The Prostate 39:23–27 (1999) Prostate Cancer Mortality Trends in Mexico, 1980–1995 Vı́ctor Tovar-Guzmán,* Carlos Hernández-Girón, Olga López-Rı́os, and Eduardo C. Lazcano-Ponce Center for Public Health Research, National Institute of Public Health, Cuernavaca, Mexico BACKGROUND. In Mexico, prostate cancer (PC) is the second highest cause of mortality by cancer in men. In 1991, there were 2,473 deaths, with a crude rate of 5.66 cases per 100,000 men age 40 or over. For this same year, the ratio of proportional mortality to the total number of malignant tumors was 12.6%. Our objectives were to determine the trends and geographic distribution of mortality for PC in Mexico for 1980–1995. METHODS. Crude and adjusted mortality rates were carried out for 1980–1995. A test for trend was done using the simple lineal regression method. The standardized mortality ratio (SMR) was calculated for 1980–1995 for each Mexican state. RESULTS. The total number of deaths by PC in Mexico for 1980–1995 was 32,349. The crude mortality rate increased during that period from 3.16 to 6.75 cases per 100,000 men over age 40. The statistical trend test with respect to time was significant (P < 0.0001). The state of Southern Baja California showed a higher SMR, and a smaller SMR was shown for the state of Quintana Roo. CONCLUSIONS. The mortality trends for PC increased for the period studied. It represents a serious problem for public health. We find that the central and northern states of the country, characterized by a greater industrial and socioeconomic development, are those with a higher SMR. Prostate 39:23–27, 1999. © 1999 Wiley-Liss, Inc. KEY WORDS: prostate cancer; mortality trends; Mexico INTRODUCTION Today, research on prostate cancer (PC) in Mexico is limited. Some descriptive studies exist with respect to the epidemiology of this disease, which mainly report the rapid increase in incidence and mortality, and above all its appearance starting at age 45 years, with an alarming rate in men 75 years and older. In Mexico, PC is the second highest cause of mortality by cancer in men; the first is lung cancer. In 1991, there were 2,473 deaths due to PC in Mexico, which is equivalent to a crude rate of 5.66 cases per 100,000 men aged 40 or over. For this same year, the ratio of proportional mortality to the total number of malignant tumors was 12.6%. For 1992, mortality rates by state did not have a logical distribution pattern. The state with the highest rate was Southern Baja California, and the one with the lowest rate was Quintana Roo (6.4 and 1.1 per 100,000 inhabitants, respectively) . The objectives of our descriptive epidemiological © 1999 Wiley-Liss, Inc. research were to determine trends of specific crude and adjusted mortality rates for PC in Mexico for 1980–1995, and their distribution by 5-year age groups starting at age 40. Also, we calculated the value of the standardized mortality ratio by state, which will provide information on the difference between observed and expected cases. Another objective was to calculate the correlation between specific mortality rates and the socioeconomic level by state in Mexico. MATERIALS AND METHODS This work was carried out through an analysis of information obtained from electronic records of PC *Correspondence to: Dr. Victor Tovar-Guzmán, Center for Public Health Research, National Institute of Public Health, Av. Universidad 655, Col. Sta. Marı́a Ahuacatitlán, Cuernavaca, Mor. C.P. 62508, Mexico. E-mail: email@example.com Received 18 March 1998; Accepted 28 September 1998 24 Tovar-Guzmán et al. Fig. 1. Mortality trends in prostate cancer among Mexican men, 1980–1995. Crude and age-adjusted rates. *P < 0.0001. Fig. 2. Mortality trends by 5-year groups for prostate cancer in Mexico, 1980–1995. *P < 0.05. mortality, edited by the National Statistics, Geography, and Informatics Institute (INEGI), which also integrates records such as age, state of residency, determinant cause of death (International Disease Classification (CIE) no. 185) and year of death . For calculations of crude mortality rates, was used as denominator the population projected by the National Population Council (CONAPO). Adjusted mortality rates for PC, for 1980–1995, were calculated according to the direct statistical method, using the population of the Mexican Republic as a reference [3,4]. The calculation of the statistical trend test, measured through the value of P, was done using the simple lineal regression method, taking as a depen- PC Trends in Mexico 25 Fig. 3. Standardized mortality ratios of prostate cancer by state in Mexico, 1980–1995. dent variable the crude and adjusted mortality and, as an independent variable, the year of death . The crude cancer mortality rate trends by 5-year groups, starting at age 40, were calculated using a robust regression statistical technique, obtaining