Cancer Death Comes Early in India

Verbal Autopsies

Nick Mulcahy

March 28, 2012

March 28, 2012 — An estimated 71% of all cancer deaths in India occur in people 30 to 69 years of age, according to a study published online March 28 in the Lancet.

This is a "landmark study" because it is a "nationally representative sample," according to an accompanying comment.

The study provides adequate coverage of often neglected rural areas where three quarters of Indians live, write Rengaswamy Sankaranarayanan, MD, from the International Agency for Research on Cancer in Lyon, France, and Rajaraman Swaminathan, PhD, from the Cancer Institute in Chennai, India, in their comment.

It "represents notable progress in obtaining direct estimates of the cancer burden, particularly in regions that are not covered by cancer registration," they write.

"Verbal autopsy" — and not systematic medical certification — was used to determine the cause of death in the study. Such autopsies, which rely on relatives' accounts of illness, are used in an Indian data-collection system that covers 1.5 million households. The data collection is contributing to a larger ongoing general mortality study, known as the Million Death Study. The study enables a population projection of mortality from the autopsy data.

The fact that so many cancer deaths occur in India before old age is an opportunity, Drs. Sankaranarayanan and Swaminathan suggest. There are "substantial social and economic gains that would be associated with a successful cancer prevention program," they write.

Among men 30 to 69 years of age, about half of all of cancer deaths were from oral and pharyngeal (23%), stomach (13%), and lung (11%) cancers. Among women in the same age group, about half of all cancer deaths were from cervix (17%), stomach (14%), breast (10%), and oral and pharyngeal (10%) cancers.

However, Drs. Sankaranarayanan and Swaminathan cast a somewhat critical eye on the specifics of the cancer mortalities. "For cancer deaths, assigning causes of death to internal cancers such as stomach, liver, bowel, and kidney can be more challenging than for sites such as head and neck, breast, and cervix," they point out.

Nevertheless, the study is a "valuable exercise in improving cancer mortality data in India," they write.

Interventions such as tobacco control, vaccination against human papillomavirus and hepatitis B, cervical cancer screening and early detection, and treatment of oral and breast cancers would have a "substantial effect" on cancer death prevention, they add.

Overall, cancer mortality rates in India are about 40% lower in men and 30% lower in women than in the United States or United Kingdom, they write. However, this will change because cancer death rates are expected to rise, "particularly with increases in age-specific exposure to tobacco smoking," they note.

Data Include Religion, State, and Education

There have been previous mortality estimates for specific cancers in India, but they have heavily relied on data from urban population-based cancer registries, according to the study authors, led by Prabhat Jha, MD, DPhil, from the University of Toronto in Ontario, Canada, who collaborated with colleagues from India and elsewhere.

The Million Death Study is one of the few large nationally representative studies of the causes of death in any low- or middle-income country, say the study authors. The study allows for a better estimation of the causes of death, including cancer, because it is not limited by locale or education level, Drs. Sankaranarayanan and Swaminathan note.

The authors found that 7137 of the 122,429 study deaths were due to cancer, which corresponds to a projected estimate of 556,400 cancer deaths across the whole of India for 2010. About 71% (395,000) of these deaths occurred in people 30 to 69 years of age (200,100 men and 195,300 women), the authors report. Cancer deaths accounted for 6% of deaths across all ages.

Surprisingly, age-standardized cancer mortality rates per 100,000 were similar for men in rural (95.6; 99% confidence interval [CI], 89.6 to 101·7) and urban (102.4; 99% CI, 92.7 to 112.1) areas, and for women in rural (96.6; 99% CI, 90.7 to 102.6) and urban (91.2; 99% CI, 81.9 to 100.5) areas. However, these rates varied greatly between Indian states, with higher rates in poor states.

Level of education influenced mortality, the authors found. The cancer death rates were 2 times higher in the least educated (illiterate) men (106.6; 99% CI, 97.4 to 115.7) than in the most educated men (45.7; 99% CI, 37.8 to 53.6), and in the least educated (illiterate) women (106.7; 99% CI, 99.9 to 113.6) than in themost educated women (43.4; 99% CI, 30.7–56.1).

Cervical cancer was far less common in Muslim women (24 deaths; age-standardized mortality ratio, 0.68; 99% CI, 0.64 to 0.71) than in Hindu women (340 deaths; 1.06 ; 99% CI, 1.05 to 1.08). Circumcision among Muslim men is a "likely explanation," the authors write, saying that it reduces the sexual transmission of human papillomavirus.

The study was funded by the Bill & Melinda Gates Foundation and the US National Institutes of Health.

Lancet. Published online March 28, 2012. Abstract, Comment


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.