Critical review of cancer mortality using hospital records and potential years of life lost

ABSTRACT Objective To determine and discuss cancer mortality rates in southern Brazil between 1988 and 2012. Methods This was a critical review of literature based on analysis of data concerning incidence and mortality of prostate cancer, breast cancer, bronchial and lung cancer, and uterine and ovarian cancer. Data were collected from the online database of the Brazil Instituto Nacional de Câncer José Alencar Gomes da Silva. Results The southern Brazil is the leading region of cancer incidence and mortality. Data on the cancer profile of this population are scarce especially in the States of Santa Catarina and Paraná. We observed inconsistency between data from hospital registers and death recorded. Conclusion Both cancer incidence and the mortality are high in Brazil. In addition, Brazil has great numbers of registers and deaths for cancer compared to worldwide rates. Regional risk factors might explain the high cancer rates.


❚ INTRODUCTION
Cancer is one of the leading causes of death worldwide. (1) From 2014 to 2015 Brazil had estimated more than 500,000 new cases of cancer, (2) which placed einstein (São Paulo). 2018;16(1):1-7 the country among those with the highest cancer incidence in current days. (3) Although the growing efforts for early screening and diagnosis, the associated risk factors for development of this disease are strongly present in Brazilian population, particularly smoking, occidental diet, obesity and sedentarism. (4) A high cancer incidence in seen in most-populous and industrialized regions of the country; the south region. This region concentrates the highest incidence of neoplasias, according to figure 1, and this fact might be associated with southern population longer life expectancy and healthy habits. (5) This high number of cancer cases among southern population still a reflex of great number of diagnoses and, consequently, increase records in official databases. (6) ❚ OBJECTIVE To discuss the cancer issue in southern Brazil within the last 20 years based on information from the database of Instituto Nacional de Câncer José Alencar Gomes da Silva particularly issues related with notification of this disease and mortality related to prostate cancer, breast cancer, bronchial and lung cancer, and uterine and ovarian cancer.

❚ METHODS
This is a critical review on estimation analysis of incidence and indicators of mortality of prostate cancers, breast cancer, bronchial and lung cancer, uterine and ovarian cancer based on consultations of online database of National Cancer Institute José Alencar Gomes da Silva (INCA -Instituto Nacional de Câncer). Number of hospital records and deaths for each topography and by States of Southern Region were obtained from online data of hospital-based cancer registry of INCA (1998INCA ( -2012, and from INCA's mortality atlas (2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012). Such registers were selected to compose this study because they included information that enabled to evaluate quality of service provide the hospital network in Brazil.
In addition, we evaluated data related with crude and adjusted rates and number of years of potential life lost for cancer from INCA's mortality atlas using based on total of Brazil population available in census from 2010 (https://mortalidade.inca.gov.br/MortalidadeWeb/ pages/Modelo01/consultar.xhtml;jsessionid=CA2C39 0AB43798B3880F4078A4345436). Adjusted mortality rate by age evaluates the number of deaths in each age range regarding total of deaths in south population and this rate was directly standardized by primary source consulted in order to reduce bias that age factor can add to studies on cancer. Proportional mortality data were used to illustrate amount of deaths in population affected by cancer in studied period and number of years of potential life lost as indicator of total sum of years lost for each death because of cancer. Such associated indicators enabled to measure impact of each cancer in the studied population within each year.
All database mentioned were searched within the INCA website. (2) The values presented were obtained from those calculated by databases, using as reference the estimated southern Brazilian population for each year. Coefficients were also adjusted by age by database searched. To discuss results, we used published studies in PubMed and Scientific Electronic Library Online (SciELO). Although the high cancer incidence and mortality rates in south Brazil, few studies have discussed possible associated factors with this disease. Some neoplasias such as ovarian and breast cancer have a strong genetic predisposition among population in this region prevalent. (7) In addition, southern population high fat diet and consumption of smoked products favor the development of obesity, one of the main factors that lead to development of cancer in the modern society. (8) Such factors contribute significantly to development of clinicopathological cancers with worse prognosis that may impact cancer morbidity and mortality.

