Preventable burden of head and neck cancer attributable to tobacco and alcohol between 1990 and 2039 in China

Abstract

Tobacco use and heavy alcohol consumption are risk factors for head and neck cancer (HNC), including oral, pharynx, and larynx cancer. No study has investigated the preventable burden of HNC attributable to tobacco and alcohol in China. We extracted data from 1990 to 2019 from the Global Burden of Disease. The preventable burden attributable to tobacco and alcohol was estimated by subtracting the overlapping fraction derived from a literature search. Descriptive analyses were performed initially, followed by joinpoint regression and age-period-cohort (APC) analysis. The future burden was forecasted using a Bayesian APC model. The crude burden increased significantly, while the age-standardized rates showed a downward trend from 1990 to 2019 in China. Both all-age and age-standardized population attributable fractions rose significantly, potentially due to the poor prognosis of tobacco- and alcohol-associated HNC. The absolute burden would continue to climb in the next 20 years from 2019, largely due to population aging. For site-specific burden, compared with total, pharynx, and larynx cancer burden, the substantial upward trend of oral cancer burden indicated a strong interaction with risk factors such as genetic susceptibility, betel nut chewing, oral microbiota, and human papillomavirus. The burden of oral cancer attributable to tobacco and alcohol is a major concern and is anticipated to become more severe than cancer in other anatomic sites. Altogether, our study provides useful information to rethink the current restrictions on tobacco and alcohol, lean healthcare resources, and develop effective HNC prevention and control strategies.

Keywords: age-period-cohort analysis; decomposition analysis; head and neck cancer; preventable burden; tobacco and alcohol.

FIGURE 1

Effects of age, period, and cohort and the resulting local drifts on head and neck cancer death attributable to tobacco and alcohol from 1990 to 2019 by anatomic site. (A) Local drifts of different age groups. (B) Fitted longitudinal age curves of mortality. (C) Relative risks of each time period referenced with 2000–2004. (D) Relative risk of each cohort compared with the reference cohort 1945–1949. The solid lines and shaded areas represent each metric and their 95% confidence interval (CI).

FIGURE 2

Effects of age, period, and cohort and the resulting local drifts on head and neck cancer DALYs attributable to tobacco and alcohol from 1990 to 2019 by anatomic site. (A) Local drifts of different age groups. (B) Fitted longitudinal age curves of disability‐adjusted life year (DALY) rate. (C) Relative risks of each time period referenced with 2000–2004. (D) Relative risk of each cohort compared with the reference cohort 1945–1949. The solid lines and shaded areas represent each metric and their 95% confidence interval (CI).

FIGURE 3

The time trends of age‐standardized mortality rate (ASMR) and average annual percent change (ASDR) for head and neck cancer attributable to tobacco and alcohol between 1990 and 2019 and their forecasts until 2039. (A) ASMR. (B) ASDR. The dotted lines show the observed values, and the solid lines represent the predicted mean values. The fan demonstrates the predictive distribution between 5% and 95% quintiles. The shaded bands represent the intervals in an increment of 10%.

FIGURE 4

The temporal trends and contribution changes of the number of head and neck cancer death and disability‐adjusted life years (DALYs) attributable to tobacco and alcohol between 1990 and 2019 and their forecasts until 2039. (A) The trend of the total death number. (B) Contribution of changes in population aging, population growth, and age‐specific rate in the total death number. The solid and empty dots represent observed and predicted values. (C) The trend of the total DALY. (D) Contribution of changes in population aging, population growth, and age‐specific rate in the total DALY. The solid dots and empty dots represent observed and predicted values.

FIGURE 5

The proportion of the absolute burden for each anatomic site attributable to tobacco and alcohol between 1990 and 2019 and their forecasts until 2039. (A) Changes in the proportion of the death number at each anatomic site. (B) Changes in the proportion of disability‐adjusted life year (DALY) at each anatomic site. The solid dots and empty dots represent observed and predicted values.