Cancer Epidemiology 41 (2016) 24–33
Contents lists available at ScienceDirect
Cancer Epidemiology The International Journal of Cancer Epidemiology, Detection, and Prevention journal homepage: www.cancerepidemiology.net
National and regional breast cancer incidence and mortality trends in Mexico 2001–2011: Analysis of a population-based database Enrique Soto-Perez-de-Celisa,b , Yanin Chavarri-Guerrac,* a
Cancer Care in the Elderly Clinic, Department of Geriatrics, Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán, Mexico City, Mexico Programa de Cáncer Post-Mastectomía, Instituto Nacional de Cancerología, Mexico City, Mexico Department of Hemato-Oncology, Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán, Avenida Vasco de Quiroga 15, Colonia Belisario Domínguez Sección XVI, Tlalpan, Mexico City, DF 14080, Mexico b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 12 August 2015 Received in revised form 7 January 2016 Accepted 7 January 2016 Available online xxx
Introduction: Breast cancer is the most common malignancy in Mexican women since 2006. However, due to a lack of cancer registries, data is scarce. We sought to describe breast cancer trends in Mexico using population-based data from a national database and to analyze geographical and age-related differences in incidence and mortality rates. Methods: All incident breast cancer cases reported to the National Epidemiological Surveillance System and all breast cancer deaths registered by the National Institute of Statistics and Geography in Mexico from 2001 to 2011 were included. Incidence and mortality rates were calculated for each age group and for 3 geographic regions of the country. Joinpoint regression analysis was performed to examine trends in BC incidence and mortality. We estimated annual percentage change (APC) using weighted least squares log-linear regression. Results: We found an increase in the reported national incidence, with an APC of 5.9% (95% CI 4.1–7.7, p < 0.05). Women aged 60–65 had the highest increase in incidence (APC 7.89%; 95% CI 5.5 10.3, p < 0.05). Reported incidence rates were significantly increased in the Center and in the South of the country, while in the North they remained stable. Mortality rates also showed a significant increase, with an APC of 0.4% (95% CI 0.1–0.7, p < 0.05). Women 85 and older had the highest increase in mortality (APC 2.99%, 95% CI 1.9–4.1; p < 0.05). Conclusions: The reporting of breast cancer cases in Mexico had a continuous increase, which could reflect population aging, increased availability of screening, an improvement in the number of clinical facilities and better reporting of cases. Although an improvement in the detection of cases is the most likely explanation for our findings, our results point towards an epidemiological transition in Mexico and should help in guiding national policy in developing countries. ã 2016 Elsevier Ltd. All rights reserved.
Keywords: Epidemiology Breast neoplasms Mexico Healthcare disparities
1. Introduction Breast cancer is the most common malignancy in women world wide, with nearly 1.7 million cases diagnosed in 2012, representing 25% of all female cancers [1]. In developing countries, the incidence of breast cancer is rising sharply due to changes in reproductive factors, lifestyle, and an increased life expectancy. These rising cancer rates in the developing world are projected to reach a 55%
Abbreviations: DGE, national epidemiological administration; SINAVE, national system of epidemiological surveillance; INEGI, national institute of statistics and geography; CONAPO, national population council; APC, annual percentage change. * Corresponding author. E-mail address: [email protected] (Y. Chavarri-Guerra). http://dx.doi.org/10.1016/j.canep.2016.01.007 1877-7821/ ã 2016 Elsevier Ltd. All rights reserved.
increased incidence and 58% increased mortality in fewer than 20 years [2]. Mexico is an upper middle-income country of 112, 336, 538 inhabitants [3] with one of the highest per capita incomes in Latin America. Even though Mexico is among the top 20 largest economies in the World, more than half of its population is poor [4]. The distribution of wealth in Mexico is unequal [5], with a 27fold difference between the average incomes of the top and bottom deciles of the population [6]. This inequality is also true for different regions of the country, with northern states experiencing larger economic growth due to their proximity to the United States, better infrastructure and a greater capacity for producing manufactured goods [7]. As in other developing countries, cancer incidence and mortality have been on the rise in Mexico. Since 2006, breast
E. Soto-Perez-de-Celis, Y. Chavarri-Guerra / Cancer Epidemiology 41 (2016) 24–33
cancer has been the leading cause of cancer-related mortality in Mexican women, surpassing cervical cancer [8]. A review of the regional geographic trends of breast and cervical cancer mortality between 1979 and 2006 showed a higher risk of dying from breast cancer in women from the northern states, while women from the southern states had higher mortality attributable to cervical cancer [9]. In order to tackle inequities in the access to healthcare, in 2003 the Mexican government instituted a national health insurance program directed at the poorest sectors of the population called Seguro Popular, which includes coverage of breast cancer screening, diagnosis and treatment [10]. One of the main barriers for designing strategies to tackle cancer in developing nations is the paucity of statistics regarding its incidence and mortality [11], and the Breast Health Global Initiative has included the assessment of the burden of breast cancer at the national level as one of its top priorities [2]. Due to the fact that Mexico lacks a national population-based cancer registry, information on these indicators usually comes from data extrapolated from neighboring countries, from hospital-based histopathological databases [8] or from the only regional cancer registry, which is located in the western state of Jalisco [12]. For example, data from GLOBOCAN 2012, (which reports an incidence of 35.4 cases of breast cancer per 100,000 person-years in Mexico) are not obtained from any registry, but rather extrapolated from the mortality data obtained from the death certificate database of the Instituto Nacional de Estadística y Geografía (National Institute of Statistics and Geography, INEGI) [13], which is considered of high quality. Furthermore, the previous edition of GLOBOCAN, published in 2008, obtained incidence data by using a regional model extrapolated from other territories in Central America and the Latin Caribbean with a high quality population-based cancer registry (Puerto Rico and Costa Rica) [14]. Thus, although the regional and national mortality figures attributable to breast cancer in Mexico come from reliable sources and have been published before [15], the true incidence of the disease is largely unknown and no direct data sources have been used to describe it. Starting in 2000, the Dirección General de Epidemiología (National Epidemiological Administration, DGE) of the Mexican Ministry of
