ABSTRACT
Objectives
To assess the effect of age‐period‐birth cohort on mortality rates related to lip/oral and oropharyngeal cancer (LOOPC) in Brazil from 1980 to 2019 and to estimate the future mortality rate for 2042.
Materials and Methods
Mortality rate per 100 000 inhabitants, and age‐standardised mortality rate (ASR) per 100 000 inhabitants were estimated. The Prais‐Winsten regression model was used to estimate the trends and the annual percent change (APC%). The age‐period‐cohort effects were calculated using the Poisson regression model. Lee‐Carter model was employed to perform projections.
Results
A total of 134 941 deaths were observed. Prais‐Winsten regression model revealed a slight upward trend in lip and oral cancer mortality among men (p = 0.04) and women (p = 0.02), as well as in oropharyngeal cancer among men (p = 0.02). Significant age‐period‐cohort was observed for LOOPC in both sexes (p < 0.01). The risk ratio declined in recent cohorts for men (Both Cancers) but increased for women (Lip/Oral Cancer). Period analysis showed a risk increase for lip/oral cancer in recent periods in both sexes and a decrease for men and women for oropharyngeal cancer. In 2042, mortality projections decrease in lip/oral cancer for men aged between 40 and 60 years and oropharyngeal cancer in men between 35 and 60 years. For women, no significant changes are projected. The model projections mortality rate reveal varied outcomes across the diverse regions of Brazil.
Conclusion
A significant age‐period cohort was observed over the 40 years assessed. Projections for 2042 indicated a significant decrease in LOOPC mortality rates for men and no change for women.
Keywords: epidemiology, mortalities, mortality rates, mouth neoplasm, oral neoplasm
1. Introduction
Cancer is the second leading cause of worldwide death, accounting for 4.45 million deaths in 2019 [1]. Squamous cell carcinoma (SCC) is the primary type of lip, oral, and oropharyngeal cancer (LOOPC), leading to 188 438 deaths worldwide in 2022 [2]. Historically, the incidence of oral cancer has been highest in South and Southeast Asia, parts of Western and Central Europe, and South America [3]. The International Agency for Research on Cancer (IARC/WHO) has shown that Brazil and Uruguay had the highest mortality rates from oral cancer in Latin America, between 2000 and 2020 [3].
Although potentially preventable, LOOPC presents a significant public health challenge, especially in low‐ and middle‐income countries [2, 4]. These cancers arise due to a multistep process involving a progression model with multiple genetic and epigenetic events [5]. While they traditionally share common non‐genetic risk factors such as tobacco and alcohol consumption [6], exposure to human papillomavirus (HPV) is also recognised as a specific risk factor primarily for the oropharyngeal region [7]. Lip cancers are strongly associated with ultraviolet radiation (UVR) from long‐term sunlight exposure, particularly in individuals with a deficiency of melanin pigment having outdoor occupation [6, 7, 8]. Recent data show a global trend toward decreased tobacco use [9], accompanied by a simultaneous increase in HPV infections [10, 11]. This increase is largely attributed to the growing number of sexual partners and the practice of unprotected sex, especially among young people [12]. These temporal changes in social behaviour patterns are essential to understanding the shift in oral and oropharyngeal cancer profiles for designing effective public health interventions [13].
To date, only a few studies have projected LOOPC mortality [14, 15]. Existing forecasts indicate an increase in oral cancer in the United Kingdom [15] and a rise in LOOPC among women in Spain, while oropharyngeal cancer is expected to decline among men [14]. Given the limited availability of forecasts for South America, it remains unclear whether similar trends are expected in this population, highlighting the need for region‐specific projections. The use of the age‐period‐cohort (APC) approach coupled with modelling techniques to forecast projections allows us to separate the effects of age, period, and birth cohort, providing a nuanced understanding of actual dynamics and projecting future mortality rates for all age groups of the population [16, 17]. While it is known that LOOPC mortality rates increase with age, it has been observed that significant variations exist in cohort and period effects in Brazil [18]. Expanding the analysis to include more recent data, particularly with regional stratification, could reveal emerging trends. Therefore, the primary objective of this study was to estimate the future mortality rate related to LOOPC for 2042 in Brazil. The secondary objective was to assess the effect of period and birth cohort on mortality rates in Brazil from 1980 to 2019.
2. Materials and Methods
The present study was reported following the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) for cohort studies.
2.1. Study Design and Setting
This retrospective ecological time‐series study covers all Brazilian municipalities (n = 5570). The research projected the modelled data to 2042 for Brazil and respective regions (South, Southeast, Midwest, Noth, and Northeast). The effect of age, period, and birth cohort on mortality rates related to LOOPC in Brazil from 1980 to 2019 was also investigated. Mortality data were obtained from the Mortality Information System (in Portuguese, Sistema de Informações sobre Mortalidade, SIM) available by the Brazilian National Health System's Information Technology Department (DATA‐SUS; http://www2.datasus.gov.br). Demographic information on the population during the specified timeframe was electronically retrieved from the Brazilian Institute of Geography and Statistics, accessible at https://datasus.saude.gov.br/populacao‐residente. Data collection was performed in January 2024.
