This week’s Lancet brings an example of the very worst in medical publishing: trendy, misleading and meaningless.  A short article tries to address  a very important question: why do women in the Palestinian Territories continue to have a lot of children?

Conventional wisdom, much loved by Hans Rosling and Bill Gates, is that improved child survival always leads to smaller families. These smaller families, of course, are healthier and more prosperous and the children in them tend to have longer, richer lives. That has happened in a lot of places: Bangladesh is always the favourite example. Bangladesh’s economy has not boomed but the country has made dramatic strides in improving the health and status of women and the survival of children. The average woman in Bangladesh now has about 2.2 children, about the level needed to keep a population constant (although Bangladesh’s total population will continue to grow fast for some time because there are many girls and adolescents who have not yet started having children)

The Palestinian Territories are an example of a place where child health and survival has improved enormously but the total fertility rate (the number of children born to each woman) has declined much more slowly. A team of Danish researchers got a DANIDA grant and went to research this. What could be wrong with that?

  • This Danish team had their “field work … supported by travel grants from DANIDA”. Yet they managed to interview just eight Palestinian women. Yes, eight. What an excellent use of taxpayer funds. I do hope that all of the researchers went to all of the interviews. It would be terrible to think that any of them had become exhausted by the demanding schedule.
  • According to the researchers, women were “sampled on the basis of the number of their children (one to six) and imprisoned family members (yes or no) to represent the widest possible group of women”. It becomes clear later in the article that they mean, “imprisonment … by the Israeli military.” This is a bizarre use of the word “sample” which would get you sanctioned by any professional market research group.  And it certainly does not represent “the widest possible group” of women. To understand just how selective a sample this is, remember that about 4,700 prisoners were held in Israeli prisons on security grounds in 2011; there are, though, about  800,000 men aged between 15 and 64 in the Territories. Assuming all of the prisoners are male (they are not), that would mean that about 1 in 200 was held by the Israelis. Had they truly taken a sample, it is highly unlikely that they would have found even one mother of a prisoner
  • The researchers seem to want to show that, “the mothers feared death or imprisonment of their adolescent sons by the Israeli military. For most mothers, these fears prompted them to have more children to counter such losses.” Inconveniently four of their eight interviewees wouldn’t go along with this hypothesis but, of course, it didn’t stop the researchers saying that they had found four of their very odd “sample” who might have thought that
  • Of the other four, our researchers write, “four of eight mothers brought up infertility and were concerned about its social and political implications.” Was this a rational fear? Has there been a wave of unexplained infertility in the Palestinian Territories? We never find out

As a piece of research this is trite, shoddy and meaningless but it is worse because the researchers seem entirely unaware of the context. The researchers write, “globally, high fertility rates are often associated with high child mortality rates, which does not seem to be the case in the [Palestinian Territories]”. Or, indeed across most of French West Africa (where there have been wonderful declines in child mortality but women still routinely have six or seven children each) and most of the Middle East: Iraqi women have more children than Palestinian ones and are, presumably, untroubled by the fear of “death or imprisonment of their adolescent sons by the Israeli military” (although they have many other things to fear). Even in prosperous, well-run, democratic Jordan, women have almost as many children as do their counterparts in the West Bank. Jewish settlers in the West Bank have even more children (an average of about 5 compared to an average of about 3.5 for Palestinian women) and they are, presumably, entirely untroubled by the fear of arrest. So, this is a vital global issue that is entirely unaddressed by the work or writings of these researchers.