The impact of reducing alcohol consumption in Australia: An estimate of the proportion of potentially avoidable cancers 2013–2037

Abstract

first concluded that alcohol causes cancer in humans in 1988. The World Cancer Research Fund has declared that alcohol causes cancer of the oral cavity, pharynx, larynx, oesophagus (squamous cell carcinoma), female breast, colon, rectum, stomach and liver. It recommended that alcohol be avoided altogether to prevent cancer. We aimed to quantify the impact of reducing alcohol consumption on future cancer incidence in Australia. We used PREVENT 3.01 simulation modelling software to estimate the proportion of cancers that could potentially be prevented over a 25-year period under two hypothetical intervention scenarios and two latency periods (20 and 30 years). Under a scenario where alcohol consumption abruptly ceases, we estimated up to 4% of alcohol-related cancers could be avoided over a 25-year period (~49,500 cancers, depending on assumed latency). If the maximum consumption of all Australian adults was ≤20 g/day (~two Australian standard drinks), up to 2% of alcohol-related cancers could be avoided (~29,600 cancers). The maximum proportions were higher for men (6% for no alcohol consumption; 5% for ≤20 g/day) than women (3%; 1%). The proportion avoidable was highest for oesophageal squamous cell carcinoma (17% no alcohol consumption; 9% ≤20 g/day), followed by cancers of the oral cavity (12%; 5%) and pharynx (11%; 5%). The cancer sites with the highest numbers of potentially avoidable cases were colon in men (11,500; 9,900) and breast in women (14,400; 4,100). Successful interventions to reduce alcohol intake could lead to significant reductions in cancer incidence.

Impact of reducing alcohol consumption through price‐based policies on cancer incidence in Germany 2020–50—a simulation study

“…In populations in which liquor consumption is more prevalent, alcohol price increase on a volumetric basis would probably yield a higher impact than in the policy scenarios based on German consumption data. Several international studies [23][24][25][26] have used similar methods to assess the impact of reducing alcohol consumption on future cancer incidence; however, a direct comparison is difficult due to different modeling assumptions. For example, using the Prevent macro-simulation model and assuming a 20-year latency period, Wilson et al [25] estimated that the alcohol-related cancer burden in Australia could be reduced by 2.3% (men = 4.8%, women = 0.9%) during a 25-year period, if the proportion of the population drinking > 20 g/day of alcohol could diminish to zero over 5 years.…”

“…Several international studies [23][24][25][26] have used similar methods to assess the impact of reducing alcohol consumption on future cancer incidence; however, a direct comparison is difficult due to different modeling assumptions. For example, using the Prevent macro-simulation model and assuming a 20-year latency period, Wilson et al [25] estimated that the alcohol-related cancer burden in Australia could be reduced by 2.3% (men = 4.8%, women = 0.9%) during a 25-year period, if the proportion of the population drinking > 20 g/day of alcohol could diminish to zero over 5 years. Our estimates of 4.7% (men = 10.1%, women = 1.4%) are larger, which is due to the use of a shorter latency period, a longer study period as well as sex-specific thresholds for the definition of high alcohol intake (men = > 20 g/day of alcohol, women = > 10 g/day of alcohol).…”

“…With respect to the interval between the reduction in the intake of alcohol and the complete decline in cancer risk, there is little evidence on the true lengths of these time lags. For cancers of the upper aerodigestive tract, different studies reported latency periods for cessation of drinking of more than 20 years [4,25]. As we did not consider cessation of drinking in any investigated policy scenario, we decided to use latency periods of approximately half of those reported interval lengths.…”

“…In similar studies examining the impact of reducing alcohol consumption on future cancer incidence [23][24][25][26], LAT and LAG times in the range from 1 to 10 years and from 4 to 20 years, respectively, were used. For our main analysis, we used a LAT time of 1 year and a LAG time of 9 years, and assumed a linear decline in excess cancer risk during the LAG period.…”

Variaciones hipotéticas en la incidencia y la fracción atribuible poblacional de cánceres hepático, colorrectal y de mama, como efectos de la reducción teórica del índice de masa corporal en Colombia, 2016 - 2050

“…simulaciones, siendo estas el "tiempo lat" (entendido como el tiempo transcurrido entre el momento en el que inicia la intervención sobre la exposición al factor de riesgo y el momento en el que aparece el efecto, sin que este periodo de cambio en la exposición se traduzca en un cambio notable en la incidencia -que en nuestro estudio tuvo una asignación de 3 años-) y el "tiempo lag" (correspondiente al periodo transcurrido desde el final del lat hasta la obtención del efecto total en términos del rr, que es lo mismo que el tiempo de duración del efecto de la intervención sobre el factor de riesgo -designado con valor de 20 años en nuestras simulaciones-) [13,19,30]. Prevent v3.01 leyó el archivo elaborado en Access® 2007 y generó simulaciones con las proyecciones de la incidencia esperada para cada cáncer según los escenarios propuestos.…”