Year: 2019
Type: letter
Abstract: We thank Dr. Ranjan for his interest in our article.1 However, we firmly stand behind our results and conclusion, including that drinking ≥700 ml/day hot tea (defined in this study as tea at temperature ≥60°C) has shown consistent associations with esophageal squamous cell carcinoma (ESCC) risk, and that “it will be reasonable to advise people in Golestan and elsewhere to wait for their hot beverages to cool down before drinking”, as stated in the “Conclusions” section of the “Discussion” in our article. In our study, drinking tea at temperatures <60°C was the reference group, and we did not report any increased risk associated with drinking tea at these lower temperatures. As such, in no part of the article have we asked people to stop drinking tea. In the “What's new?” section mentioned by Dr. Ranjan, the recommendation is “to advise the public to wait for beverages to cool to <60°C before consumption”. Even when we reviewed the media report mentioned by Dr. Ranjan, the recommendation was “it is therefore advisable to wait until hot beverages cool down before drinking.”2 It is therefore unclear to us how Dr. Ranjan has reached his interpretation of our study. We also appreciate Dr. Ranjan's efforts to indirectly differentiate any effects of smoking and hot tea drinking on ESCC risk. However, as we have mentioned in the “Methods” section of our article, we did several additional analyses, including an analysis limited to only those who never smoked and never drank alcohol. As we have reported in the “Results” section, “The HRs were generally similar in additional subgroup analyses by sex or history of cigarette and alcohol use … (Supporting Information Table S3).” In other words, results were similar when we limited our analysis to never-smokers and never-alcohol drinkers only, supporting our conclusion that the association between hot tea drinking and esophageal cancer is independent from smoking and alcohol drinking status. The number of cancer cases and unadjusted and adjusted hazard ratios in our additional analyses are presented in Supporting Information Table S3 of our article. For Dr. Ranjan's information, in case he has not already reviewed that table, among never-smokers/never-alcohol drinkers the number of ESCC cases drinking ≥700 ml/day tea at ≥60°C was 152 cases. We are not sure if we clearly understand the concern raised in the third paragraph of Dr. Ranjan's letter. However, in a prospective cohort study, like our study, participants recruited at the baseline include both those who are exposed to the exposure of interest and those who are not (in our study, hot tea drinkers and not-drinkers.) Then the participants are followed over time, and the risk of health outcome of interest (incident ESCC cases in our study) by exposure status is calculated after the follow-up (in our study, a median of over 10 years). We calculated the risk after adjustments for multiple potentially confounding factors, including sociodemographic, dietary and lifestyle factors listed in Table 1 of our article. Moreover, we repeated our multivariate analyses after stratification by sex and smoking/alcohol drinking status to eliminate residual confounding from these important factors. We think this is similar to what Dr. Ranjan suggests. Results from these additional analyses were comparable to our overall results. We agree with Dr. Ranjan that the association between hot beverage drinking and oral cancer should be studied. Further information is also needed on the usual duration of hot beverages remaining in the oral cavity before being swallowed, because if people swallow their hot beverages quickly, the risk of any thermal injury to the oral cavity might decline while the duration of the hyperthermal episode in the esophagus might increase.3 We also agree with Dr. Ranjan that epidemiological evidence could be misinterpreted. One example is a “protective effect” of tea drinking vs. esophageal cancer. As we have reviewed in our article, from six prospective cohort studies on the effect of the amount of tea (five previously published plus our article), only one study showed an inverse association, whereas the other five showed no inverse associations. Similarly, none of three cohort studies and only one out of four case–control studies reviewed in one of the articles cited by Dr. Ranjan did show an inverse association between green tea drinking and bladder cancer risk in humans.4 Interestingly, in another article cited by Dr. Ranjan,5 no association between green tea drinking and gastric cancer was observed in prospective cohort studies (relative risk 1.05; 95% confidence interval [CI], 0.90–1.21), and an inverse association was observed only in case–control studies (odds ratio, 0.84; 95% CI, 0.74–0.95); the dose–response analysis was mostly based on case–control studies. It should be noted that although case–control studies are important to study very rare diseases or to generate hypotheses, because of a strong probability of recall bias and higher possibility of selection bias in case–control studies, prospective cohort studies provide more reliable information on the association between dietary factors and cancer. We think it is important to carefully review evidence from the published literature with considering its strengths and limitations to make an unbiased conclusion. In our article, we did not make a conclusion on the association between amount of tea consumed at lower temperatures and ESCC risk and recommended further research on this matter. Yours sincerely
Source: International Journal of Cancer
DOI: 10.1002/ijc.32342
Authors: Farhad Islami, Farin Kamangar, Reza Malekzadeh
Language: English
Subfield: Surgery
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W2965219834