25
Health instituted the weekly mandatory reporting of all new breast cancer cases as part of the Sistema Nacional de Vigilancia Epidemiológica (National System of Epidemiological Surveillance, SINAVE) [16]. SINAVE was created in the 1970’s and since 1995 it has national coverage, encompassing all the public institutions that constitute the Sistema Nacional de Salud (National Health System) and several private institutions. Data reported by SINAVE include confirmed and suspected cases of 142 different diseases, including not only breast cancer but also cervical cancer and cervical dysplasia [17]. These suspected cases are reported on a weekly basis by physicians at nurses at every medical unit or hospital belonging to the institutions included in SINAVE using an online computerized format called Sistema Único de Información para la Viligancia Epidemiologica (Unified Information System for Epidemiological Surveillance, SUIVE) [18]. The reported cases are reviewed by the Sanitary Jurisdiction to which each medical unit belongs and then reported on a state level weekly, monthly and yearly [17]. In this paper, we describe national and regional breast cancer incidence trends using population-based data from the SINAVE database and mortality trends using the INEGI records for the first decade of the 21st century, in order to understand the epidemiological variations generated by recent changes in the country, including growing access to healthcare, increasing coverage of screening methods and the quality of data reporting. 2. Patients and methods 2.1. Study area Mexico is a federation comprising 31 states and a Federal District (DF), geographically limited by United States border on the north and by the Guatemala–Belize border on the south. We divided the country into three geographic regions and 8 sub regions: North (Northwest and Northeast); Central (North-Central, South Central, West and East) and South (Southeast and Southwest). The location of each region and subregion within the country can be seen in Fig. 1.
Fig. 1. Geographic regions of Mexico.
26
E. Soto-Perez-de-Celis, Y. Chavarri-Guerra / Cancer Epidemiology 41 (2016) 24–33
the South of the country, while the North showed a stabilization of reported incidence rates.
2.2. Data sources and variables Data on the number of incident breast cancer cases reported between 2001 and 2011 were obtained from the SINAVE morbidity database [16]. Breast cancer related deaths between 2001 and 2011 were obtained from the INEGI death certificate database [13]. Incidence and mortality data were obtained for each age group and for each state, subregion and region of the country. Yearly populations were obtained from national census data and from the population growth estimates published by the Consejo Nacional de Población (National Population Council, CONAPO) [19]. Age group-specific rates and standardized (overall and truncated) rates were calculated using the direct method (using the world standard population) [20]. These were expressed as rates/100,000 persons. Only women over 15 years of age were considered for analysis. Incidence and mortality rates were calculated for each state, subregion and region. 2.3. Statistical methods We used joinpoint regression analysis to describe trends over time at the state, subregional, regional and national level, with each joinpoint indicating a statistically significant change in trend. We estimated the annual percentage change (APC) for each of those trends using weighted least squares log-linear regression and assuming a Poisson distribution [21]. The APC over the 2001–11 time period is a weighted average of the slope coefficients of the underlying joinpoint regression line with the weights equal to the length of each segment over the interval. Joinpoint analyses were performed using the ‘Joinpoint’ software from the Surveillance Research Program of the US National Cancer Institute. 3. Results From 2001 to 2011, 69,651 new breast cancer cases were reported to SINAVE (Table 1). For the same period, 48,817 deaths attributed to breast cancer were registered in the death certificate database (Table 2). 3.1. National and regional reported incidence The modeled age-standardized national breast cancer incidence rate had a significant increase, rising from 14.2/100,000 personyears (Standard Error [SE] 0.23) in 2001, to 25.2/100,000 personyears (SE 0.25) in 2011, with an APC of 5.9% (95% CI 4.1 to 7.7, p < 0.05) (Fig. 2a). Results of the joinpoint analysis of regional agestandardized incidence trends are shown in Table 3 and in Fig. 3. For the studied period, we found a significant increase in the number of new breast cancer cases reported in the Center and in
3.2. Subregional variations in reported incidence When the age-standardized breast cancer incidence rates were analyzed at the subregional level, a significant increase was seen for all subregions except the Northwest and the Northeast (Table 3). The Northeast showed no increase in incidence APC (0.7%; 95% CI 5.3–7.1), with a downward trend in the number of reported cases after 2009 in each of the three states comprising the subregion (Fig. 4). 3.3. National and regional mortality The modeled age-standardized national breast cancer mortality rate also showed an increase, rising from 14/100,000 personyears (SE 0.23) in 2001 to 14.6/100,000 person-years (SE 0.2) in 2011, with an APC of 0.4% (95% CI 0.1–0.7, p < 0.05) (Fig. 2b). Regional age-standardized mortality rates were significantly increased in all three regions of the country during the studied period, as shown in the joinpoint analysis presented in Table 3 and in Fig. 3. 3.4. Subregional variations in mortality When the breast cancer mortality rates were analyzed at the subregional level, a significant increase was seen only in the Southwest [APC 2.7% (95% CI 1.2–4.3), p < 0.05] and Center-North [APC 1.7% (95% CI 0.7–2.25), p < 0.05]. On the other hand, the Center-South subregion showed a decrease in the mortality rate, with an APC of 0.7% (95% CI 1.1– 0.3, p < 0.05). The rest of the subregions showed non-significant changes in mortality. These data are illustrated in Table 3. 3.5. Age associated trends A rise in the reported number of breast cancer cases was seen in all age groups except for women age 25 and younger. The largest increase was seen in women aged 60–65, with an APC of 7.89% (95% CI 5.5–10.3; p < 0.05), followed by women 65 and older with 6.7% (95% CI 4.9 8.6; p < 0.05). The highest observed incidence of new cases was seen in the 60–65 age group, with 62.82 cases per 100,000 women reported in 2011. During the studied period mortality rates in the younger age groups (50 years old) either decreased or remained stable, while older groups (60 years old) showed an increase (Table 4). Women older than 85 showed the highest APC in mortality of any age group (2.99%, 95% CI 1.9–4.1; p < 0.05), and the highest observed mortality rate for 2011 (88.72 deaths per 100,000). Trends according to age group can be seen in Table 4.
