HIV in numbers

Summary

Podcast Summary: More or Less – “HIV in Numbers”

Host: Ruth Alexander (BBC World Service)
Originally Aired: March 11, 2013

---

Overview

This episode of “More or Less” examines the complexities around global HIV statistics: how they’re gathered, what they really mean, and whether claims of progress match the reality. Prompted by news of a baby being cured of HIV, the show digs beneath the headline figures and questions what we can genuinely know about the HIV epidemic and the success of prevention programmes.

---

Key Discussion Points & Insights

1. How Reliable Are Global HIV Numbers?

• Presenter: Ruth Alexander

• Opening with news that global HIV infection rates are reportedly falling, the episode immediately casts doubt, noting “there is a lot more uncertainty about these figures than the headlines suggest.” (00:20)

• Stats (2011, UNAIDS):

  • 34 million people were living with HIV globally
  • 2.5 million new cases that year
  • 20% reduction in new infections compared to a decade prior (00:44)

2. Problems with Estimating HIV Cases

• Wesley Stevenson and Professor Ron Brookmeyer (Biostatistics, Univ. of California) dissect why numbers fluctuated dramatically after 2007:

  • “Many of the statistics were based on women attending antenatal clinics… [these] are not representative necessarily of the whole country... there are differences between urban and rural areas…” – Ron Brookmeyer (01:34)
  • Key Issue: Antenatal data can overestimate prevalence due to its sampling bias.

• Dramatic revision in 2007:

  • UNAIDS reduced global HIV estimates by 6 million.
  • Revised methodology decreased the global figure by 15%, from 39 to 33 million.
  • New cases dropped around 40% to 2.5 million. (02:40)

  “In 2007, UNAIDS reduced their estimates of the numbers of people living with HIV worldwide by 6 million.” – Wesley Stevenson (01:09)

3. Modelling, Life Expectancy, and Statistical Uncertainty

• Numbers are now derived from a “combination of surveys, longitudinal studies, and computer models.” (03:08)

• “The completeness and accuracy of the data sources varies a lot by country.” – Ron Brookmeyer (03:08)

• Concerns about comparability:

  • “Do those changes in the figures reflect real underlying changes in the epidemic? Or do they simply reflect changes to new data sources... or the inputs into the various computer models?... It’s, in a sense, comparing apples to oranges.” – Ron Brookmeyer (03:39)

• NGOs & Correction: Organizations like WHO and UNAIDS attempt to ‘back-calculate’ earlier data to account for model improvements, but uncertainty persists. (04:03)

4. Does This Mean Success Stories Are Overstated?

• Ruth Alexander asks if the drop in new infections “paints an artificial picture of success.” (04:17)

• Professor Brookmeyer insists prevention still deserves some credit:

  “We’ve had a lot of prevention successes, including antiretrovirals for infected persons… for high-risk uninfected persons… increases in testing, counselling, condom usage, circumcision. All of these have been shown in rigorous epidemiological studies and clinical trials to be effective.” – Ron Brookmeyer (04:38)

• Bottom Line: Data are “a bit messy,” but real-world prevention advances have had a measurable effect.

5. Looking Ahead: Progress vs. Perfection

• “New methods of surveying and testing are being developed all the time. But for the time being, it’s clear that the numbers aren’t perfect.” – Wesley Stevenson (05:12)
• Continuing progress is acknowledged, but listeners are cautioned to treat HIV stats with a critical eye.

---

Notable Quotes & Memorable Moments

• “These are women who have been sexually active, they’re women in a certain age range and often there are differences between urban and rural areas... the prevalence of HIV infection appears to be higher when it’s based on the antenatal clinic data than when we look at the nationally representative surveys.” – Ron Brookmeyer (01:34)
• On data revisions: “It’s, in a sense, comparing apples to oranges.” – Ron Brookmeyer (03:39)
• “We’ve had a lot of prevention successes... All of these have been shown in rigorous epidemiological studies and clinical trials to be effective.” – Ron Brookmeyer (04:38)

---

Important Timestamps

• 00:13: Introduction and episode premise
• 00:44: UNAIDS’ headline HIV statistics for 2011
• 01:09 – 02:40: Detailed discussion of how estimates changed (antenatal survey bias, new models, large revisions in 2007)
• 03:08: Sources of uncertainty and how estimates are formed
• 03:39: Challenges in interpreting revised statistics
• 04:17 – 05:12: Are prevention programmes really working? Prevention evidence
• 05:12: Conclusion on ongoing uncertainty and progress

---

Tone and Style

The episode is analytical yet accessible, presenting complex statistical issues in everyday language with a healthy dose of skepticism and pragmatism. Both host and guests balance optimism about genuine progress with a clear-eyed view of data limitations.

