs of the group of women: for the first segment (1998–2013) APC-tertiary=+ 6.3% against APC-secondary=?3.4%, and for the second segment (2000–2010) APC-primary=?4.1% against APC-tertiary=+ 11% between 2013 and 2015. The Joinpoint analysis showed a final period of increase for both sexes at the secondary and tertiary educational levels. All these values were significant.
Regarding the RR, it was higher among young men, showing a decrease in older adults (65 years of age and over). Unlike men, inequality in women grew with age, going from 2.26 (95%CI: 2.22–2.34) in the group of young women to 2.89 (95%CI: 2.79–2.93) in older women (Table 1).
Regarding the RII, the joinpoint analysis (Figure 4) presents a turning point for each sex: between 1998 and 2015, the increase in APC was + 1.0% for men, and between 2005 and 2013, the decrease in APC was ?1.8% for women, thus showing a small but significant gap between the inequality of men versus women. All the exposed values are statistically significant.

Discussion
This study found a reduction in gastric cancer adult (25+ years of age) mortality rates in Colombia during the study period (1998–2015), with much higher mortality rates and relative inequalities among men, particularly the poor in informal employment and covered by the healthcare-subsidized insurance scheme, since 2003. The result is similar to that found in international studies. This progressive reduction could be attributed to the reduction in access to healthcare barriers, concomitant to the progressive achievement of universal health insurance coverage since 2010. On the other hand, the relative educational inequalities in gastric cancer mortality did not reduce during the period, which can be partially appointed to persistent limited access to endoscopic early diagnosis and timely treatment. Older lower-educated men bore the very highest toll of mortality, similar to the findings of other studies, and in the higher increase in relative inequalities.

Comparison with previous studies
Gastric cancer ASMR trends in Colombia (1998-2015) found in this study are in line with worldwide evidence since most countries have presented mortality decreases in stomach cancer during recent decades (36,37,38). For instance, a study in different European settings (Denmark, Finland, metropolitan France, the Netherlands, Norway, Sweden, England, and Wales), found a significant average annual gastric cancer mortality decrease of 4.2% between 1980 and 2005 (39) greater than in this study: -2.4% among men and -3.2% among women. This difference reveals that Colombia has to improve actions to further reduce gastric cancer mortality among adults.
When we compare the studied sex/age groups (young, middle-aged, and senior men and women), less-educated men, particularly the elderly, have the highest mortality rates and the lowest reduction in both adjusted mortality rates and relative inequalities. This coincides with a study conducted on gastric cancer morbidity in Korea that found that among elderly men (60–69 years old), the lowest income group had a higher mortality ratio if compared with the highest income group (40).

Explanations of the results
Health insurance coverage greatly increased in Colombia, particularly since 2003, due to a reform in the decentralization regulations, achieving universal coverage in health insurance since 2010 (18), using more than 95% of the population with health insurance coverage as a reference, accompanied by important improvements in funding and institutional strengthening (41). This probably contributed to the continuous reduction in adult gastric cancer mortality rates in the study period. In general, adult gastric cancer mortality rates reached a peak in 2000 and, from then on, a decrease parallel to the increase in insurance coverage since 2003 and, later, to the growth in healthcare services coverage since 2008 for those subsidized. This is particularly true for those identified as more vulnerable, affiliated to a subsidized insurance scheme (42,43). In contrast, limitations in access to preventive healthcare services for cancer prevention remain, in particular regarding early diagnosis and timely treatment for poor and rural populations (44).
Therefore, despite the decreasing trend of ASMRs in stomach cancer mortality, the relative educational inequalities did not reduce significantly during the study period. This is due to the ASMRs in the lowest educational groups decreasing slowly, while those with higher educational levels decreased more quickly or remained stable with insignificant increases. A previous study found similar trends in the Colombian population from 1998–2003, with a significant decrease in gastric cancer mortality rates, observing particularly strong and statistically significant decreases among men with primary educational levels (45).