2.2. Outcomes
The outcome of the present study was the mortality rate due to LOOPC analysed between 1980 and 2019 and stratified by anatomical location of cancer type (lip/oral and oropharyngeal) and sex. The anatomical sites of the neoplasms were determined according to the codes used in the 9th or 10th edition of the International Classification of Diseases (ICD‐9, ICD‐10) [18, 19] as follows: C00 and 140 (lip); C01‐02 and 141 (tongue); C03 and 143 (gingiva), C04 and 144 (floor of the mouth); and C05‐06 and 145 (other sites of the mouth); C09‐10 and 146 (oropharynx) [20]. Codes C00/140 to C05/145 were analysed together, named ‘lip/oral cancer’. Although the base of the tongue is considered part of the oropharynx [21], DATA‐SUS does not distinguish it from other parts of the tongue between 1979 and 1995. Malignant neoplasms of the nasopharynx, hypopharynx and major salivary glands were not included.
Mortality rates per 100.000 inhabitants were calculated for each age group. All age‐standardised measures were performed considering the reference population as proposed by Segi and modified by Doll [22]. The age‐standardised mortality rate (ASR) was calculated according to cancer (lip/oral and oropharyngeal) and sex in each period evaluated (between 1980 and 2019).
2.3. Statistical Methods
Descriptive analysis was performed estimating the number of cases and ASR. Stata 17 (Stata Corp, College Station, TX, USA) was used to estimate temporal trends through the Prais‐Winsten regression model, applying log‐transformed age‐standardised mortality rates. The regression coefficient (β) was then converted into the annual percent change (APC%) using the formula: APC% = [−1 + (10β)] × 100 [23]. Lexis diagrams were constructed for each outcome considering a rate per 100 000 inhabitants. The outcome was plotted according to age (stratified by period and cohort), period and cohort.
Additional statistical analysis was performed using R version 4.3.2 software (R Core Team, Vienna, Austria) and the packages ‘dplyr’, and ‘Epi’. The analyses of the Age‐Period‐Cohort (APC) models had periods grouped into 5‐year intervals (1980–1984, 1985–1999, …, 2015–2019), totalling eight periods and 24 cohorts (from 1900–1904 to 2015–2019). The variable age was estimated in 5‐year intervals. One of the main challenges in modelling APC effects is their linear dependency, known as the non‐identifiability problem. There is no consensus on the optimal approach to address this issue. In this study, the APC parameters were estimated using deviations, curvatures, and drift, following the method proposed by Holford (1983) [24], which is widely recognised and frequently applied in cancer mortality research. Holford's method focuses on analysing linear combinations and curvatures of the effects. The overall linear trend is decomposed into two components: a primary linear effect attributed to age and drift, which represents the combined linear effects of period and cohort.
A Poisson regression model was employed considering the data distribution. The association effect generated by the ACP model is the relative risk (RR) for each period. The period of reference was 2000–2004 and the reference cohort born was 1960–1964. These references were chosen considering that central cohorts and periods show greater stability. The deviance statistic was used to evaluate the model fit: the model's fit improves as its deviance decreases. The contribution of effects was assessed by comparing the deviance of the estimated model with the specific effect of the full model (age‐period‐cohort). The modelling was performed using the parameter natural splines. Statistical significance was attributed to models with a p < 0.05. For sensitivity analysis, the parameters ‘factor’ to model the regression.
Projections were performed by a non‐linear (in age and time) model named Lee‐Carter‐a model for rates in a Lexis diagram [17]. The Lee‐Carter model is originally defined as a model for rates observed in A‐sets (age by period) of a Lexis diagram, as log(rate(x,t)) = a(x) + b(x)k(t) + εxt, using one parameter per age(x) and period(t). This function uses natural splines for a(), b(), and k(), placing knots for each effect such that the number of events is the same between knots. An ‘ACa’ was used with the reference age 50 years, and the period reference 2019. Projections were performed for 2042 using Bootstrap 1000 simulated samples. The 95% confidence intervals were calculated using the 2.5th and 97.5th percentiles.
3. Results
Over the 40‐year evaluation period, a total of 134 941 deaths attributed to LOOPC cancer were documented within the Brazilian population (Table S1). The mortality for men accounted for 80.1% (n = 108 074) of the total deaths. Specifically, men contributed to 77.5% and 84.7% of fatalities related to lip/oral and oropharyngeal cancers, respectively. ASR was consistently higher in men, regardless of cancer type (Table 1). Prais‐Winsten regression model revealed a slight upward trend in lip and oral cancer mortality among men (p = 0.04) and women (p = 0.02), as well as in oropharyngeal cancer among men (p = 0.02). Lexis diagram for mortality rate is displayed in Tables S2 and S3.
TABLE 1.