---

Summary for Non-Listeners

This “More or Less” episode explores whether reported declines in HIV infections are due to genuine public health progress or shifting statistical methods. After dramatic downward revisions to HIV estimates in 2007, the show reveals how data were previously skewed by non-representative sampling and improved modelling. While numbers remain subject to significant uncertainty, prevention methods such as antiretrovirals, increased testing, and education have proven efficacy. The overarching message: treat headline HIV statistics with caution, but don’t doubt that substantial progress has been made.

---

End of summary. Skip to [00:13 – 05:12] in the episode for this segment; following time covers unrelated financial news.

Transcript

[BBC Announcer]

Thank you for downloading from the BBC. For details of our complete range of podcasts and our terms of use, go to bbcworldservice.com podcasts

[Ruth Alexander]

hello and welcome to More or Less on the BBC World Service. I'm Ruth Alexander. The news that a baby has been cured of HIV has got us thinking about how the data on HIV cases are gathered. The disease has been one of modern medicine's greatest challenges since it first appeared about 30 years ago. So it seemed encouraging to see that in recent years the number of new cases has been falling. But there's a lot more uncertainty about these figures than the headlines suggest. Wesley Stevenson's been looking at this.

[Wesley Stevenson]

Yes, according to the Joint United Nations Programme on HIV AIDS, or UNAIDS, as it's known, in 2011 there were 34 million people living with HIV and had 2.5 million new cases in that year. That's 20% fewer new infections than was the case 10 years earlier.

[Ruth Alexander]

So you'd think that tells a simple and strong story. Incident rates are falling, prevention programmes must be working.

[Wesley Stevenson]

But this might not be the entire story. New methods of estimating these figures are being introduced all the time in order to improve the accuracy of these figures. And sometimes this can have a dramatic effect. For example, in 2007, UNAIDS reduced their estimates of the numbers of people living with HIV worldwide by 6 million. Why? I asked Ron Bruckmeier, professor of Biostatistics at the University of California in the

[Ron Bruckmeier]

us, Many of the statistics were based on women attending antenatal clinics. Obviously, surveys among pregnant women in antenatal clinics are not representative necessarily of the whole country, because first there's no data from that that reflects men. These are women who have been sexually active, they're women in a certain age range and often there are differences between urban and rural areas who are attending these antenatal clinics. And in fact, what we have seen is when estimates based on antenatal clinic data compared to what comes out of more rigorously designed nationally representative probability based surveys, is that the prevalence of HIV infection appears to be higher when it's based on the antenatal clinic data than when we look at the nationally representative surveys.

[Wesley Stevenson]

UNAIDS also changed its assumptions about the life expectancy of people with HIV when it realised people were living longer than they thought. And the and this change, along with the adjustments made to the antenatal clinic data, changed the estimates dramatically. The numbers living with HIV worldwide fell 15% to 33 million, and the number of new cases dropped around 40% to 2.5 million.

[Ruth Alexander]

So where are we now?

[Wesley Stevenson]

Well, Professor Brookmeyer says there's still uncertainty in the figures.

[Ron Bruckmeier]

The reason has to do with the quality and reliability of the statistics that we have to work with. The statistics are based on a combination of surveys, longitudinal studies, and computer models. And the completeness and accuracy of the data sources varies a lot by country.

[Wesley Stevenson]

So along with the past changes to the way these figures are calculated, this adds up to a problem.

[Ron Bruckmeier]

Do those changes in the figures reflect real underlying changes in the epidemic? Or do they simply reflect changes to new data sources that are coming online or changes to the inputs into the various computer models? It's, in a sense, comparing apples to oranges.

[Wesley Stevenson]

Now, efforts are made by NGOs such as the World Health Organization and UNAIDS to work these changes backwards and recalculate the figures in the past to say what they would have been had they had the benefit of better models in previous years.