It is important to underline that endoscopic early diagnosis for gastric cancer has been included in the National Health Benefit Plan for those in the subsidized regime a decade later (2008) and those in the contributory scheme, i.e., those with formal employment (46). In fact, the survival rates for stomach cancer had been analyzed in previous research, and it was found that approximately 65% of the population that survived were in the contributory regime (47). In the same vein, and in line with our finding of a higher burden of gastric cancer mortality and educational inequalities among less-educated senior men, a study conducted in several Central Latin-American countries geographically close to Colombia concluded that patients who received timely treatment were more likely to survive, unlike patients with low incomes, particularly those over 55, with a lower probability of receiving treatment (48). Comparable results were obtained in Korea (1998–2012), identifying persistent healthcare access barriers characterized by a manifest inequality in access to gastric cancer detection among lower-income quintiles versus the highest quintiles (49). Also, in line with our study, higher mortality rates particularly among senior men in China were found (50).
According to the results obtained, it can also be hypothesized that the decrease observed during the entire period of the ASMRs among men and women with lower educational levels, particularly among young and adult age groups, could be explained by timely access to universal healthcare insurance coverage, as well as the commitment of reducing access barriers linked to the shared payments of the subsidized regime (51), and by the strengthening of surveillance, detection, timely treatment, and other actions contemplated in the public health plan comprehensive approach to cancer control (52).

Limitations of the study
There are two major limitations in this study: Firstly, the ecological bias inherent in the design of the study, for which the estimated associations cannot be interpreted as causal effects. However, the ample size of the dataset contributes to reducing the chance of ecological error (53).
Secondly, although Colombia has an appropriate Vital Statistics registration, the results of the study depend on the completeness and quality of the cause of death records in the use of secondary data.
Regarding the completeness of the dataset, non-coverage in Colombia reached 9.9% in 2002, as referenced in a study (54). A previous study, which used life tables as a reference to calculate under-registration in the Colombian deaths dataset, found that this is greater among less prosperous subnational entities (32 in total, the so-called "departments"), and that under-registration noticeably decreased during the studied period (2000–2010) (55). Another study confirmed that under-registration is highest in the poorest regions (56). These findings suggest that estimates of disparities in mortality by educational level are likely to have been underestimated. This may also have led to an underestimation of the extent to which inequalities have increased because under-registration decreased over the study period (55).
Regarding the size of the garbage codes (R00–R99) in the Colombian mortality dataset, a comparative study showed that garbage codes in Colombia in 2015 (22.9%) were even lower than those of the average of highly-developed countries (23.8%), and much lower than those in middle-developed countries (39.7%)(57). According to our calculations (web appendix Table 1), the percentage of malignant neoplasms garbage codes (C76–C80) for the study period was 14.4%, and the percentage of the garbage codes in the overall dataset (R00–R99) was 34.9%. These percentages for both malignant neoplasms and the total of deaths were greatly reduced during the study period (results not shown). As the percentage of garbage codes tend to be higher in less developed settings (48), the garbage codes in the Colombian dataset also suggest that our results underestimate the educational inequalities and their increase.
Finally, the data corresponding to mortality come from a different source than the population distribution by educational group and could have given rise to a numerator-denominator bias.

Conclusions
While stomach cancer mortality rates reduced during the study period, slow reductions in mortality among the less-educated groups contributed to upholding the social inequality gaps. These results serve as input for Colombian health authorities in the work of planning, reorientation prevention (at all levels), and health education aimed at strengthening the quality of life of the populations diagnosed (or identified with risk) of stomach cancer of all socioeconomic levels. The differences found in the ASMR and RR for stomach cancer according to age group, sex, and educational level suggest the need to propose specific interventions, considering specific population characteristics.
On the other hand, the gap identified between the RII of men versus women leads us to analyze the orientation of public health programs (classically oriented to the promotion of protective behaviors in the female population), suggesting an orientation of the promotion of health from a gender perspective. This is advantageous since specific interventions can have differential effects on educational inequalities for men and women.
?
Conflict of interest:
The authors have no conflict of interest.

?