Table 1 Total number of incident breast cancer cases reported by SINAVE for each subregion of the country during the studied period. Region
Southwest Southeast East West Center-North Center-South Northwest Northeast Total
Year 2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
166 111 500 780 583 367 774 690 3971
267 122 601 820 678 534 996 710 4728
230 85 499 696 974 497 846 628 4455
189 78 561 827 920 748 957 738 5018
239 26 532 966 1125 733 825 677 5123
288 200 693 1008 1214 726 987 927 6043
337 222 1234 1204 1730 755 873 975 7330
453 253 1290 1164 1950 785 1133 902 7930
463 258 1418 1130 1764 862 1177 1356 8428
576 335 1591 1427 1795 716 1293 652 8385
430 431 1922 1501 1824 835 1352 597 8892
E. Soto-Perez-de-Celis, Y. Chavarri-Guerra / Cancer Epidemiology 41 (2016) 24–33
27
Table 2 Total number of Breast Cancer deaths reported by INEGI for each subregion of the country during the studied period. Region
Southwest Southeast East West Center-North Center-South Northwest Northeast Total
Year 2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
179 125 436 458 326 1111 510 455 3600
180 127 467 543 355 1134 541 508 3855
231 117 486 524 342 1155 577 452 3884
238 155 495 543 421 1202 616 503 4173
226 135 491 590 389 1228 650 520 4229
250 147 531 662 433 1231 666 537 4457
246 138 592 580 468 1283 699 598 4604
291 150 569 627 489 1355 770 579 4830
268 180 620 660 492 1335 736 614 4905
323 190 590 682 491 1392 751 640 5059
313 192 626 710 535 1429 773 643 5221
Fig. 2. National incidence (a) and mortality (b) trends 2001–2011, joinpoint analysis.
4. Discussion Breast cancer in Mexico presented a continuous increase in the reported incidence and mortality rates between 2001 and 2011. A significant increase in both rates was seen in Central and Southern regions and subregions of the country. On the other hand, the reported incidence and mortality rates in the North remained unchanged, with a downturn in more recent years. Women 60– 69 years old showed the largest rise in reported incidence, while women 60 years of age and older were the group with the highest increase in mortality. 4.1. Population aging The observed increase in the incidence of breast cancer could reflect population growth, aging and lifestyle modifications seen in developing countries transitioning to a more “developed” epidemiological landscape. Life expectancy in Mexico doubled during the second half of the 20th century, and by 2005 the life expectancy of Mexican women was 77.9 years. Although in the year 2000 older adults represented only 7% of the population, this proportion is expected to increase to 12.5% by 2020 and to 28% by 2050, owing to the aging of people born during the Mexican economic expansion of 1960–1980 (the Mexican baby boomers) [19]. This will inevitably lead to an increase in the incidence of diseases associated with aging, including breast cancer. Additionally, since women in the 50–59 and 60–64 age group are specially targeted by breast cancer screening programs, this is expected to rise even further. As a matter of fact, an increase in
the incidence and mortality of breast cancer in women 60 years of age and older seems to be the main factor driving the upward trends we observed. These changes might lead to an increase in the median age at diagnosis for breast cancer patients in Mexico, which has historically been reported to be of 50 years of age [8]. Another interesting observation is the fact that the number of new cases reported in women 65 years and older is lower than that in the 50–59 and 60–64 subgroups, in contrast to what is seen in other countries like the United States [22]. Although there is no clear explanation for this, it may be due to the fact that older women in Mexico have lower participation in screening programs or that they fail to seek medical attention for breast symptoms. 4.2. Changes in access and availability of healthcare facilities An important landmark which could partially explain some of the changes seen in our study is the aforementioned establishment of the Seguro Popular program in 2003. In the first 9 years after its inception, 52.6 million Mexicans, previously uninsured, were incorporated into the program [10]. Currently, the public sector accounts for 88% of the available medical facilities, while the private sector is responsible for the remaining 12% [23]. 44% of the population is covered by Social Security (for salaried workers) and 45% is covered by Seguro Popular, while 7% are covered by other public schemes (directed at government employees) and less than 3% have private insurance [24]. From 2003 to 2013, the number of existing medical facilities (including both ambulatory clinics and hospitals) in the public sector grew from 19,505 to 22,228,
28
E. Soto-Perez-de-Celis, Y. Chavarri-Guerra / Cancer Epidemiology 41 (2016) 24–33
Table 3 Modeled regional age - s tandardized incidence and mortality rates for women 15 years of age in Mexico, 2001–2011. Region
Rate
Southwest
Incidence. Mortality Incidence Mortality Incidence Mortality Incidence Mortality Incidence Mortality Incidence Mortality Incidence Mortality Incidence Mortality Incidence Mortality Incidence Mortality Incidence Mortality Incidence Mortality
Southeast South East West Center-North Center-South Center Northwest Northeast North National
a b c d
Modeled ASR per year/100,000 women (WHO Standard Population) [13,14]
APC (2001–11).