from this the regression coefficient, 95% confidence interval, and statistical P value . The working hypothesis distinguishes the gradient, which is unequal to zero (␤ ⫽ 0). The corresponding graph of rate trends was done using the x-axis for the year of the study and the y-axis for the corresponding rate. For this, the arithmetic scale was changed into a logarythmic one in order to obtain a better understanding of the phenomenon under study. The standardized mortality ratio (SMR) was calculated for 1980–1995 with a corresponding confidence interval (CI) of 95% for each Mexican state. The SMR is an intermediary step in the calculation of adjusted statistical rates, using the indirect method . Statistical analysis was carried out considering the rural level, which was calculated by CONAPO using different social indicators for the year 1990 for each Mexican state  and the mortality rate due to PC. Statistical analysis was carried out using the STATA 3.1  and EXCEL  packages. RESULTS The total number of deaths by PC in Mexico for 1980–1995 was 32,349, as recorded by death certificates. The crude mortality rate increased during that period from 3.16 to 6.75 cases per 100,000 men over 40; this represents an increase of 113.6% between 1980– 1995. The adjusted rate by age for the same period varied between 2.71–7.01 cases per 100,000 men over 40, representing an increase of 191.9%. The statistical trend test for the two rates with respect to time were significant (P < 0.0001) (Fig. 1). When analyzing mortality trends for each 5-year age group, a statistically significant increase was observed in the slopes, starting from the 50–54-year age group (P < 0.05). The greatest regression slope was for the group of men older than 75 years, with a ␤ coefficient slope of 10.98 (P < 0.0001). The crude mortality rate varied, and in the group of men 75 and older, it 26 Tovar-Guzmán et al. TABLE 1. Standardized Mortality Ratio of Prostate Cancer by State in Mexico, 1980–1995* State BCS Jalisco Aguascalientes Colima Michoacán Sonora Zacatecas Nayarit Guanajuato Chihuahua Tabasco Sinaloa Coahuila BCN Tamaulipas Distrito Federal SMR 95% CI State SMR 95% CI 183.28 161.81 152.21 151.83 145.05 139.69 135.49 135.18 127.05 126.24 121.96 121.95 119.10 112.67 111.62 104.86 (158.36–208.18) (156.18–167.44) (136.15–168.27) (132.78–170.88) (138.35–151.74) (131.32–148.07) (124.99–146.00) (122.88–147.48) (120.93–133.18) (119.14–133.34) (111.90–132.02) (114.69–129.21) (111.27–126.92) (104.05–121.28) (104.90–118.35) (101.78–107.93) SLP Nuevo León Morelos Yucatán Querétaro Tlaxcala Hidalgo Durango Chiapas Campeche Veracruz Puebla Oaxaca México Guerrero Quintana Roo 102.01 99.93 98.22 97.23 95.41 95.06 88.64 86.44 83.60 83.27 82.72 81.99 60.49 59.91 57.69 47.87 (94.96–09.06) (94.41–105.46) (89.21–107.22) (89.23–105.22) (84.74–106.09) (83.13–106.99) (81.77–95.50) (78.47–94.40) (77.78–89.42) (71.30–95.24) (79.29–66.16) (77.43–86.56) (55.94–65.04) (57.46–62.36) (52.89–62.49) (35.86–59.89) *SMR, standardized mortality ratio; 95% CI, 95% confidence interval. increased from 170.1 in 1980 to 331.1 in 1995, which represented an increase of 94% (Fig. 2). The standard mortality ratio (SMR) for the different Mexican states showed a dispersed relationship between the different regions, with higher SMRs in the states of Southern Baja California (SMR, 183.28; 95% CI, 158.36208), Jalisco (SMR, 161.81; 95% CI, 156.18– 167.44), and Aguascalientes (SMR, 136.15; 95% CI, 136.15–168.27). The smaller SMRs corresponded to the states of Quintana Roo (SMR, 47.87; 95% CI, 35.86– 59.89), Guerrero (SMR, 57.69; 95% CI, 52.89–62.49), and Mexico (SMR, 59.91; 95% CI, 57.46–62.36) (Table I, Fig. 3). A statistical analysis was carried out using mortality rate values for PC and rural level for 1990, which gave us a value of r = −0.15 (P < 0.005), showing that at higher rural levels, mortality rates were lower. DISCUSSION It is important to point out the change in the mortality behavior pattern that is occurring in Mexico. In the last few decades, this has changed from a mortality where causes were mostly infectious or due to deficiencies to a pattern of chronic diseases, particularly cerebral-vascular and cancer. These causes are turning into an important public health problem. In this context, PC is not an exception: its risks, characterized by the influence of environmental and hormonal factors, as well as life styles, have had an influence on the incidence of the disease, with truly specific risk factors, such as its appearance in age groups older than 50 and primarily in the black race [10–13]. When analyzing SMR distribution in Mexico, we found that the central and northern states of the country, characterized by greater industrial and socioeconomic development, were the ones with a higher SMR, i.e., the number of observed cases was greater than the number of