Number of cases, age‐standardised rates (ASR), and Prais‐Winsten regression model for estimating trends in ASR and their APC% with 95% confidence intervals (95% CI) according to cancer type and sex.
| No of cases | Lip and oral cancer | Oropharyngeal cancer | ||
|---|---|---|---|---|
| Men | Women | Men | Women | |
| Total % | 66 288 (77.5%) | 19 314 (22.5%) | 41 786 (84.7%) | 7553 (15.3%) |
| ASR | ||||
| 1980–1984 | 2.31 | 0.55 | 1.07 | 0.16 |
| 1985–1989 | 2.18 | 0.49 | 1.32 | 0.23 |
| 1990–1994 | 2.32 | 0.50 | 1.44 | 0.25 |
| 1995–1999 | 2.39 | 0.54 | 1.51 | 0.23 |
| 2000–2004 | 2.52 | 0.54 | 1.65 | 0.22 |
| 2005–2009 | 2.58 | 0.61 | 1.69 | 0.25 |
| 2010–2014 | 2.42 | 0.58 | 1.57 | 0.23 |
| 2015–2019 | 2.49 | 0.58 | 1.64 | 0.23 |
| Prais‐Winsten regression model | |||
|---|---|---|---|
| Cancer type and sex | β (95% CI) | p‐value | APC% (95% CI) a |
| Lip and oral cancer in men | 0.02 (0.00–0.03) | 0.04 | 0.33 (0.08–0.57) |
| Lip and oral cancer in women | 0.02 (0.00–0.04) | 0.02 | 0.47 (0.16–0.77) |
| Oropharyngeal cancer in men | 0.05 (0.01–0.09) | 0.02 | 1.12 (0.46–1.79) |
| Oropharyngeal cancer in women | 0.03 (−0.01–0.07) | 0.19 | 2.94 (−0.95–6.98) |
Note: Brazil, from 1980 and 2019. ASR, age‐standardised mortality rates per 100 000 inhabitants. APC%, Age percentage change for age‐standardised mortality rates.
APC% were estimated for the full period (1980–2019).
3.1. Age‐Period‐Cohort Effect
Table S4 and Figure S1 show the full adjustments in the age‐period‐cohort models. For men, the age‐drift (p < 0.01), age‐cohort (p < 0.01) and age‐period effect (p < 0.01) substantially improve model fit, with the age‐period‐cohort model presenting the best fit (larger decline in deviance), reinforcing the relevance of both cohort and period effects. For women, while neither the age‐cohort nor the age‐period models individually showed a significant improvement in fit, the age‐period‐cohort model was significant (p < 0.01) and presented the lowest deviance. This suggests that although cohort and period effects alone do not strongly influence mortality trends, their combined presence—along with age—provides a better explanation of the observed patterns. Sensitivity analysis showed that age‐period‐cohort maintained the best fit in all analyses (Table S2). Figures S2–S5 present the description of the mortality rate due to LOOPC according to age, period and cohort.
The risk ratios for mortality due to lip/oral and oropharyngeal cancers by cohort and period are presented in Table 2. Regarding the cohort, the results indicate a decline in the risk ratio for men across the most recent cohorts for both lip/oral and oropharyngeal cancers compared to the 1960–1964 cohort. In contrast, among women, there has been a progressive increase in the risk for lip/oral cancer in the more recent cohorts. Regarding the period, a slight increase in oral cancer risk for men and women was observed in recent periods (2005–2009 and 2015–2019 for men; 2005–2009 to 2015–2019 for women), while a decrease in risk was observed in oropharyngeal cancer for men in the periods 2010–2014 and 2015–2019, and women in the period 2015–2019.
TABLE 2.
Risk ratio (RR) and their respective 95% confidence intervals (95% CI) for mortality rate of lip/oral cancer, and oropharyngeal cancer by sex according to Cohort and Period.
| Mortality rates for lip/oral cancer | Mortality rates for oropharyngeal cancer | |||
|---|---|---|---|---|
| Men | Women | Men | Women | |
| RR (95% CI) | RR (95% CI) | RR (95% CI) | RR (95% CI) | |
| Cohort | ||||
| 1900–1904 | 0.8 (0.7–0.8) | 0.7 (0.6–0.8) | 0.5 (0.5–0.5) | 0.8 (0.6–0.9) |
| 1905–1909 | 0.8 (0.7–0.8) | 0.8 (0.7–0.8) | 0.5 (0.5–0.6) | 0.8 (0.7–0.9) |
