Posted by: Aadisht on: October 31, 2007
P. Sainath being innumerate is actually the most charitable explanation for this editorial. A less kind explanation is that his bias is making him too lazy to do his research properly, and a very unkind explanation is that he’s actively using scare tactics to push an agenda.
I refer specifically to this section:
Let’s revert to the latest maternal mortality figures released by the WHO and others. Some 536,000 women died in childbirth in 2005. Of these, every fifth one of them, at least, was an Indian. That is, 117,000 of them. A total that could only be matched by Nigeria, Afghanistan and Congo together.
Does Sainath not understand the concept of per-capita mortality rates (which makes him innumerate at best and stupid at worst), or is he intentionally not bringing them up (which makes him dishonest)?
The report Sainath is referring to is here. Scroll to Page 23 of Section 1 (which is Page 29 of the PDF file). This is the table which has the estimates of maternity deaths. Page 24 has the India figure: as Sainath says, it’s 117,000.
What about the three other countries? The figures are:
which totals to 117,000 as well.
What Sainath omits, of course, is that India’s population is one billion people, much more than that of Nigeria, Afghanistan, and the DRC taken together. What is very curious is that the report puts the lifetime risk of dying in childbirth and the deaths per hundred thousand childbirths in the same table, and Sainath doesn’t use these measures, which are far more useful and worthwhile. Incidentally, here they are:
Lifetime risk of maternal death:
In other words, you are four times less likely to die giving birth in India than Nigeria.
What about the number of deaths per 100,000 childbirths (referred to as the Maternal Mortality Rate, or MMR)? In the same table, we get the figures:
Now, if 1 out of 70 mothers is going to die giving birth, that is still an obscene figure. And there is a long way to go. Similarly, if approximately one of every two hundred pregnancies is going to end in the death of the mother, that’s still nothing to be proud of.
For comparison, here are the MMR figures for the Asian tigers, which started independence poorer than India:
So there’s a long way to go. But to twist statistics to make India seem worse off than countries with actual mortality risks four to eight times worse smacks of scare tactics. Moreover, an unbiased person would look at the table, and see links with levels of urbanisation, the rule of law, and how soon a country started economic reform. Sainath looks at it and goes off on a tangent to abuse the media for talking about the Sensex instead of this (and doesn’t that argument sound very similar to the one which abuses the media for talking about the Gujrat riots instead of the ethnic cleansing of Kashmiri Pandits?).
Sainath also writes:
In fact, it would be good to devise a health index spanning the reform years. One that looks at how both rich and poor have done health-wise. How many years of life, for instance, are taken away from you by ill-health if you are one of India’s less well off citizens?
Excellent idea. Let’s look at the WHO’s 2000 report on Maternal Mortality. Scroll to page 26. The MMR in 1998 was 540. In other words, the maternal mortality rate has seen a 20% drop in 7 years.
So let’s close with Sainath’s parting shot:
Maybe we need a media relevance index. An MRI scan of mass-produced mediocrity.
Like the mediocrity of his research and grasp of statistics? Pot, kettle, black.