JNJ-42153605 Data sources for cancer rates were national can
Data sources for cancer rates were national cancer registry reports (http://www.ncdirindia.org) [, , ]; estimates of cancer incidence in 1991  and estimates of cancer incidence in 2012 . Data sources for life-expectancy, tobacco prevalence, per-capita consumption of alcohol, and prevalence in overweight and obesity were from world health organization, global health observatory data (http://www.who.int/gho/en)  and for Gross Domestic Product (GDP) were from world bank (https://data.worldbank.org/country/india) .
The various measures used are age-standardised (world population) incidence rate, to compare the rates with various countries (ASR) and annual percent change (APC), to assess the transition in rates. Data on a long time period are available only from registries such as Bangalore, Barshi, Bhopal, Chennai, Delhi and Mumbai. We fitted linear regression to the natural logarithm of the estimated rates APC on incidence rates of these registries. JoinPoint linear regression was used to determine trends in cancer incidence (APC), a statistical algorithm detects Joinpoints, or points in time where the slope of the regression line significantly changes. Thus, the model described trends during different time segments. At each segment, trends in rates were measured using the estimated APC, which assumes that rates changes by a constant percentage each year. The JoinPoint regression software was used for the trend analysis .
Discussion A decline in tobacco prevalence was reported in the past 20–25 years in both genders in India. Of the several causes investigated for cancer, the use of smoking and smokeless tobacco has shown strong and consistent associations with several cancer sites such as lung, oral cavity, other pharynx (oropharynx & hypopharynx), larynx, oesophagus and urinary JNJ-42153605 in the body [13,14]. However, no decline in the major tobacco associated cancer sites was observed in the present analysis except a slight decline in pharyngeal (excludes nasopharynx) and oesophageal cancers. It is alarming to note the increase in the incidence of lung cancer in both genders. It has been reported that cancers such as cervix-uteri, Hodgkin’s disease, nasopharynx, stomach and liver are mainly infection related [15,16]. In India, increase in the economic status and thereby reduction in the poverty clearly indicated reduction in the infection-related cancers such as cervix-uteri. However, life-style modification happened in many parts of the country reflect increased incidence of several related cancers such as breast, colo-rectum, prostate, corpus-uteri etc. What is probably more worrisome is, India witnessing increased incidence of key risk factors such as obesity, physical inactivity, alcohol etc. that contribute to the occurrence of many cancers. Life-style modification such as delayed child birth and changing breastfeeding pattern are likely to result in increased breast cancer incidence. Excessive use of alcohol is linked to other pharynx, breast, oesophagus (squamous cell carcinomas), liver, colo-rectum [17,18]. Further, changing dietary patterns are linked to the risk in overweight and obesity. It has been reported that increasing overweight/obesity rates and physical inactivity led to an increase in the risk of many cancers such as corpus-uteri, colo-rectum, breast, and oesophagus (adeno carcinomas) [, , ]. Thus increased alcohol consumption, increased prevalence in overweight/obesity and physical inactivity and with the tobacco epidemic, the burden of related cancers in India will be very high. A substantial portion of cancer cases could be prevented by broadly applying effective prevention measures. For instance, cancers such as breast and cervix-uteri can be cured if detected early and treated adequately. In the present analysis, an up-down variations was observed in some of the transition graphs (Figs. 6c–d, 7 a–f), which might be due to the variation in the completeness in case-ascertainment existing in cancer registries. Hence, it is very important that cancer incidence data needs to be assessed more accurately.