| 1910–1914 | 0.8 (0.8–0.8) | 0.8 (0.7–0.8) | 0.6 (0.5–0.6) | 0.8 (0.7–0.9) |
| 1915–1919 | 0.8 (0.8–0.9) | 0.8 (0.8–0.9) | 0.6 (0.6–0.6) | 0.8 (0.7–0.9) |
| 1920–1924 | 0.8 (0.8–0.9) | 0.9 (0.8–0.9) | 0.6 (0.6–0.7) | 0.8 (0.7–0.9) |
| 1925–1929 | 0.9 (0.8–0.9) | 0.9 (0.8–1.0) | 0.7 (0.6–0.7) | 0.8 (0.8–0.9) |
| 1930–1934 | 0.9 (0.9–0.9) | 0.9 (0.8–1.0) | 0.7 (0.7–0.7) | 0.8 (0.8–0.9) |
| 1935–1939 | 0.9 (0.9–0.9) | 0.9 (0.9–1.0) | 0.7 (0.7–0.8) | 0.8 (0.8–0.9) |
| 1940–1944 | 0.9 (0.9–1.0) | 0.9 (0.9–1.0) | 0.8 (0.8–0.8) | 0.9 (0.8–0.9) |
| 1945–1949 | 1.0 (0.9–1.0) | 0.9 (0.9–1.0) | 0.8 (0.8–0.9) | 0.9 (0.8–1.0) |
| 1950–1954 | 1.0 (1.0–1.1) | 1.0 (0.9–1.0) | 0.9 (0.9–1.0) | 1.0 (0.9–1.0) |
| 1955–1959 | 1.1 (1.0–1.1) | 1.0 (1.0–1.0) | 1.0 (1.0–1.0) | 1.0 (1.0–1.0) |
| 1960–1964 a | Ref. | Ref. | Ref. | Ref. |
| 1965–1969 | 0.9 (0.9–0.9) | 1.0 (1.0–1.1) | 0.9 (0.9–0.9) | 1.0 (1.0–1.0) |
| 1970–1974 | 0.8 (0.8–0.8) | 1.1 (1.0–1.1) | 0.8 (0.8–0.9) | 1.0 (0.9–1.1) |
| 1975–1979 | 0.7 (0.7–0.8) | 1.1 (1.0–1.2) | 0.8 (0.7–0.8) | 1.0 (0.9–1.1) |
| 1980–1984 | 0.7 (0.6–0.7) | 1.1 (1.0–1.2) | 0.7 (0.6–0.7) | 1.0 (0.9–1.2) |
| 1985–1989 | 0.6 (0.5–0.6) | 1.2 (1.0–1.3) | 0.6 (0.6–0.7) | 1.0 (0.9–1.2) |
| 1990–1994 | 0.5 (0.5–0.6) | 1.2 (1.0–1.3) | 0.6 (0.5–0.6) | 1.0 (0.8–1.2) |
| 1995–1999 | 0.5 (0.4–0.5) | 1.2 (1.0–1.4) | 0.5 (0.5–0.6) | 1.0 (0.8–1.3) |
| 2000–2004 | 0.4 (0.4–0.5) | 1.3 (1.1–1.5) | 0.5 (0.4–0.5) | 1.0 (0.8–1.3) |
| 2005–2009 | 0.4 (0.3–0.4) | 1.3 (1.1–1.6) | 0.4 (0.4–0.5) | 1.0 (0.8–1.4) |
| 2010–2014 | 0.3 (0.3–0.4) | 1.3 (1.1–1.6) | 0.4 (0.3–0.5) | 1.0 (0.7–1.4) |
| 2015–2019 | 0.3 (0.3–0.4) | 1.4 (1.1–1.7) | 0.4 (0.3–0.4) | 1.0 (0.7–1.5) |
| Period | ||||
| 1980–1984 | 1.0 (1.0–1.0) | 1.1 (1.0–1.1) | 0.9 (0.8–0.9) | 0.9 (0.8–0.9) |
| 1985–1989 | 1.0 (0.9–1.0) | 1.0 (1.0–1.0) | 0.9 (0.9–0.9) | 0.9 (0.9–1.0) |
| 1990–1994 | 0.9 (0.9–1.0) | 1.0 (0.9–1.0) | 1.0 (0.9–1.0) | 1.0 (0.9–1.1) |
| 1995–1999 | 1.0 (0.9–1.0) | 1.0 (0.9–1.0) | 1.0 (1.0–1.0) | 1.0 (1.0–1.1) |
| 2000–2004 a | Ref. | Ref. | Ref. | Ref. |
| 2005–2009 | 1.0 (1.0–1.0) | 1.1 (1.0–1.1) | 1.0 (0.9–1.0) | 1.0 (0.9–1.0) |
| 2010–2014 | 0.9 (0.9–1.0) | 1.0 (1.0–1.1) | 0.9 (0.8–0.9) | 0.9 (0.9–1.0) |
| 2015–2019 | 1.0 (1.0–1.0) | 1.0 (1.0–1.0) | 0.9 (0.9–0.9) | 0.9 (0.8–0.9) |
Note: The analysis is derived from Age‐Period‐Cohort Analysis.
The period of reference (Ref.) was 2000–2004 and the reference cohort born was 1960–1964.
3.2. Projections
Projections for Brazil in 2042 show a slight, but significant, decrease in lip/oral cancer mortality rates for men with ages between 40 and 60 years old when compared with rates of 2019. A significant decrease in oropharyngeal cancer in men is projected for 2042 for individuals between 35 and 60 years old (Figure 1). For women, no significant changes are projected to 2014 for both cancers at any age.
FIGURE 1.
Mortality rates per 100 000 inhabitants due to lip/oral and oropharyngeal cancer for 2019 and projections to 2042 in men and women.
Figure 2 displays the projection models for LOOPC stratified by sex and Brazilian regions. The model projections mortality rate reveal varied outcomes across the diverse regions of Brazil. While an increase in mortality rates due to lip/oral cancer is projected for men over 60 years old in the Northeast, a significant decrease is observed in men aged 30–50 in the Southeast, in men around 65 years old in the South, and in men aged 50–60 in the Midwest. An increase in mortality from oropharyngeal cancer is projected for men over 50 in the Northeast, while a decrease is observed in men aged 30–70 in the Southeast and at 65 years old in the South. No other significant differences were projected.