References

1. Wang H, Naghavi M, Allen C, Barber RM, Carter A, Casey DC, et al. Global, regional, and national life expectancy, all-cause mortality, and cause-specific mortality for 249 causes of death, 1980–2015: a systematic analysis for the Global Burden of Disease Study 2015. The Lancet. 2016;388(10053):1459–544.
2. Thrumurthy SG, Chaudry MA, Hochhauser D, Mughal M. The diagnosis and management of gastric cancer. BMJ (Online). 2013;347(November):1–9.
3. McGuire S. World Cancer Report 2014. Advances in Nutrition. 2016;7(2):418–9.
4. Fitzmaurice C, Allen C, Barber RM, Barregard L, Bhutta ZA, Brenner H, et al. Global, regional, and national cancer incidence, mortality, years of life lost, years lived with disability, and disability-adjusted life-years for 32 cancer groups, 1990 to 2015: A Systematic Analysis for the Global Burden of Disease Study Global Burden. JAMA Oncol. 2017;3(4):524–48.
5. Ferlay J, Soerjomataram I, Dikshit R, Eser S, Mathers C, Rebelo M, et al. Cancer incidence and mortality worldwide: Sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer. 2015;136(5):E359–86.
6. Allemani C, Weir HK, Carreira H, Harewood R, Spika D, Bannon F, et al. Global surveillance of cancer survival 1995–2009: analysis of individual data for 25 676 887 patients from 279 population-based registries in 67 countries (CONCORD-2). Lancet. 2015;385(9972):977–1010.
7. Gelband H, Jha P, Gauvreau CL, Horton S. Summary. In: Cancer: Disease Control Priorities. 3rd ed. The World Bank; 2015. p. 113–34.
8. Instituto Nacional de Cancerología. Análisis de Situación del Cáncer en Colombia [Internet]. 2017 [cited 2019 Dec 18]. Available from: https://www.cancer.gov.co/Situacion_del_Cancer_en_Colombia_2015.pdf
9. Stringhini S, Berkman L, Dugravot A, Ferrie JE, Marmot M, Kivimaki M, et al. Socioeconomic status, structural and functional measures of social support, and mortality: The British Whitehall II Cohort Study, 1985-2009. Am J Epidemiol. 2012;175(12):1275–83.
10. Glied S, Lleras-muney A. Technological Innovation and Inequality in Health. 2008;45(3):741–61.
11. World Health Organization. Reducir las inequidades sanitarias actuando sobre los determinantes sociales de la salud. Resolución WHA62.14. 62a Asamblea Mundial de la Salud. 2009;1–5.
12. WHO, Commission on Social Determinants of Health. Closing the gap in a generation. Health equity through action on the social determinants of health. 2008;
13. Sen A. Development as freedom. New York: Oxford University Press; 2001. 1–365 p.
14. Krieger N. Epidemiology and the People’s Health: Theory and Context.
15. Fiva JH, Hægeland T, Rønning M, Syse A. Access to treatment and educational inequalities in cancer survival. J Health Econ [Internet]. 2014;36(1):98–111. Available from: http://dx.doi.org/10.1016/j.jhealeco.2014.04.001
16. Deaton A. La salud en el mundo moderno. In: El gran escape: salud, riqueza y los orígenes de la inequidad. México D.F: Fondo de Cultura Economica / Mexico; 2015.
17. Fryers T, Melzer D, Jenkins R, Brugha T. The distribution of the common mental disorders: Social inequalities in Europe. Clinical Practice and Epidemiology in Mental Health. 2005;1:1–12.
18. de Vries E, Arroyave I, Pardo C. Time trends in educational inequalities in cancer mortality in Colombia, 1998–2012. BMJ Open [Internet]. 2016 Apr 5 [cited 2019 Dec 2];6(4):e008985. Available from: http://bmjopen.bmj.com/lookup/doi/10.1136/bmjopen-2015-008985
19. Hoffmann R, Hu Y, De Gelder R, Menvielle G, Bopp M, Mackenbach JP. The impact of increasing income inequalities on educational inequalities in mortality - An analysis of six European countries. Int J Equity Health [Internet]. 2016;15(1):1–12. Available from: http://dx.doi.org/10.1186/s12939-016-0390-0
20. Observatorio Nacional de Salud. Sexto informe ONS, Informe Nacional sobre Inequidades Sociales en Salud en Colombia. 2015;1–366.
