Life-course approach to behavioural risk factors and quality of life for cancers of the upper aero-digestive tract in an Indian population: a case-control study

Abstract

The incidence and mortality rates of cancers of upper aero-digestive tract (UADT) remain significantly high in India. These cancers, which occur more commonly in later adulthood, are influenced by social and lifestyle behaviours carried out during childhood, adolescence and early adult life. Thus, life-course epidemiology attempts to assess varying health effects of various risk factors, according to timing, duration and frequency of exposure, which may give important clues to the causes of cancer. These cancers have serious impacts on quality of life (QoL) as they involve anatomical structures essential for mastication, speaking, cosmetic appearance and psychological wellbeing. Moreover, these cancers frequently present late and progress rapidly. Therefore, there is a need to develop a model for cost-effective screening and detection of individuals at high-risk of these cancers in the near future as well as early detection of cancer cases. Thereby, a study was designed with the following hypotheses:  Early age at initiation, frequency and duration of use of tobacco in its all forms, exposure to second hand tobacco smoke (SHS) at home and drinking alcohol has a linear dose-response relationship with the incidence of UADT cancers.  There is an association between tobacco, alcohol drinking, diet, oral hygiene and anthropometry measures with incidence of oral cancers.  Cancer site, staging, gender and age at diagnosis have an impact on QoL.  A risk-factor based screening model for UADT cancers has strong predictive ability to detect high-risk individuals. A bi-centre hospital-based frequency matched case-control study was conducted in Pune, Maharashtra State, India, from June 2014 and May 2015. Cases were histopathologically confirmed new cases of squamous cell carcinoma of UADT. Controls were patients diagnosed with a disease other than UADT or any other cancer, selected from the same hospital during the same period as the cases were recruited. Data were collected by medical-record abstraction, face-face interviews and by visual inspection of the oral cavity. A closed-ended questionnaire with a life-course perspective was used to collect patient‟s self-reported information on sociodemographics, lifestyle habits (chewing and smoking tobacco, drinking alcohol, SHS, oral hygiene) and QoL. The interviewer recorded anthropometry measures and number of missing teeth. Unconditional logistic regression was used to estimate the odds ratios (ORs) and their corresponding 95% confidence intervals (CIs). A reciever-operater characterstic curve was plotted against sensitivity and false-positive rate to produce a cut off point for the presence of UADT cancers. Examination of all QoL domains was done using oneway analysis of variance and the Bonferroni adjustments for post-hoc comparisons. Data were analysed by the Statistical Package for Social Sciences version 22. A total of 480 patients participated in the study. Chewing tobacco emerged as the strongest predictor for UADT cancers (OR=7.61; 95% CI 4.65-12.45) followed by smoking and drinking alcohol. Exposure to SHS during childhood (<16 years) rather than after ≥16 years increased the risk significantly (OR=4.05; 95% CI 2.06-7.95). There is a linear dose-response relationship between duration, frequency and early age at initiation for lifestyle risk factors (chewing and smoking tobacco; drinking alcohol) with incidence of UADT cancers in adulthood. Combined effects of tobacco and alcohol consumption habits elevated the risk (OR=12.05; 95% CI 4.61-31.49) in comparison to never users of these habits. Furthermore, the combination of these three lifestyle risk factors accounted for 86.82% of population attributable risk. Overall, the most affected QoL domains were anxiety and mood both among cases and controls. Oropharynx and hypopharynx cancer cases had the worst mean scores across all QoL domains. Stage IV cancer patients showed the worst QoL. Amongst UADT cancers in Pune, cancer of the oral cavity was the most common. Chewing tobacco showed higher odds (OR=8.51; 95% CI 4.90-14.77) for oral cancer risk as compared to UADT cancers. Poor oral hygiene emerged as significant predictor for oral cancer risk (OR=6.98; 95% CI 3.72-13.05). A screening model was derived for detection of individuals at high-risk for UADT cancers. This model has high sensitivity (93.5%), specificity (71.1%), false positive rate (28.8%), false negative rate (6.4%), positive predictive value (74.8%) and negative predictive value (96.6%). Our research recognises the framework of life-course influences of early exposure to behavioural risk-factors as independent and combined predictors of UADT cancers. The significantly compromised QoL in UADT cancer needs to be incorporated as an outcome measure in an individualized approach to therapeutic and palliative care planning of these cases to enable a better quality of survival. If validated in other studies, our proposed screening model can be applicable to many other high-risk UADT cancer populations with behavioural risk factors similar to our study population.