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
6.43 7.04 6.11 9.58 6.10 7.79 9.32 10.35 20.94 14.87 16.82 11.90 8.45 16.79 12.88 13.95 23.10 16.36 21.86 18.41 24.24 17.52 14.17 14.03
7.00 7.23 6.88 9.66 6.74 7.95 10.55 10.42 21.91 15.03 17.31 12.10 9.70 16.67 13.86 13.99 23.51 16.69 23.06 18.51 24.49 17.54 15.00 14.08
7.62 7.43 7.76 9.73 7.45 8.11 11.94 10.50 22.93 15.19 17.83 12.30 11.14 16.56 14.92 14.04 23.94 17.03 24.34 18.60 24.74 17.57 15.89 14.14
8.29 7.63 8.75 9.81 8.23 8.28 13.52 10.57 23.99 15.36 18.35 12.51 12.79 16.45 16.06 14.08 24.37 17.38 25.69 18.70 24.99 17.59d 16.83 14.20
9.02 7.84 9.86 9.89 9.10 8.45 15.30 10.64 25.10 15.52 18.89 12.72 14.68 16.33 17.29 14.13 24.81 17.73 27.11 18.79 25.25 18.00 17.83 14.25
9.81 8.05 11.11 9.97 10.05 8.63 17.32 10.72 26.27 15.69 19.45 12.94 16.85 16.22 18.61 14.18 25.26 18.09 28.60 18.89 25.51 18.41 18.88 14.31
10.67 8.27 12.53 10.04 11.11 8.81 19.61 10.79 27.49 15.86 20.02 13.15 19.34 16.11 20.03 14.22 25.71 18.45 30.19 18.98 25.77 18.84 20.00 14.36
11.61 8.50 14.12 10.12 12.28 8.99 22.19 10.87 28.76 16.03 20.61 13.37 22.20a 16.00 21.56 14.27 26.18 18.83c 31.85 19.08 26.03 19.28d 21.18 14.42
12.63 8.73 15.92 10.20 13.57 9.18 25.12 10.95 30.10 16.20 21.22 13.60 20.96 15.89 23.20 14.31 26.65 18.32 33.62b 19.18 26.30 18.85 22.44 14.48
13.75 8.97 17.94 10.28 14.99 9.37 28.44 11.02 31.49 16.38 21.85 13.83 19.79 15.79 24.97 14.36 27.13 17.82 22.59 19.28 26.56 18.43 23.76 14.53
14.96 9.21 20.22 10.37 16.57 9.56 32.19 11.10 32.96 16.55 22.49 14.06 18.69a 15.68 26.88 14.41 27.62 17.34c 15.18b 19.37 26.84 18.02d 25.17 14.59
Center-South. Incidence. 2001–2008 APC 14.8% (95% CI 9.7–20.2, p < 0.05). 2008–2011 APC 5.6% (95% CI 16.8–7.1, p NS). Northeast. Incidence. 2001–2009 APC 5.5% (95% CI 1.6–9.6, p < 0.05). 2009–2011 APC 32.8 (95% CI 1.1–54.3, p < 0.05). Northwest. Mortality. 2001–2008 APC 2% (95% CI 1.1–2.9, p < 0.05). 2008–2011 APC 2.7% (95% CI 5.5–0.2, p NS). North. Mortality. 2001–2004 APC 0.1% (95% CI 2.8–3.1, p NS). 2004–2008 APC 2.3% (95% CI 0.3–5.1, p NS). 2008–2011 APC
representing a 14% increase[23]. Likewise, the number of available medical offices in the public sector grew from 51,598 in 2001 to 70,196 in 2011 [25]. Regarding available personnel in the public sector, the number of physicians (regardless of their specialization) grew from 112,681 in 2001 to 170,565 in 2011, and the number of nurses went from 192,046 to 250,931 during the same period [25].
8.8 2.7 12.7 0.8 10.5 2.1 13.2 0.7 4.6 1.1 2.9 1.7 8.5 0.7 7.6 0.3 1.8 0.8 0.7 0.5 1.0 0.7 5.9 0.4
2.2% (95% CI
95% CI.