21. Solar O, Irwin A. A Conceptual Framework for Action on the Social Determinants of Health. Social Determinants of Health Discussion Paper 2 (Policy and Practice). 2010. p. 79.
22. Braveman P, Gottlieb L. The social determinants of health: It’s time to consider the causes of the causes. Public Health Reports. 2014;129(SUPPL. 2):19–31.
23. World Bank. World Development Indicators (WDI). World Bank Data (WBD) 2020 March 2, 2011; Colombia - World Bank Group - country indicators]. [Internet]. Available from: https://data.worldbank.org/country/colombia
24. Departamento Administrativo Nacional de Estadística - DANE. Proyecciones de población por sexo y edades simples hasta 80 años y más, a nivel nacional y departamental. 2005-2020.
25. Fundación Universitaria del Área Andina. Geografía física de Colombia. 2017.
26. Bulla P, Guarín S. Rural security in Colombia: An opportunity for state consolidation. Stability. 2015;4(1).
27. Abril L. Access policy and social justice in higher education: the Colombian case. In: Fairness in Access to Higher Education in a Global Perspective Reconciling Excellence, Efficiency, and Justice. 2013. p. 111–28.
28. The World Bank. Education Statistics: Education Attainment.
29. UNESCO-UIS. "Browse by theme: Education," Data Center, UNESCO Institute for Statistics [Internet]. 2015 [cited 2020 Apr 27]. Available from: http://uis.unesco.org/en/country/co
30. Horton NJ, Kleinman KP. Much ado about nothing: A comparison of missing data methods and software to fit incomplete data regression models. Am Stat. 2007;61(1):79–90.
31. Bray F, Ferlay J. Age standardization. In: Forman D, Bray F, D.H. Brewster D, Gombe Mbalawa C, Kohler B, Piñeros M, et al., editors. Cancer Incidence in Five Continents Vol X. Lyon, France 2014; 2014. p. 112–5.
32. Wagstaff A, Paci P, Doorslae E Van. On the measurement of inequalities health. Sm Sci Med. 1991;33(5):545–57.
33. Mackenbach JP, Kunst AE. Measuring the magnitude of socio-economic inequalities in health: An overview of available measures illustrated with two examples from Europe. Soc Sci Med. 1997;44(6):757–71.
34. Moreno-Betancur M, Latouche A, Menvielle G, Kunst AE, Rey G. Relative Index of Inequality and Slope Index of Inequality: A Structured Regression Framework for Estimation. Epidemiology. 2015;26(4):518–27.
35. Kim HJ, Fay MP, Feuer EJ, Midthune DN. Permutation tests for joinpoint regression with applications to cancer rates. Stat Med. 2000;19(3):335–51.
36. Ferlay J, Steliarova-Foucher E, Lortet-Tieulent J, Rosso S, Coebergh JWW, Comber H, et al. Cancer incidence and mortality patterns in Europe: Estimates for 40 countries in 2012. Eur J Cancer [Internet]. 2013;49(6):1374–403. Available from: http://dx.doi.org/10.1016/j.ejca.2012.12.027
37. Yako-suketomo H, Katanoda K. Comparison of time trends in stomach cancer mortality (1990-2006) in the World, from the WHO mortality database. Jpn J Clin Oncol. 2009;39(9):622–3.
38. Healthdata.org. GBD Compare | IHME Viz Hub.
39. Amiri M, Janssen F, Kunst AE. The decline in stomach cancer mortality: Exploration of future trends in seven European countries. Eur J Epidemiol. 2011;26(1):23–8.
40. Kim NY, Oh JS, Choi Y, Shin J, Park EC. Relationship between socioeconomic status and accessibility for endoscopic resection among gastric cancer patients: using National Health Insurance Cohort in Korea: poverty and endoscopic resection. Gastric Cancer. 2017;20(1):61–9.