FIGURE 2.
Mortality rates per 100 000 inhabitants due to lip/oral and oropharyngeal cancer for 2019 and projections to 2042 in men and women according to Brazilian Region.
4. Discussion
The present findings provide new insights regarding the LOOPC cancer mortality rate, representing the longest mortality study of those cancers and the first to perform forecasting with Brazilian data. The main finding from the projected model indicated a reduction in LOOPC mortality among men, particularly in their fourth and sixth decades of life, while no changes are projected for women. These results partially align with projections from Spain for 2040, which forecast a decrease in oropharyngeal cancer mortality in men and an increase in both lip/oral and oropharyngeal cancers in women [14]. Significant regional variations were observed in the present study, with an increase in LOOPC mortality among men in the Northeast and a decrease in the Southeast and South regions. These findings can be explained by several Brazilian sociodemographic factors, including the notable economic disparities between macro‐regions, which are crucial points for consideration in the development of public health policies.
It is important to acknowledge potential limitations in the study. As an ecological study, the findings are based on aggregated data rather than individual‐level information, which prevents the establishment of direct causal relationships. Also, this approach does not allow for the inclusion of risk factors in the model, which represents an important limitation to consider when interpreting the data. Nonetheless, ecological studies play a crucial role in public health by identifying broad trends and informing cancer prevention and control policies. Variations in anatomical site classification for oral and oropharyngeal cancer in recent literature may reduce comparability among studies. Despite efforts to mitigate under‐reporting of deaths by the Brazilian Unified Health System's Informatic Department, limitations inherent in national mortality databases may persist. Furthermore, the projections presented in this study are subject to uncertainties, as they are based on modelled data that do not account for unforeseen changes in risk factors or healthcare interventions. Consequently, these estimates may either overestimate or underestimate future mortality trends, as they do not incorporate emerging exposures or preventive measures that could impact the cancer burden in Brazil. Although the Age‐Period‐Cohort model accounts for variations in age and cohort effects, the projections themselves are based on the assumption that current trend effects remain stable. Thus, the dynamic demographic shifts, such as population aging or changes in birth cohort sizes, are not explicitly modelled. Additionally, while the Age‐Period‐Cohort approach used in this study addresses the mathematical non‐identifiability problem, it does not resolve the inherent dependency in reality, meaning that trends observed in a given age group may still reflect period or cohort effects that cannot be fully decomposed.
A significant age, period, and cohort effect on mortality rates was observed in the present study. Age strongly influences mortality rates, which is expected in chronic diseases [25]. The increased mortality rate among the elderly highlights the age effect. Also, men experience higher rates and an earlier onset of an upward trajectory for LOOPC, which can be explained by the higher exposure to risk factors for a more extended duration compared to women [9, 26]. Period effects were less pronounced, with small changes occurring in recent generations and with an increase in oral/lip cancer for both sexes, while a decrease is observed for oropharyngeal cancer. This reduction of oropharyngeal cancer could be explained by the HPV vaccination, which still finds important regional differences in its coverage [27], as well as the continuous preventive activities with the population [28]. The cohort effect reveals a decreasing trend in the risk ratio for men across the most recent cohorts for both lip/oral and oropharyngeal cancers. On the other hand, among women, there has been a continuous increase in the risk for lip/oral cancer in the more recent cohorts.
When stratifying the analyses for Brazilian macro‐regions, different patterns are observed in the forecasting. The southeast and south region, the macro‐regions with the higher human development index (HDI), show a decrease in the projection for LOOPC in men. Otherwise, the northeast, which presents the lower HDI, shows an increase in the mortality rate of LOOPC for men. These findings highlight the crucial importance of understanding the behaviour of oral cancer in a country with remarkable inequalities. These socioeconomic discrepancies are intrinsically related to access to health services and must guide public health policies toward screening and controlling disease. Moreover, areas of high UV index such as those observed in the northeast region could also be a regional factor in explaining the increased mortality of lip cancer [29].
It's crucial to recognise that the projections are based on certain assumptions, rendering them potentially imprecise. Projection models rely on modelled data, rendering them susceptible to significant inaccuracies, as they cannot predict events deviating from the norm. For example, the potential increase in electronic cigarette use already observed among adolescents could not be incorporated into the projections due to its recent observation in the Brazilian population [30]. However, it is noteworthy that the estimations were performed considering age‐period‐cohort modelling using data from 40 years of available information from all 5570 Brazilian municipalities. It is important to acknowledge that these projections may either underestimate or overestimate the real burden of these cancers. This can also be explained by the non‐inclusion in the models of factors such as population exposures, known and unknown risk factors, and potential preventive measures that can be directed in the future to the Brazilian population. Despite these limitations, these projections offer a valuable tool for public health policies, which can prioritise resources and strategically plan cancer control efforts.
