The Sunday entertainment section of Dainik Bhaskar, Ras Rang, screamed about the havoc that cancer was likely to wreak on the nation. The headline read, “Jang jari hai” (The war is on). Hence, presumably, four patently Caucasian looking archers were planted alongside the story; the relation between the archers and cancer is still something that one is trying to work out. But what left no doubt about the problematic nature of the piece was the statement attributed to the Ministry of Health that cancer was likely to increase 500 times by the year 2025. The next sentence said that 280 percent of this increase was likely to be from tobacco related cancers. Now even if you did not quite make the grade in basic arithmetic, please try to figure out what 280 percent of a 500 times increase means.
Since the alarm of the disease had been raised with archers in tow, it was necessary to read on. Quotes from two practitioners from Mumbai, an assortment of figures in a sidebar, and paragraphs that flowed with the cadence of a malignant hiccough were enough to make one want to find the author of the piece. On the top right, high above the archers, Ms. Labonita Ghosh was mentioned as the journalistic auteur. When you add cancer, quotes from Mumbai, Dainik Bhaskar, and a name like Labonita Ghosh in the Gwalior edition of the newspaper, basic forensics would lead one to DNA. A quick web search revealed that on 3 February, 2007, the original article, so one hazards, had appeared in DNA.
This translation, including the lead, had changed a 500 percent increase to a 500 times increase for the poor Hindi reader. The chances of malignancy in Hindi are greater than in English. Let us understand what this error means for the reader. If something that occurs once (1) undergoes a 500 percent increase, the occurrence becomes 5; however, if that same thing increases 500 times, the occurrence becomes 500. The error is effectively of 10,000 percent. But let us not be bored with figures. Let us only be inordinately alarmed because in an article that does not have any other fact in the main story, and the lead has such an error, it is not too much to be shocked to read such figures. But that was the least of problems.
Even in the English version the attribution of this figure was to the Ministry of Health. One would presume that such a figure was the result of some new study. The key point being that no mention was made about the source of these figures, how it was arrived at, and what were the baselines used for such estimates. On consulting the ministry’s website one accessed the report prepared by the National Commission on Macroeconomics and Health called “Burden of Disease in India” (September, 2005). In the chapter on cancer (pp.219-225), the key points about cancer in India are mentioned. Besides a map of India there may be found, among other points, the following: 8 lakh new cases of cancer per year, 24 lakh prevalent cases, relatively low rates of cancer occurrence, tobacco-related cancers amenable for primary prevention (48% in men and 20% in women). Further, the chapter projected that the incidence of cancer in 2015 will be 998,453, the prevalent cases will be 2,496,133, and deaths will be 666,563. The 2004 figure for incidence is 806,680, prevalent cases is 2,016,700, and deaths is 538,858. The increase from 2004 to 2015, hence, is 23.77% for incidence and prevalence, and 23.70% for deaths. Now, this may be some erroneous conclusion that has been drawn in a Ministry of Health background paper. Other accessible data did not throw the 500 per cent figure. A symposium article in the Annals of Oncology (17, Supplement 8, viii15-viii23, 2006), titled “The rising burden of cancer in the developing world (P. Kanavos) notes that cancer will increase by 158% in India by 2020. A socio-economic research report from the Centre for Policy Research available on the Planning Commission website, “Healthcare in India: Vision 2020” notes that “In India cancer is a leading cause of death with about 1.5 to 2 million cases at anytime to which 7 lac new cases are added every year with 3 lakh deaths…Studies by WHO show that by 2026 with the expected increase in life expectancy, cancer burden in India will increase to about 14 lac cases” (R. Srinivasan, http://planningcommission.nic.in/reports/sereport/ser/vision2025/health.pdf). A third report, that of the National Commission on Macroeconomics and Health, chaired by the present finance and health ministers, notes the following about cancer incidences in India: “In India, cancers account for about of (sic) 3.3% of the disease burden and about 9% of all deaths. These estimates will, however, surely change…It is estimated that the number of people living with cancers will rise by nearly one-quarter between 2001 and 2016. Nearly 10 lakh new cases of cancers will be diagnosed in 2015 compared to about 807,000 in 2004, and nearly 670,000 people are expected to die.” The purpose of providing these figures from different reports is to point out the problematic way in which one statistic is used without any pointer to its source. (One looks forward to the source of this information. The World Cancer Report could not be accessed for this piece.) Interestingly enough, Ms. Ghosh quotes the National Cancer Registry as saying that incidences of cancer would rise 2% each year, which by 2025 would mean an increase of about 55 per cent if the base figure is of 2004.
Reading the English version further one notes, “Among men, for instance, the three top killers – lung, head and neck, and oesophageal cancer – have everything to do with smoking and tobacco abuse.” One is not sure what head and neck cancers are for there is an appalling lack of precision in these terms. In fact, the Burden of Disease report mentions that among men the three highest incidences of cancer are those of Oral cavity, Lung, Pharynx, and the fourth is that of oesophagus. To put oral cavity and pharynx as part of head and neck can only be done by someone whose understanding of both anatomy and the English language is equally challenged, as it is now fashionably put. The information that Ms. Ghosh provides about cancer among women is also at variance with the Burden of Disease report. She also tells us that prostrate cancer is now alarming: “Prostate cancer is also up among Indian men, with one in every 100 affected (in the US it’s one in every 20 men). Since there were 531,277,078 men in 2001, Ms. Ghosh would have the reader believe that more than 53 lakh men are affected by prostatrate cancer. Since total prevalent cases were about 20 lakhs, we seem to be stuck with some error. Now one has to admit that the total population for men that has been used here is the Census data, while Ms. Ghosh might suggest a different parameter for men, as just being those males who are above say the age of 35. But that would call for some qualification, some elaboration.
The problem with such reporting where some figure is pulled out, quotes are sourced and a story is strung together to meet the deadline is that the possible impact on the readers could be devastating. Medical reporting, like all specialized domain reporting, begs for a minimum knowledge of health issues, medicine, research methodologies, and the ability to translate all these for the lay reader. Medical reporters have an added burden. They need to be mindful of the fact that their reports may be used by readers to solve their own medical problems, to confront doctors, to change lifestyles. Some random picking of datum is heinous. Stringing a story around it is misleading and dangerous. But a close perusal of medical stories, some based on a scant understanding of early stage clinical trials or even a peer-reviewed journal article, will more than amply prove the erroneous way in which medicine is represented. Ms. Ghosh’s contribution is merely illustrative, not exemplary.
The minimum that the readers can hope is that medical reporting is done in a slightly more informed manner than political reporting, or even business reporting, which is now more of puff pieces for corporations and corporate officers than any diligent reporting and analysis of facts. Or may be statutory warnings need to be attached to medical reporting: Any and every information contained in this piece may be false, outdated, inaccurate, or may have been used out of context. Any alarm, anxiety, or relief experienced on reading the piece may be symptomatic of a closed and lazy mind.