41. Espinal-Piedrahita JJ, Restrepo-Zea JH. Financiamiento del sistema de salud colombiano: Tendencias y desafíos. Revista de Salud Pública. 2022 Jan 1;24(1):1–8.
42. Homedes N, Ugalde A. Las reformas de salud neoliberales en América Latina: una visión crítica a través de dos estudios de caso. Revista Panamericana de Salud Pública. 2005;17(3):210–20.
43. Giedion U, Uribe MV. Colombia’s Universal Health Insurance System. Health Aff. 2009;28(3):853–63.
44. de Vries E, Buitrago G, Quitian H, Wiesner C, Castillo JS. Access to cancer care in Colombia, a middle-income country with universal health coverage. J Cancer Policy. 2018 May;15:104–12.
45. De Vries E, Arroyave I, Pardo C, Wiesner C, Murillo R, Forman D, et al. Trends in inequalities in premature cancer mortality by educational level in Colombia, 1998-2007. J Epidemiol Community Health (1978). 2015;69(5):408–15.
46. Ministerio de Salud. Aseguramiento del sistema general de salud [Internet]. [cited 2023 Jul 5]. Available from: https://www.minsalud.gov.co/proteccionsocial/Regimensubsidiado/Paginas/aseguramiento-al-sistema-general-salud.aspx
47. de Vries E, Uribe C, Pardo C, Lemmens V, Van de Poel E, Forman D. Gastric cancer survival and affiliation to health insurance in a middle-income setting. Cancer Epidemiol [Internet]. 2015;39(1):91–6. Available from: http://dx.doi.org/10.1016/j.canep.2014.10.012
48. Estevez-Ordonez D, Montalvan-Sanchez EE, Wong RE, Montalvan-Sanchez DM, Rodriguez-Murillo AA, Dominguez RL, et al. Health barriers and patterns of gastric cancer care in rural Central American resource-limited settings. JAMA Oncol. 2018;4(8):1131–3.
49. Kim S, Hwang J. Assessment of trends in socioeconomic inequalities in cancer screening services in Korea, 1998-2012. Vol. 15, International Journal for Equity in Health. 2016.
50. Yang L, Zheng R, Wang N, Yuan Y, Liu S, Li H, et al. Incidence and mortality of stomach cancer in China, 2014. Chinese Journal of Cancer Research. 2018;30(3):291–8.
51. Vega W, Guzmán Ó. Conceptos sobre copagos y cuotas moderadoras en el régimen contributivo [Internet]. Monitor Estratégico. Bogotá; 2012. p. 01–7. Available from: https://www.minsalud.gov.co/sites/rid/Lists/BibliotecaDigital/RIDE/IA/SSA/Conceptos-sobre-copagos-cuotas-moderadoras.pdf
52. Orjuela, Alexandra, González, Lyda, Marulanda J. Derechos en salud de los pacientes con cáncer [Internet]. Serie derechos de los usuarios en salud en seguridad social. 2015. 12–110 p. Available from: http://www.defensoria.gov.co/public/pdf/Cartilla_pacientes_Cancer.pdf
53. Cohen BL. Ecological versus Case-Control Studies for Testing a Linear-No Threshold Dose-Response Relationship. Int J Epidemiol. 1990;19(3):680–4.
54. Bermúdez Quintero CD, Bermúdez Quintero SL. Sistema de predicción de estadísticas de nacimientos y defunciones en Colombia soportadas por TIC. 2019.
55. Arroyave I, Hessel P, Burdorf A, Rodriguez-Garcia J, Cardona D, Avendaño M. The public health impact of economic fluctuations in a Latin American country: mortality and the business cycle in Colombia in the period 1980–2010. Int J Equity Health. 2015 Dec 27;14(1):48.
56. Rodríguez García J. Inequidades socioeconómicas entre departamentos y su asociación con indicadores de mortalidad en Colombia en 2000. Panam Salud Publica. 2007;21(2–3):111–24.
57. Iburg KM, Mikkelsen L, Adair T, Lopez AD. Are the cause of death data fit for purpose? Evidence from 20 countries at different levels of socio-economic development. PLoS One. 2020 Aug 24;15(8):e0237539.