❚ RESULTS
Results obtained highlighted Southern Region as national leading region of cancer incidence and mortality from 1988 to 2012. This comparative analysis of number of hospital-register cancer between 2001 and 2012 for prostate cancer, breast cancer, bronchial and lung cancer, uterine and ovarian cancer showed expression values for Southern Region with special emphasizes to Rio Grande do Sul State, followed by Paraná and Santa Catarina (Table 1).
Prostate and breast cancer had higher incidence than number of deaths in Paraná and Rio Grande do Sul States. The same result was seen in Santa Catarina after 2004. In Southern Region, we also observed increase in registers of prostate cancer diagnosis, although this type of cancer did not show increase in mortality (Table 1).
Regarding bronchial and lung cancer, we observed an increase of number of hospital register and deaths, particularly in the State of Rio Grande do Sul. This was the leading State in number of deaths because of gynecologic cancers. The same result was seen in this State to the five cancer types analyzed.
Still, south region had the highest mortality rates for cancer. Table 2 shows that standardized world mortality rate was achieved and overpassed within the age 40 to 49 years, excepted to prostate cancer. Mortality rate increased gradually from 60 years of age, and higher indexes occurred among individuals aged 70 to 79 years (196.37/100,000) reaching 538.54/100,000 of those older than 80 years.
For breast cancer, 91.3% of cases of mortality occurred after 50 years, achieving rate of 116 cases/100,000 in age range ≥80 years. In relation to bronchial and lung cancer, highest rates of mortality were observed after 60 years, reaching 264.24/100,00 in men aged 70 to 79 years and 281.79/100,000 in men aged over 80 years.
Still this rate was lower among women, and relationship between sexes was 3.1 men for each women. To each uterine cancer, we observed higher number of deaths among those aged 40 to 49 years. From this age range, this rate (10.18/100,000) was double of standardized Brazilian rate (5.66/100,000) and world rate (5.27/100,000), reaching 22.32/100,000 in age of 80 years. Similar results can be seen in death rates due to ovarian cancer in which age range from 60 to 69 years had 26.06% of deaths (14.37/100,000). From the 40 years, death rate increased significantly compared with age 30 to 39 years (0.95/100,000).
Non-adjusted proportional death rate ( Figure 2) enabled to observe and analyze deaths of main cancers types (lung, breast, uterine, ovarian, and prostate) that occurred in south Brazil between 1988 and 2012. The association of data from Southern Region with hospital records showed that in 2003 a estimation of 4,980 prostate cancer cases occurred per 100,000 inhabitants in southern Brazil, however, the number found was 1,028. Although hospital register did not correspondent to incidence, the value obtained in hospital-based cancer register was lower than estimated. In 2012, however, the estimation by INCA was 9,490 new cases per each 100,000 inhabitants and, again, hospital registers showed a number about three times lower than expected, i.e., 3,022 registers.
We observed a gradual increase of proportional mortality per breast cancer between 1988 to 2012. In case of bronchial and lung cancer death, there was an increase, with raising in rate from 1.98/100,000 in 1998 to 3.12/100,000 in 2012. Non-adjusted proportional mortality for uterine cancer had a decrease from 1.   Figure 3 shows mean number of years of potential life lost because for cancer from 1998 to 2012 per each 1,000 inhabitants in Southern Region population aged no older than 80 years old. of cancer in one or more institutions and it can enable calculation of incidence of any cancer. (9) Mortality rate by age that evaluate the number of deaths in each age range regarding total of deaths in a specific period and among population living in specific geographic area, (10) was higher compared with world rates. For this reason, it is important to highlight that prostate cancer, breast cancer, bronchial cancer, and lung cancer have higher mortality rates in Brazil, and high rate can be observed in south of the country. Aspects related to weakness of Brazilian system in cancer prevention, it diagnosis in advanced phase cases, population's habits and precarious life conditions, difficult to access health system and get screening (11) can affect the observed numbers.
Southern Brazilian population has a good socioeconomic development profile that positively reflect prevention and early treatment of diseases. (12) This aspects help to understand low crude mortality rate observed in data collected. Age seems to be the main risk factor for cancer because of increase in life expectation followed by increase of chronic-degenerative diseases such as cancer. (13) Southern Region is highlighted because it presents better life expectations than the rest of the country throughout the years, and this represents an important fact to explain increased incidence of cancers registered in the study. This fact explain, e.g., the high incidence of diagnosed prostate cancer, once longevity constitutes an important risk factor for development of this cancer. (14) Cancers that were most common among men, we observed high number of prostate cancer records within this population in the studied period, although the predominance of European descent individuals. Southern Region has predominance of Caucasian population and prostate cancer has a higher incidence among African-descent individuals. This suggests that other regional factors, besides age, may influence high prevalence of this neoplasia in studied population with active investigation of prostate cancer by screening exams. (15) Despite high incidence and high mortality rates by prostate cancer are relatively low compared with other types of cancers, (14) mainly because this entails a relatively aggressive cancer. (16) In bronchial and lung cancer the high number of registers and deaths can be closed correlated with smoking, mainly because this region has the highest prevalence of smokers in the country. (17,18) This increase of mortality rate by bronchial and lung cancer is Mean number of years of life lost by men and women diagnosed with bronchial and lung cancer appeared in all age ranges, and it was higher between 60 and 69 years, achieving 13.74 of potential years of life lost. In general, based on potential years of life lost indexes and potential years of life lost rate in relation to studied cancers, the prostate cancer revealed rate of 6.69, and for breast 9.08, bronchial and lung 13.74, uterine 3.49 and for ovarian cancer 1.9. Age range regarding number of deaths for prostate, bronchial and lung cancers was higher among those aged 60 to 69 years. Number of deaths for breast, uterine and ovarian cancer was higher among those aged 50 to 59 years. Women diagnosed with breast cancer in the analyzed period lost, on average, 9 years of life.