4.3–13.5 1.2–4.3 5.9–20.0 1.0–2.6 6.0–15.2 1.0–3.2 9.3–17.3 0.2–1.6 2.7–6.7 0.1–2.2 0.2–6.2 0.7–2.7 4.3–12.8 1.1– 0.3 5.7–9.6 0–0.6 0.5–4.2 0.1–1.7 5.3–7.1 0.3–1.3 1.6–3.7 0–1.3 4.1–7.7 0.1–0.7
p
Contents lists available at ScienceDirect
Cancer Epidemiology The International Journal of Cancer Epidemiology, Detection, and Prevention journal homepage: www.cancerepidemiology.net
National and regional breast cancer incidence and mortality trends in Mexico 2001–2011: Analysis of a population-based database Enrique Soto-Perez-de-Celisa,b , Yanin Chavarri-Guerrac,* a
Cancer Care in the Elderly Clinic, Department of Geriatrics, Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán, Mexico City, Mexico Programa de Cáncer Post-Mastectomía, Instituto Nacional de Cancerología, Mexico City, Mexico Department of Hemato-Oncology, Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán, Avenida Vasco de Quiroga 15, Colonia Belisario Domínguez Sección XVI, Tlalpan, Mexico City, DF 14080, Mexico b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 12 August 2015 Received in revised form 7 January 2016 Accepted 7 January 2016 Available online xxx
Introduction: Breast cancer is the most common malignancy in Mexican women since 2006. However, due to a lack of cancer registries, data is scarce. We sought to describe breast cancer trends in Mexico using population-based data from a national database and to analyze geographical and age-related differences in incidence and mortality rates. Methods: All incident breast cancer cases reported to the National Epidemiological Surveillance System and all breast cancer deaths registered by the National Institute of Statistics and Geography in Mexico from 2001 to 2011 were included. Incidence and mortality rates were calculated for each age group and for 3 geographic regions of the country. Joinpoint regression analysis was performed to examine trends in BC incidence and mortality. We estimated annual percentage change (APC) using weighted least squares log-linear regression. Results: We found an increase in the reported national incidence, with an APC of 5.9% (95% CI 4.1–7.7, p < 0.05). Women aged 60–65 had the highest increase in incidence (APC 7.89%; 95% CI 5.5 10.3, p < 0.05). Reported incidence rates were significantly increased in the Center and in the South of the country, while in the North they remained stable. Mortality rates also showed a significant increase, with an APC of 0.4% (95% CI 0.1–0.7, p < 0.05). Women 85 and older had the highest increase in mortality (APC 2.99%, 95% CI 1.9–4.1; p < 0.05). Conclusions: The reporting of breast cancer cases in Mexico had a continuous increase, which could reflect population aging, increased availability of screening, an improvement in the number of clinical facilities and better reporting of cases. Although an improvement in the detection of cases is the most likely explanation for our findings, our results point towards an epidemiological transition in Mexico and should help in guiding national policy in developing countries. ã 2016 Elsevier Ltd. All rights reserved.
Keywords: Epidemiology Breast neoplasms Mexico Healthcare disparities
1. Introduction Breast cancer is the most common malignancy in women world wide, with nearly 1.7 million cases diagnosed in 2012, representing 25% of all female cancers [1]. In developing countries, the incidence of breast cancer is rising sharply due to changes in reproductive factors, lifestyle, and an increased life expectancy. These rising cancer rates in the developing world are projected to reach a 55%
Abbreviations: DGE, national epidemiological administration; SINAVE, national system of epidemiological surveillance; INEGI, national institute of statistics and geography; CONAPO, national population council; APC, annual percentage change. * Corresponding author. E-mail address: [email protected] (Y. Chavarri-Guerra). http://dx.doi.org/10.1016/j.canep.2016.01.007 1877-7821/ ã 2016 Elsevier Ltd. All rights reserved.
increased incidence and 58% increased mortality in fewer than 20 years [2]. Mexico is an upper middle-income country of 112, 336, 538 inhabitants [3] with one of the highest per capita incomes in Latin America. Even though Mexico is among the top 20 largest economies in the World, more than half of its population is poor [4]. The distribution of wealth in Mexico is unequal [5], with a 27fold difference between the average incomes of the top and bottom deciles of the population [6]. This inequality is also true for different regions of the country, with northern states experiencing larger economic growth due to their proximity to the United States, better infrastructure and a greater capacity for producing manufactured goods [7]. As in other developing countries, cancer incidence and mortality have been on the rise in Mexico. Since 2006, breast
E. Soto-Perez-de-Celis, Y. Chavarri-Guerra / Cancer Epidemiology 41 (2016) 24–33
cancer has been the leading cause of cancer-related mortality in Mexican women, surpassing cervical cancer [8]. A review of the regional geographic trends of breast and cervical cancer mortality between 1979 and 2006 showed a higher risk of dying from breast cancer in women from the northern states, while women from the southern states had higher mortality attributable to cervical cancer [9]. In order to tackle inequities in the access to healthcare, in 2003 the Mexican government instituted a national health insurance program directed at the poorest sectors of the population called Seguro Popular, which includes coverage of breast cancer screening, diagnosis and treatment [10]. One of the main barriers for designing strategies to tackle cancer in developing nations is the paucity of statistics regarding its incidence and mortality [11], and the Breast Health Global Initiative has included the assessment of the burden of breast cancer at the national level as one of its top priorities [2]. Due to the fact that Mexico lacks a national population-based cancer registry, information on these indicators usually comes from data extrapolated from neighboring countries, from hospital-based histopathological databases [8] or from the only regional cancer registry, which is located in the western state of Jalisco [12]. For example, data from GLOBOCAN 2012, (which reports an incidence of 35.4 cases of breast cancer per 100,000 person-years in Mexico) are not obtained from any registry, but rather extrapolated from the mortality data obtained from the death certificate database of the Instituto Nacional de Estadística y Geografía (National Institute of Statistics and Geography, INEGI) [13], which is considered of high quality. Furthermore, the previous edition of GLOBOCAN, published in 2008, obtained incidence data by using a regional model extrapolated from other territories in Central America and the Latin Caribbean with a high quality population-based cancer registry (Puerto Rico and Costa Rica) [14]. Thus, although the regional and national mortality figures attributable to breast cancer in Mexico come from reliable sources and have been published before [15], the true incidence of the disease is largely unknown and no direct data sources have been used to describe it. Starting in 2000, the Dirección General de Epidemiología (National Epidemiological Administration, DGE) of the Mexican Ministry of