expected cases, and this agrees with the international literature, in the sense that this type of cancer is more frequent in developed countries [14–17]. In this study, we found an inverse determination coefficient between rural level and crude PC rates. This confirms what we stated above, that the states with a higher rural level (Chiapas, Oaxaca, Campeche, Guerrero, Tabasco, and Quintana Roo) are the ones with lower rates for this kind of disease. Environmental and geographic patterns, as well as lifestyles that together become causes for PC, are distributed differentially throughout the Mexican regions; for example, the prevalence of determined risk factors, such as lifestyles (a diet rich in saturated fats, low physical activity, obesity, smoking, alcoholism) are mainly manifested in the northern Mexican states. These findings are similar to those reported in the world literature [18–21]. A potential bias of mortality rates could be due to the change in the quality of records. Some years ago, there were problems in the notification and recording of information, which created difficulties in diagnostic certainty. Presently, due to state policies implemented in the last several years, there is now greater certainty in the statistics of secondary information on general mortality. PC Trends in Mexico CONCLUSIONS In Mexico, massive PC detection campaigns have not yet been implemented. Techniques most often used are rectal exploration by touch, transrectal ultrasonography, and serological markers ; these have been mainly used in the diagnostic clinical phase. The control and reduction of PC could be carried out in two ways: 1) with a decrease in new cases, and 2) through an increase in survival rates, through early detection, and through improvements in treatment. This would lead to a decrease in PC mortality. In order to complement a better epidemiological scenario for PC, it is important to carry out more descriptive and analytical research to obtain a better comprehension of this problem. REFERENCES 1. Olaiz G, Hernández M. Cáncer: un encuentro con el futuro epidemiológico. Cuadernos de Salud. Secretarı́a de Salud. México; 1994. 2. OPS. Clasificación internacional de enfermedades. IX revisión. Washington, DC. 1978. 3. Lilienfield D, Stolley P. Foundations of epidemiology, 3rd ed. New York: Oxford University Press; 1994. 4. Dever A. Epidemiologı́a y administración de servicios de salud. OPS. Washington, DC. 1991. 5. Daniel W. Bioestadı́stica. Base para el análisis de las ciencias de la salud, 3rd ed. Mexico: Limusa Noriega; 1990. 6. Hamilton L. Statistics with Stata 3. Belmont, CA: Duxbury Press; 1993. 7. Indice de marginación social. Mexico: CONAPO; 1990. 8. Stata Corporation. Stata release 3.1. College Station, MA: Stata Corporation; 1993. 9. Microsoft Corporation. Microsoft Excel 5.0. Cambridge, MA: Microsoft Corporation; 1994. 27 10. Ernester V, Selvin S, Sacks S, Austin D, Brown S, Winkelstein W. Prostatic cancer: mortality and incidence rates by race and social class. Am J Epidemiol 1978;107:311–320. 11. Greenwald P, Damon A, Kirmss V, Polan A. Physical and demographic features of men before developing cancer of the prostate. JNCI 1974;53:341–346. 12. Guilleyardo J, Johnson W, Weish R, Akazaki K, Correa P. Prevalence of latent prostate carcinoma in two U.S. populations. JNCI 1980;65:311–316. 13. Krongrad A, Lai H, Lai S. Competing risk of mortality in prostate cancer. J Urol 1997;87:865–868. 14. Bolumar F, Vioque J, Cayuela A. Changing mortality patterns for major cancers in Spain, 1951–1985. Int J Epidemiol 1991;20: 20–25. 15. Magrath I, Litvak J. Cancer in developing countries: opportunity and challenge. J Natl Cancer Inst 1993;85:862–874. 16. Wynder E, Fujita Y, Harris R, Hirayama T, Hiyama T. Comparative epidemiology of cancer between the United States and Japan. Cancer 1991;67:746–763. 17. Hill C, Benhamou E, Doyon F. Trends in cancer mortality, France 1950–1985. Br J Cancer 1991;63:587–590. 18. Hiatt R, Armstrong M, Klatsky A, Sidney S. Alcohol consumption, smoking, and other risk factors and prostate cancer in a large health plan cohort in California (United States). Cancer Causes Control 1994;5:66–72. 19. Anderson S, Baron J, Bergstrom R, Lindgren C, Wolk A, Adami H. Lifestyles factors and prostate cancer risk: a case-control study in Sweden. Cancer Epidemiol Biomarkers Prev 1996;48: 509–513. 20. Rohan T, Howe G, Burch J, Jain M. Dietary factors and risk of prostate cancer: a case-control study in Ontario, Canada. Cancer Causes Control 1995;48:145–154. 21. Whitemore A, Kolonel N, Wu A, John E, Gallagher R, Howe G. Prostate cancer in relation to diet, physical activity and body size in blacks, whites and Asians in the United States and Canada. J Natl Cancer Inst 1995;87:652–661. 22. Gerber G, Chodak G. Routine screening for cancer of the prostate. J Natl Cancer Inst 1991;83:329–335.