❚ DISCUSSION Data presented in this study enabled to affirm that south region concentrated the highest national incidence of hospital register for cancer with high mortality rate in the studied period. Among analyzed markers, hospital cancer registers used as basis for our study are important to measure relevance of this disease in public health because they represent assistances yearly for each type  because of the growing number of women who smoke, and because of higher risk of development of different cancers. (19) Mortality profile observed in bronchial and lung cancer in older population can still be associated with smoking history, (19) once the incubation period of this pathology can last for 30 years. (20) The reduced tendency of death among younger men probably reflect national actions to reduce smoking rates in Brazil.
Regarding cancer among women, the prevalence of smoking in the region is still an important risk factor to increase incidence of gynecology cancers, such as uterine cancer. (2) Reduction of non-adjusted mortality observed in this study between 1999 and 2001 could be for gradual broad of screening services as well as strengthening and qualification of basic care network within last 20 years.
Age of ovarian cancer diagnosis ranged from 42.7 to 48 years, whereas mean described for sporadic cases is closed to 61 years old. These findings suggest possible existence of family heritage about one in each 10 cases of women with ovarian cancer that present standard that gives susceptibility for early occurrence of hereditary breast and ovarian cancer syndrome. (21) In this sense, we observed in Rio Grande do Sul a high prevalence of breast cancer among young women. In this situation, this neoplasia is presented as aggressive disease of difficult treatment, (22) which explain the high mortality rate observed after 50 years. Indeed, we observed that around 6% of southern population of metropolitan region have hereditary predisposition for breast cancer that commonly occurs in individuals younger 50 years and who present quite aggressive characteristics with high mortality rate. (23) In addition, a study done by Cadaval Gonçalves et al., revealed that, between 1980 and 2005, Rio Grande do Sul had highest concentration in the country of death rates for breast cancer. (24) Therefore, the study suggested that morbidity and mortality of the disease is a true fact in this population and it has been associated with socioenvironmental facts such as low formal education level, chronic exposition to agriculture products and overweight. (25)(26)(27) The unclear data observed between incidence registers and number of deaths for cancer included in our study pointed out the need to standardize data register about cancer in Brazil.
Other indicators analyzed regarding mortality, the number of productive years of life lost, is an indicator that reflect total sum of years of potential life lost. (28) However, the index called years of productive life lost enabled to compare different causes of death in a specific population. (29) Both indexes allowed to measure the amount of years of life lost by population because of cancer. In this study, we observed that high-incidence cancers were also those that potentially debilitate population, and they reduced individuals' years of life up to 14 years, which is case of individuals affected by lung cancer. Of note is that use of tobacco products, a risk factor to all studied cancers, often start during individuals' adolescence. For this reason, a residue of this tobacco use is expected in more advanced age ranges, especially among men.
Great number of years of life lost for ovarian cancer is because of this disease aggressiveness and also because it is discovered in more advanced phases, which results in extremely low survival rate. On the other hand, in uterine cancer, we observed low rate of years of potential life lost compared with prostate cancer, breast cancer, and bronchial and lung cancer. This finding emphasizes the importance of oncological care programs in Brazil to promote women health that massive investments occurred since 2005 to implement a national oncological care plan.
Prostate cancer had a reduced impact on years of potential life lost compared with cancer of high incidence in south region. The number of years of life lost by men can have an important relationship with extended survival of patients with prostate cancer that can achieve up to 5 years after diagnosis and treatment. (30) Based on mean life expectancy of Brazilian population that in 2010 was considered around 75 years old, we can affirm that almost one fifth of total years of individual's life is lost when people had suffered neoplasias such as bronchial and lung cancer. There is also the socioeconomic impact because of productive years of life lost, especially in cases of uterine cancer, because this latter is a potentially avoidable cancer. Our data suggest that these significant losses of productive years in Southern Region is a negative socioeconomic impact to Brazil.

❚ CONCLUSION
Both incidence and mortality of cancer are still high in Brazil with significant number of registers and deaths compare with worldwide rates. There is no agreement between number of hospital-based cancer registers and number of deaths because of cancer considering that, in some years, the register is lower than number of deaths. In addition, we observed a great number of death because of uterine cancer in south Brazil. Few studies were carried out to collected data on cancer profile in south Brazilian population, especially in States of Santa Catarina and Paraná.