25
Health instituted the weekly mandatory reporting of all new breast cancer cases as part of the Sistema Nacional de Vigilancia Epidemiológica (National System of Epidemiological Surveillance, SINAVE) [16]. SINAVE was created in the 1970’s and since 1995 it has national coverage, encompassing all the public institutions that constitute the Sistema Nacional de Salud (National Health System) and several private institutions. Data reported by SINAVE include confirmed and suspected cases of 142 different diseases, including not only breast cancer but also cervical cancer and cervical dysplasia [17]. These suspected cases are reported on a weekly basis by physicians at nurses at every medical unit or hospital belonging to the institutions included in SINAVE using an online computerized format called Sistema Único de Información para la Viligancia Epidemiologica (Unified Information System for Epidemiological Surveillance, SUIVE) [18]. The reported cases are reviewed by the Sanitary Jurisdiction to which each medical unit belongs and then reported on a state level weekly, monthly and yearly [17]. In this paper, we describe national and regional breast cancer incidence trends using population-based data from the SINAVE database and mortality trends using the INEGI records for the first decade of the 21st century, in order to understand the epidemiological variations generated by recent changes in the country, including growing access to healthcare, increasing coverage of screening methods and the quality of data reporting. 2. Patients and methods 2.1. Study area Mexico is a federation comprising 31 states and a Federal District (DF), geographically limited by United States border on the north and by the Guatemala–Belize border on the south. We divided the country into three geographic regions and 8 sub regions: North (Northwest and Northeast); Central (North-Central, South Central, West and East) and South (Southeast and Southwest). The location of each region and subregion within the country can be seen in Fig. 1.
Fig. 1. Geographic regions of Mexico.
26
E. Soto-Perez-de-Celis, Y. Chavarri-Guerra / Cancer Epidemiology 41 (2016) 24–33
the South of the country, while the North showed a stabilization of reported incidence rates.
2.2. Data sources and variables Data on the number of incident breast cancer cases reported between 2001 and 2011 were obtained from the SINAVE morbidity database [16]. Breast cancer related deaths between 2001 and 2011 were obtained from the INEGI death certificate database [13]. Incidence and mortality data were obtained for each age group and for each state, subregion and region of the country. Yearly populations were obtained from national census data and from the population growth estimates published by the Consejo Nacional de Población (National Population Council, CONAPO) [19]. Age group-specific rates and standardized (overall and truncated) rates were calculated using the direct method (using the world standard population) [20]. These were expressed as rates/100,000 persons. Only women over 15 years of age were considered for analysis. Incidence and mortality rates were calculated for each state, subregion and region. 2.3. Statistical methods We used joinpoint regression analysis to describe trends over time at the state, subregional, regional and national level, with each joinpoint indicating a statistically significant change in trend. We estimated the annual percentage change (APC) for each of those trends using weighted least squares log-linear regression and assuming a Poisson distribution [21]. The APC over the 2001–11 time period is a weighted average of the slope coefficients of the underlying joinpoint regression line with the weights equal to the length of each segment over the interval. Joinpoint analyses were performed using the ‘Joinpoint’ software from the Surveillance Research Program of the US National Cancer Institute. 3. Results From 2001 to 2011, 69,651 new breast cancer cases were reported to SINAVE (Table 1). For the same period, 48,817 deaths attributed to breast cancer were registered in the death certificate database (Table 2). 3.1. National and regional reported incidence The modeled age-standardized national breast cancer incidence rate had a significant increase, rising from 14.2/100,000 personyears (Standard Error [SE] 0.23) in 2001, to 25.2/100,000 personyears (SE 0.25) in 2011, with an APC of 5.9% (95% CI 4.1 to 7.7, p < 0.05) (Fig. 2a). Results of the joinpoint analysis of regional agestandardized incidence trends are shown in Table 3 and in Fig. 3. For the studied period, we found a significant increase in the number of new breast cancer cases reported in the Center and in
3.2. Subregional variations in reported incidence When the age-standardized breast cancer incidence rates were analyzed at the subregional level, a significant increase was seen for all subregions except the Northwest and the Northeast (Table 3). The Northeast showed no increase in incidence APC (0.7%; 95% CI 5.3–7.1), with a downward trend in the number of reported cases after 2009 in each of the three states comprising the subregion (Fig. 4). 3.3. National and regional mortality The modeled age-standardized national breast cancer mortality rate also showed an increase, rising from 14/100,000 personyears (SE 0.23) in 2001 to 14.6/100,000 person-years (SE 0.2) in 2011, with an APC of 0.4% (95% CI 0.1–0.7, p < 0.05) (Fig. 2b). Regional age-standardized mortality rates were significantly increased in all three regions of the country during the studied period, as shown in the joinpoint analysis presented in Table 3 and in Fig. 3. 3.4. Subregional variations in mortality When the breast cancer mortality rates were analyzed at the subregional level, a significant increase was seen only in the Southwest [APC 2.7% (95% CI 1.2–4.3), p < 0.05] and Center-North [APC 1.7% (95% CI 0.7–2.25), p < 0.05]. On the other hand, the Center-South subregion showed a decrease in the mortality rate, with an APC of 0.7% (95% CI 1.1– 0.3, p < 0.05). The rest of the subregions showed non-significant changes in mortality. These data are illustrated in Table 3. 3.5. Age associated trends A rise in the reported number of breast cancer cases was seen in all age groups except for women age 25 and younger. The largest increase was seen in women aged 60–65, with an APC of 7.89% (95% CI 5.5–10.3; p < 0.05), followed by women 65 and older with 6.7% (95% CI 4.9 8.6; p < 0.05). The highest observed incidence of new cases was seen in the 60–65 age group, with 62.82 cases per 100,000 women reported in 2011. During the studied period mortality rates in the younger age groups (50 years old) either decreased or remained stable, while older groups (60 years old) showed an increase (Table 4). Women older than 85 showed the highest APC in mortality of any age group (2.99%, 95% CI 1.9–4.1; p < 0.05), and the highest observed mortality rate for 2011 (88.72 deaths per 100,000). Trends according to age group can be seen in Table 4.