[Ruth Alexander]

But all the same, should we be suspicious that the falling numbers of new infections are painting an artificial picture of success and are actually down to not progress in prevention, but changes in the way new cases are counted?

[Wesley Stevenson]

Well, actually, Professor Brookmeyer doesn't think so. He says that although the data are a bit messy, there is other evidence that prevention programs are making a difference.

[Ron Bruckmeier]

We've had a lot of prevention successes, including antiretrovirals for infected persons that decrease their risk of transmitting to others, antiretrovirals for high risk uninfected persons which reduce their risk of acquiring infections, increases in testing and counseling, and condom usage, circumcision. All of these have been shown in rigorous epidemiological studies and clinical trials to be effective.

[Wesley Stevenson]

So there's still progress to be made, and new methods of surveying and testing are being developed all the time. But for the time being, it's clear that the numbers aren't perfect.

[Ruth Alexander]

Thanks, Wes. There's been celebration at the New York Stock Exchange. The Dow Jones Industrial Average recently closed at an all time high.

[Wesley Stevenson]

Well, that sounds like time for a party.

[Ruth Alexander]

Yes, maybe, because the index is seen as a pretty important indicator of the health of the US economy. It's made up of the shares of 30 of the country's biggest industrial players, like Exxon, Microsoft and Walmart. The record, which was broken three times in three days, was last set in 2007, before the economic crash. It was hailed as a sign that the US Economy was getting over its hangover and getting back to work. But. But it didn't take long for the party poopers to come along. Let's take up the story with Justin lahart from the Wall Street Journal.

[Justin Lahart]

Yeah, it certainly has hit a record high numerically. But some people are quick to point out that we have had inflation since the last record was hit in October 2007. So if you do an inflation adjustment, the dow is about 9% below where it was then. And the idea is that if you were to sell all those stocks, you wouldn't be able to buy as many hamburgers or cars or whatever you wanted to buy with your proceeds from your investment.

[Wesley Stevenson]

So if you'd invested in these shares, you would have less money now than you would have had in 2007.

[Ruth Alexander]

But this forgets dividend payments. Share prices aren't an investors only measure of profit and loss. These companies can also pay dividends. Dividends are a portion of a company's profits paid out to shareholders as a kind of reward for investing. So if you'd reinvested these dividends, used the money to buy more shares in the Dow Jones Index, how would you have done then?

[Justin Lahart]

Last I checked, you'd be up about 5% adjusting for inflation in general, you'd have to say that owning the Dow Jones industrial average since 2007, it would be better than just holding out of cash since then.

[Wesley Stevenson]

So it is party time.

[Ruth Alexander]

But Wes?

[Justin Lahart]

Yeah.

[Ruth Alexander]

How much did you have invested in the Dow?

[Wesley Stevenson]

Well. Yeah. Nothing. So back to the grindstone.

[Ruth Alexander]

Yes, please. Meanwhile, the fortunes of another giant of our times are being questioned. Apple has lost its crown as the world's most valuable publicly traded company. The oil company ExxonMobil has retaken the throne. Apple's stock market value, or market cap, has fallen to around $400 billion. Last year, when it had reached 500 billion. It it was claimed that Apple was so successful it was bigger than Poland, whose gross domestic product or GDP was about 470 billion. Now, the same comparison would make Apple about the size of South Africa. But we showed on this program at the time that this is actually a false comparison. The market value of a company is linked to how people expect it to perform in the future. Gdp, on the other hand, is a measure of the value of goods and services a company has produced minus imports in a single year. You could make a more sensible comparison if you took Apple's annual sales figure and subtracted the value of anything the company hadn't made itself but had bought in from other suppliers. This would reduce Apple's value considerably by around 75 or 80%. A year ago it meant Apple was about the size of Bangladesh or Vietnam. Now it's below Angola, about the same size as Slovakia. If you spot any numbers you think need a little investigating, please do email us at more or lessbc.co.uk and for more information about the programme and to listen to previous episodes, you can go to our BBC.co.uk more or less there

[BBC Announcer]

are dozens of different podcasts now available from the BBC, including news, documentaries, science, business, arts and sports. The details of them all go to bbcworldservice.com podcasts.