Few studies forecasting the mortality of oral and oropharyngeal cancer are available, and none of them were performed in the South American population, making comparisons difficult [14, 15, 31]. Infante‐Cossio et al. [14] analysed the mortality rates for oral and oropharyngeal cancer between 1980 and 2019, and generated a projection model for the next 25 years in Spain. The authors observed more deaths in females than in males for oral cancer in the period 2040–2044, while deaths for oropharyngeal cancer will decrease in males and gradually increase in females. Similarly, a large average annual percentage increase in deaths is forecasted from oral cancer (males: 2.97%, females 3.09%) from 2015 to 2035, based on 1979–2014 UK data [15]. Also, Alsharif et al. [31] assess the mortality rates of head and neck cancer (HNC) within the Gulf Cooperation Council countries. Their projections for 2040 reveal a substantial increase in related deaths, being higher among females compared to males in most countries [15].
The observed regional disparities in LOOPC mortality highlight the need for both universal and targeted public health policies. Universal measures—such as strengthening tobacco and alcohol control policies, and addressing broader social determinants of health—remain essential to reduce the overall burden of disease. However, regions with rising projected mortality rates, such as the Northeast, may benefit from targeted interventions tailored to local epidemiological and socioeconomic contexts. Additionally, the growing burden of lip/oral cancer among women suggests the necessity of gender‐specific interventions and awareness campaigns. These projections provide critical insights for long‐term planning and resource allocation, emphasising the reduction of health inequities across Brazilian regions.
Author Contributions
Luiz Alexandre Chisini conceived the ideas, collected the data, analysed the data, and wrote the paper. Luana Carla Salvi collected the data and wrote the paper. Francine da costa Santos conceived the ideas and reviewed the paper. Rodrigo Varella de Carvalho conceived the ideas and reviewed the paper. Ana Carolina Uchoa Vasconcelos conceived the idea and contributed to the writing and reviewing process.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Data S1: Supporting Information.
Table S1: Number of cases according to cancer type and sex. Brazil, from 1980 to 2019.
Table S2: Lexis diagram for mortality rates per 100 000 inhabitants due to lip/oral and oropharyngeal cancer according to sex by age‐cohort, Brazil, 1980–2019.
Table S3: Lexis diagram for mortality rates per 100 000 inhabitants due to lip/oral and oropharyngeal cancer according to sex by age‐period, Brazil, 1980–2019.
Table S4: Adjustments in the APC models for mortality rates due to lip/oral and oropharyngeal cancer according to sex, Brazil, 1982–2019.
Table S5: Sensitivity analysis for the adjustments in the APC models for mortality rates due to lip/oral and oropharyngeal cancer according to sex, Brazil, 1982–2021. We used the parameters ‘ns’ and ‘bs’ to model the regression.
Figure S1: Age, period and cohort effects in mortality rates due to lip/oral and oropharyngeal cancer in Brazil according to sex, 1980–2019.
Figure S2: Mortality rates per 100 000 inhabitants due to lip/oral cancer in men according to age, period, and cohort, Brazil, 1980–2019. (a) Mortality rate by age; each line represents a different period. (b) Mortality rate by age; each line represents a different cohort. (c) Mortality rate by period; each line represents a different age. (d) Mortality rate by cohort; each line represents a different age.
Figure S3: Mortality rates per 100 000 inhabitants due to lip/oral cancer in women according to age, period, and cohort, Brazil, 1980–2019. (a) Mortality rate by age; each line represents a different period. (b) Mortality rate by age; each line represents a different cohort. (c) Mortality rate by period; each line represents a different age. (d) Mortality rate by cohort; each line represents a different age.
Figure S4: Mortality rates per 100 000 inhabitants due to oropharyngeal cancer in men according to age, period, and cohort, Brazil, 1980–2019. (a) Mortality rate by age; each line represents a different period. (b) Mortality rate by age; each line represents a different cohort. (c) Mortality rate by period; each line represents a different age. (d) Mortality rate by cohort; each line represents a different age.
Figure S5: Mortality rates per 100 000 inhabitants due to oropharyngeal cancer in women according to age, period, and cohort, Brazil, 1980–2019. (a) Mortality rate by age; each line represents a different period. (b) Mortality rate by age; each line represents a different cohort. (c) Mortality rate by period; each line represents a different age. (d) Mortality rate by cohort; each line represents a different age.
