Name:
Location: Vancouver, B.C., Canada

I'm a PhD student in econ at UBC. For fun, I write this blog.

Tuesday, August 08, 2006

The Globe and Mail's AIDS trainwreck.

Yes, it's time for more bitching about the Globe and Mail... this time its goony A4 feature story on AIDS in Canada, offered up in advance of the International AIDS Conference to be held in Toronto next weekend. Here's the gist of the piece:

When we think of HIV and AIDS now, we think of African countries, where HIV infection rates are skyrocketing each year. But the epidemic at home is more real than ever. Why is Canada, a developed nation that prides itself on its health-care system and one that helps fund international research for the disease, continuing to see people contract the virus that causes AIDS?

And here is the graphic that accompanies the piece:



I'll start with the obvious bloopers. Close inspection reveals that three of the HIV--infection source categories in the pie charts at the top are "male homosexual contact" (on the decline as a source of HIV), "heterosexual contact/other" (on the rise) and -- as a wholly separate category -- "sexual contact with a person at risk" (also apparently on the rise). Maybe I'm just too innocent, but if sexual contact isn't homosexual or heterosexual (or "other"!), just what kind of sexual contact is it?

Two of the next graphs show infection rates by province and ethnicity without any information about the underlying population distribution by province and ethnicity. But the real kicker comes in the final bar-graphs, in which we learn that Canadian doctors are diagnosing about 300 new AIDS victims per year. Leaving aside the G&M's breathless report, can the correct number really be that low, or is it a per-million-adult-population figure or something? According to the CIA World Factbook, there were 1500 Canadian deaths from AIDS in 2003 vs. what looks about 100 on the G&M's graphic. And even the 58,000 Canadians-living-with-HIV number reported by both the G&M and the CIA factbook indicates a national incidence of 0.3%, which -- although certainly the most startling number in the report -- actually seems pretty benign when you consider Dorothy Aken'ova's op-ed reporting that, worldwide "7,000 girls and women are infected with HIV every day." Every day.

But it gets better. Midway through the piece we read: Unlike many developing countries, Canada doesn't have tens of thousands of people becoming infected with HIV annually, but it is still seeing a climb in infection rates in all major groups...Canada needs to implement some basic strategies before it can expect progress in its efforts against HIV and AIDS, Dr. Montaner said.

All major groups? Before we can expect progress? Unless they're severely mislabelled, or unless HIV is showing a wildly different trend from full-blown AIDS, the bar graphs show that both total new diagnoses and total deaths have been declining monotonically in Canada for over a decade! It may be the case that women are making up a historically high percentage among total new HIV infections as the article goes on to discuss, but the overall AIDS pandemic is, according to the graphics, very much on the decline in Canada and has been since about 1992. Is this totally due to anti-retroviral therapy? What about people who don't go in for HIV tests? And what are the actual annual numbers on HIV infections? Who edited this article?

To be sure, I'm not saying AIDS isn't a problem in Canada (one in three hundred residents isn't something to sneeze at). I'm not even disagreeing with the various experts quoted in the piece that more federal money is needed to fight it. I don't honestly know. But this piece of reporting is just goony.


Update: Oddthoughts has some further thoughts. On further reflection, what seems incredible is that of a population of 58,000 HIV positive residents (a quarter of them undiganosed according to estimates reported at CBC), AIDS is causing so few actual deaths. Based on some more Googling, the numbers for exactly how many are all over the map. But GlobalHealthFacts says less than 1000. So maybe the trend is due almost entirely to anti-retroviral therapy and not to falling HIV infection rates. It would be nice if the G&M would tell us.

0 Comments:

Post a Comment

<< Home

Web Site Hit Counters
Web Site Hit Counters