Table 1 Total number of incident breast cancer cases reported by SINAVE for each subregion of the country during the studied period. Region
Southwest Southeast East West Center-North Center-South Northwest Northeast Total
Year 2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
166 111 500 780 583 367 774 690 3971
267 122 601 820 678 534 996 710 4728
230 85 499 696 974 497 846 628 4455
189 78 561 827 920 748 957 738 5018
239 26 532 966 1125 733 825 677 5123
288 200 693 1008 1214 726 987 927 6043
337 222 1234 1204 1730 755 873 975 7330
453 253 1290 1164 1950 785 1133 902 7930
463 258 1418 1130 1764 862 1177 1356 8428
576 335 1591 1427 1795 716 1293 652 8385
430 431 1922 1501 1824 835 1352 597 8892
E. Soto-Perez-de-Celis, Y. Chavarri-Guerra / Cancer Epidemiology 41 (2016) 24–33
27
Table 2 Total number of Breast Cancer deaths reported by INEGI for each subregion of the country during the studied period. Region
Southwest Southeast East West Center-North Center-South Northwest Northeast Total
Year 2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
179 125 436 458 326 1111 510 455 3600
180 127 467 543 355 1134 541 508 3855
231 117 486 524 342 1155 577 452 3884
238 155 495 543 421 1202 616 503 4173
226 135 491 590 389 1228 650 520 4229
250 147 531 662 433 1231 666 537 4457
246 138 592 580 468 1283 699 598 4604
291 150 569 627 489 1355 770 579 4830
268 180 620 660 492 1335 736 614 4905
323 190 590 682 491 1392 751 640 5059
313 192 626 710 535 1429 773 643 5221
Fig. 2. National incidence (a) and mortality (b) trends 2001–2011, joinpoint analysis.
4. Discussion Breast cancer in Mexico presented a continuous increase in the reported incidence and mortality rates between 2001 and 2011. A significant increase in both rates was seen in Central and Southern regions and subregions of the country. On the other hand, the reported incidence and mortality rates in the North remained unchanged, with a downturn in more recent years. Women 60– 69 years old showed the largest rise in reported incidence, while women 60 years of age and older were the group with the highest increase in mortality. 4.1. Population aging The observed increase in the incidence of breast cancer could reflect population growth, aging and lifestyle modifications seen in developing countries transitioning to a more “developed” epidemiological landscape. Life expectancy in Mexico doubled during the second half of the 20th century, and by 2005 the life expectancy of Mexican women was 77.9 years. Although in the year 2000 older adults represented only 7% of the population, this proportion is expected to increase to 12.5% by 2020 and to 28% by 2050, owing to the aging of people born during the Mexican economic expansion of 1960–1980 (the Mexican baby boomers) [19]. This will inevitably lead to an increase in the incidence of diseases associated with aging, including breast cancer. Additionally, since women in the 50–59 and 60–64 age group are specially targeted by breast cancer screening programs, this is expected to rise even further. As a matter of fact, an increase in
the incidence and mortality of breast cancer in women 60 years of age and older seems to be the main factor driving the upward trends we observed. These changes might lead to an increase in the median age at diagnosis for breast cancer patients in Mexico, which has historically been reported to be of 50 years of age [8]. Another interesting observation is the fact that the number of new cases reported in women 65 years and older is lower than that in the 50–59 and 60–64 subgroups, in contrast to what is seen in other countries like the United States [22]. Although there is no clear explanation for this, it may be due to the fact that older women in Mexico have lower participation in screening programs or that they fail to seek medical attention for breast symptoms. 4.2. Changes in access and availability of healthcare facilities An important landmark which could partially explain some of the changes seen in our study is the aforementioned establishment of the Seguro Popular program in 2003. In the first 9 years after its inception, 52.6 million Mexicans, previously uninsured, were incorporated into the program [10]. Currently, the public sector accounts for 88% of the available medical facilities, while the private sector is responsible for the remaining 12% [23]. 44% of the population is covered by Social Security (for salaried workers) and 45% is covered by Seguro Popular, while 7% are covered by other public schemes (directed at government employees) and less than 3% have private insurance [24]. From 2003 to 2013, the number of existing medical facilities (including both ambulatory clinics and hospitals) in the public sector grew from 19,505 to 22,228,
28
E. Soto-Perez-de-Celis, Y. Chavarri-Guerra / Cancer Epidemiology 41 (2016) 24–33
Table 3 Modeled regional age - s tandardized incidence and mortality rates for women 15 years of age in Mexico, 2001–2011. Region
Rate
Southwest
Incidence. Mortality Incidence Mortality Incidence Mortality Incidence Mortality Incidence Mortality Incidence Mortality Incidence Mortality Incidence Mortality Incidence Mortality Incidence Mortality Incidence Mortality Incidence Mortality
Southeast South East West Center-North Center-South Center Northwest Northeast North National
a b c d
Modeled ASR per year/100,000 women (WHO Standard Population) [13,14]
APC (2001–11).