Acknowledgements
This study was conducted in a Graduate Programme supported by CAPES, Brazil.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. GBD 2019 Cancer Risk Factors Collaborators , “The Global Burden of Cancer Attributable to Risk Factors, 2010‐19: A Systematic Analysis for the Global Burden of Disease Study 2019,” Lancet 400, no. 10352 (2022): 563–591. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Ferlay J., Colombet M., Soerjomataram I., et al., “Estimating the Global Cancer Incidence and Mortality in 2018: GLOBOCAN Sources and Methods,” International Journal of Cancer 144, no. 8 (2019): 1941–1953. [DOI] [PubMed] [Google Scholar]
- 3. Serna B. Y. H., Betancourt J. A. O., Soto O. P. L., Amaral R. C. D., and Correa M., “Trends of Incidence, Mortality, and Disability‐Adjusted Life Years of Oral Cancer in Latin America,” Revista Brasileira de Epidemiologia 25 (2022): e220034. [DOI] [PubMed] [Google Scholar]
- 4. Bray F., Ferlay J., Soerjomataram I., Siegel R. L., Torre L. A., and Jemal A., “Global Cancer Statistics 2018: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries,” CA: A Cancer Journal for Clinicians 68, no. 6 (2018): 394–424. [DOI] [PubMed] [Google Scholar]
- 5. Mello F. W., Melo G., Pasetto J. J., Silva C. A. B., Warnakulasuriya S., and Rivero E. R. C., “The Synergistic Effect of Tobacco and Alcohol Consumption on Oral Squamous Cell Carcinoma: A Systematic Review and Meta‐Analysis,” Clinical Oral Investigations 23, no. 7 (2019): 2849–2859. [DOI] [PubMed] [Google Scholar]
- 6. Gilligan G., Panico R., Lazos J., et al., “Oral Squamous Cell Carcinomas and Oral Potentially Malignant Disorders: A Latin American Study,” Oral Diseases 30, no. 5 (2023). [DOI] [PubMed] [Google Scholar]
- 7. D'Souza G., Tewari S. R., Troy T., et al., “Oncogenic Oral Human Papillomavirus Clearance Patterns Over 10 Years,” Cancer Epidemiology, Biomarkers & Prevention 33 (2024): 516–524. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Petti S., “Lifestyle Risk Factors for Oral Cancer,” Oral Oncology 45, no. 4–5 (2009): 340–350. [DOI] [PubMed] [Google Scholar]
- 9. Dai X., Gakidou E., and Lopez A. D., “Evolution of the Global Smoking Epidemic Over the Past Half Century: Strengthening the Evidence Base for Policy Action,” Tobacco Control 31, no. 2 (2022): 129–137. [DOI] [PubMed] [Google Scholar]
- 10. Garolla A., Graziani A., Grande G., Ortolani C., and Ferlin A., “HPV‐Related Diseases in Male Patients: An Underestimated Conundrum,” Journal of Endocrinological Investigation 47 (2023): 261–274. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Sathish N., Wang X., and Yuan Y., “Human Papillomavirus (HPV)‐Associated Oral Cancers and Treatment Strategies,” Journal of Dental Research 93, no. 7 Suppl (2014): 29S–36S. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Moran‐Torres A., Pazos‐Salazar N. G., Tellez‐Lorenzo S., et al., “HPV Oral and Oropharynx Infection Dynamics in Young Population,” Brazilian Journal of Microbiology 52, no. 4 (2021): 1991–2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Tsai Y. S., Chen Y. C., Chen T. I., et al., “Incidence Trends of Oral Cavity, Oropharyngeal, Hypopharyngeal and Laryngeal Cancers Among Males in Taiwan, 1980‐2019: A Population‐Based Cancer Registry Study,” BMC Cancer 23, no. 1 (2023): 213. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Infante‐Cossio P., Duran‐Romero A. J., Castano‐Seiquer A., Martinez‐De‐Fuentes R., and Pereyra‐Rodriguez J. J., “Estimated Projection of Oral Cavity and Oropharyngeal Cancer Deaths in Spain to 2044,” BMC Oral Health 22, no. 1 (2022): 444. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Olsen A. H., Parkin D. M., and Sasieni P., “Cancer Mortality in the United Kingdom: Projections to the Year 2025,” British Journal of Cancer 99, no. 9 (2008): 1549–1554. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Carstensen B., “Age‐Period‐Cohort Models for the Lexis Diagram,” Statistics in Medicine 26, no. 15 (2007): 3018–3045. [DOI] [PubMed] [Google Scholar]
- 17. Lee R. D. and Carter L. R., “Modelling and Forecasting U.S. Mortality,” Journal of the American Statistical Association 87 (1992): 659–671. [Google Scholar]
- 18. Perea L. M. E., Antunes J. L. F., and Peres M. A., “Oral and Oropharyngeal Cancer Mortality in Brazil, 1983‐2017: Age‐Period‐Cohort Analysis,” Oral Diseases 28, no. 1 (2022): 97–107. [DOI] [PubMed] [Google Scholar]
- 19. Fritz A., Percy C., Jack A., et al., International Classification of Diseases for Oncology, 3rd ed. (World Health Organisation, 2000), 240. [Google Scholar]
- 20. Duran‐Romero A. J., Infante‐Cossio P., and Pereyra‐Rodriguez J. J., “Trends in Mortality Rates for Oral and Oropharyngeal Cancer in Spain, 1979‐2018,” Oral Diseases 28, no. 2 (2022): 336–344. [DOI] [PubMed] [Google Scholar]
- 21. Kato M. G., Baek C. H., Chaturvedi P., et al., “Update on Oral and Oropharyngeal Cancer Staging ‐ International Perspectives,” World Journal of Otorhinolaryngology‐Head and Neck Surgery 6, no. 1 (2020): 66–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Seigi M., Cancer mortality for selected sites in 24 countries (1950–1957) (Department of Public Health, Tohoku University School of Medicine, 1960). [Google Scholar]