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
6.43 7.04 6.11 9.58 6.10 7.79 9.32 10.35 20.94 14.87 16.82 11.90 8.45 16.79 12.88 13.95 23.10 16.36 21.86 18.41 24.24 17.52 14.17 14.03
7.00 7.23 6.88 9.66 6.74 7.95 10.55 10.42 21.91 15.03 17.31 12.10 9.70 16.67 13.86 13.99 23.51 16.69 23.06 18.51 24.49 17.54 15.00 14.08
7.62 7.43 7.76 9.73 7.45 8.11 11.94 10.50 22.93 15.19 17.83 12.30 11.14 16.56 14.92 14.04 23.94 17.03 24.34 18.60 24.74 17.57 15.89 14.14
8.29 7.63 8.75 9.81 8.23 8.28 13.52 10.57 23.99 15.36 18.35 12.51 12.79 16.45 16.06 14.08 24.37 17.38 25.69 18.70 24.99 17.59d 16.83 14.20
9.02 7.84 9.86 9.89 9.10 8.45 15.30 10.64 25.10 15.52 18.89 12.72 14.68 16.33 17.29 14.13 24.81 17.73 27.11 18.79 25.25 18.00 17.83 14.25
9.81 8.05 11.11 9.97 10.05 8.63 17.32 10.72 26.27 15.69 19.45 12.94 16.85 16.22 18.61 14.18 25.26 18.09 28.60 18.89 25.51 18.41 18.88 14.31
10.67 8.27 12.53 10.04 11.11 8.81 19.61 10.79 27.49 15.86 20.02 13.15 19.34 16.11 20.03 14.22 25.71 18.45 30.19 18.98 25.77 18.84 20.00 14.36
11.61 8.50 14.12 10.12 12.28 8.99 22.19 10.87 28.76 16.03 20.61 13.37 22.20a 16.00 21.56 14.27 26.18 18.83c 31.85 19.08 26.03 19.28d 21.18 14.42
12.63 8.73 15.92 10.20 13.57 9.18 25.12 10.95 30.10 16.20 21.22 13.60 20.96 15.89 23.20 14.31 26.65 18.32 33.62b 19.18 26.30 18.85 22.44 14.48
13.75 8.97 17.94 10.28 14.99 9.37 28.44 11.02 31.49 16.38 21.85 13.83 19.79 15.79 24.97 14.36 27.13 17.82 22.59 19.28 26.56 18.43 23.76 14.53
14.96 9.21 20.22 10.37 16.57 9.56 32.19 11.10 32.96 16.55 22.49 14.06 18.69a 15.68 26.88 14.41 27.62 17.34c 15.18b 19.37 26.84 18.02d 25.17 14.59
Center-South. Incidence. 2001–2008 APC 14.8% (95% CI 9.7–20.2, p < 0.05). 2008–2011 APC 5.6% (95% CI 16.8–7.1, p NS). Northeast. Incidence. 2001–2009 APC 5.5% (95% CI 1.6–9.6, p < 0.05). 2009–2011 APC 32.8 (95% CI 1.1–54.3, p < 0.05). Northwest. Mortality. 2001–2008 APC 2% (95% CI 1.1–2.9, p < 0.05). 2008–2011 APC 2.7% (95% CI 5.5–0.2, p NS). North. Mortality. 2001–2004 APC 0.1% (95% CI 2.8–3.1, p NS). 2004–2008 APC 2.3% (95% CI 0.3–5.1, p NS). 2008–2011 APC
representing a 14% increase[23]. Likewise, the number of available medical offices in the public sector grew from 51,598 in 2001 to 70,196 in 2011 [25]. Regarding available personnel in the public sector, the number of physicians (regardless of their specialization) grew from 112,681 in 2001 to 170,565 in 2011, and the number of nurses went from 192,046 to 250,931 during the same period [25].
8.8 2.7 12.7 0.8 10.5 2.1 13.2 0.7 4.6 1.1 2.9 1.7 8.5 0.7 7.6 0.3 1.8 0.8 0.7 0.5 1.0 0.7 5.9 0.4
2.2% (95% CI
95% CI.
4.3–13.5 1.2–4.3 5.9–20.0 1.0–2.6 6.0–15.2 1.0–3.2 9.3–17.3 0.2–1.6 2.7–6.7 0.1–2.2 0.2–6.2 0.7–2.7 4.3–12.8 1.1– 0.3 5.7–9.6 0–0.6 0.5–4.2 0.1–1.7 5.3–7.1 0.3–1.3 1.6–3.7 0–1.3 4.1–7.7 0.1–0.7
p