- 23. Ricci J. M. S., Romito A. L. Z., Silva S. A. D., Carioca A. A. F., and Lourenco B. H., “Food Intake Markers in Sisvan: Temporal Trends in Coverage and Integration With e‐SUS APS, Brazil 2015‐2019,” Ciência & Saúde Coletiva 28, no. 3 (2023): 921–934. [DOI] [PubMed] [Google Scholar]
- 24. Holford T. R., “The Estimation of Age, Period and Cohort Effects for Vital Rates,” Biometrics 39, no. 2 (1983): 311–324. [PubMed] [Google Scholar]
- 25. Ebeling M., Rau R., Malmstrom H., Ahlbom A., and Modig K., “The Rate by Which Mortality Increase With Age Is the Same for Those Who Experienced Chronic Disease as for the General Population,” Age and Ageing 50, no. 5 (2021): 1633–1640. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Hussein A. A., Helder M. N., de Visscher J. G., et al., “Global Incidence of Oral and Oropharynx Cancer in Patients Younger Than 45 Years Versus Older Patients: A Systematic Review,” European Journal of Cancer 82 (2017): 115–127. [DOI] [PubMed] [Google Scholar]
- 27. Moura L. L., Codeco C. T., and Luz P. M., “Human Papillomavirus (HPV) Vaccination Coverage in Brazil: Spatial and Age Cohort Heterogeneity,” Revista Brasileira de Epidemiologia 24 (2020): e210001. [DOI] [PubMed] [Google Scholar]
- 28. Cunha A. R. D., Prass T. S., and Hugo F. N., “Mortality From Oral and Oropharyngeal Cancer in Brazil: Impact of the National Oral Health Policy,” Cadernos de Saúde Pública 35, no. 12 (2019): e00014319. [DOI] [PubMed] [Google Scholar]
- 29. Unsal A. A., Unsal A. B., Henn T. E., Baredes S., and Eloy J. A., “Cutaneous Squamous Cell Carcinoma of the Lip: A Population‐Based Analysis,” Laryngoscope 128, no. 1 (2018): 84–90. [DOI] [PubMed] [Google Scholar]
- 30. Silva A. and Moreira J. C., “The Ban of Eletronic Cigarettes in Brazil: Success or Failure?,” Ciência & Saúde Coletiva 24, no. 8 (2019): 3013–3024. [DOI] [PubMed] [Google Scholar]
- 31. Alsharif A., Alsharif M. T., Samman M., et al., “Forecasting Head and Neck Cancer Trends in GCC Countries: Implications for Public Health Policy and Strategy,” Risk Management and Healthcare Policy 16 (2023): 2943–2952. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data S1: Supporting Information.
Table S1: Number of cases according to cancer type and sex. Brazil, from 1980 to 2019.
Table S2: Lexis diagram for mortality rates per 100 000 inhabitants due to lip/oral and oropharyngeal cancer according to sex by age‐cohort, Brazil, 1980–2019.
Table S3: Lexis diagram for mortality rates per 100 000 inhabitants due to lip/oral and oropharyngeal cancer according to sex by age‐period, Brazil, 1980–2019.
Table S4: Adjustments in the APC models for mortality rates due to lip/oral and oropharyngeal cancer according to sex, Brazil, 1982–2019.
Table S5: Sensitivity analysis for the adjustments in the APC models for mortality rates due to lip/oral and oropharyngeal cancer according to sex, Brazil, 1982–2021. We used the parameters ‘ns’ and ‘bs’ to model the regression.
Figure S1: Age, period and cohort effects in mortality rates due to lip/oral and oropharyngeal cancer in Brazil according to sex, 1980–2019.
Figure S2: Mortality rates per 100 000 inhabitants due to lip/oral cancer in men according to age, period, and cohort, Brazil, 1980–2019. (a) Mortality rate by age; each line represents a different period. (b) Mortality rate by age; each line represents a different cohort. (c) Mortality rate by period; each line represents a different age. (d) Mortality rate by cohort; each line represents a different age.
Figure S3: Mortality rates per 100 000 inhabitants due to lip/oral cancer in women according to age, period, and cohort, Brazil, 1980–2019. (a) Mortality rate by age; each line represents a different period. (b) Mortality rate by age; each line represents a different cohort. (c) Mortality rate by period; each line represents a different age. (d) Mortality rate by cohort; each line represents a different age.
Figure S4: Mortality rates per 100 000 inhabitants due to oropharyngeal cancer in men according to age, period, and cohort, Brazil, 1980–2019. (a) Mortality rate by age; each line represents a different period. (b) Mortality rate by age; each line represents a different cohort. (c) Mortality rate by period; each line represents a different age. (d) Mortality rate by cohort; each line represents a different age.
Figure S5: Mortality rates per 100 000 inhabitants due to oropharyngeal cancer in women according to age, period, and cohort, Brazil, 1980–2019. (a) Mortality rate by age; each line represents a different period. (b) Mortality rate by age; each line represents a different cohort. (c) Mortality rate by period; each line represents a different age. (d) Mortality rate by cohort; each line represents a different age.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.