Crunching HIV/AIDS data during a time of fear (and loathing)

This year marks four decades since scientists developed an antibody test for HIV, the breakthrough coming a year after it had been confirmed as the virus that causes AIDS.

It was an important milestone – and not just for clinicians and policymakers. Professionals across a range of industries had been grappling with the potential impact of a devastating and highly infectious disease.

For Professor Steven Haberman, appointed a decade earlier to teach on the pioneering new BSc in Actuarial Science at what is now Bayes Business School, HIV/AIDS was a uniquely sombre and difficult challenge as he and other experts assessed the potential impact on the life insurance industry.

In early 1987 he joined a group of actuaries tasked by the UK actuarial profession with providing a range of potential scenarios to the sector. Through the 1980s insurance companies – several of them mutuals – faced the prospect of many male policyholders dying relatively young.

Fear of the 'gay plague'

It was a time when unfettered tabloids ran front page stories referring to 'the gay plague'. One story was headlined: "I'd shoot my son if he had AIDS," says vicar. Lest readers were unsure what this might look like, the article was accompanied by a picture of a man pointing a gun at a younger man.

On a more sober level, the very survival of some insurance companies was thought to be at stake. That is clear from the title of the report Steven's group prepared for the Institute of Actuaries in December 1988: The first report on the potential impact of HIV/AIDS on life insurance premiums and insurance company solvency.

Steven recalls: “When we started our work in the spring of 1987, it was a really serious crisis. Everyone was trying to respond quite quickly. Nothing had happened like this before and there was not really any data available for insured populations.

“Our early ideas were shared with actuaries and insurance companies in a series of bulletins and a discussion. Feedback plus further analysis led to our first report at the end of 1988. The lack of data meant the initial work around death rates for our first report in 1988 was very cautious. We over-estimated the number of deaths. That affected premium levels and how much money we were recommending insurance companies should keep in reserve.

“But the data was being monitored all the time and we were working and sharing data with epidemiologists at Imperial College and elsewhere. By the following year, we felt that we had been too cautious in the first report and we published a sensitivity analysis in 1990 exploring ranges of assumptions which were widely regarded as more realistic.”

Different times

The group’s work was set against a backdrop of controversy around insurance companies’ willingness to provide life cover to gay or bisexual men – and whether premiums should be higher for that group.

“Insurance companies’ view of fairness is that you've got to be fair both to those at low risk and those at high risk. For example, drivers who’ve had 8-9 motor accidents are likely to face higher premiums for vehicle insurance than those with a clean driving record.”

That, however, is not quite the same as asking people about their sex lives and basing premiums and policy decisions on the answers. Steven and his co-authors were also being lobbied by early HIV campaign organisations such as the Terrence Higgins Trust. (The image above shows the first public meeting in London, in 1983, organised by groups that later became the Terrence Higgins Trust.  The trust kindly provided the above picture).

“We've got the equality legislation now but it was very different times then. People were very concerned about these questions, and there was a sombre mood in terms of thinking that the situation was bad enough that it could actually be absolutely devastating for certain sections of the population.”

By 1990 however, the group was able to refer in their report to the potential impact of the first effective treatment, AZT.

AIDS-related deaths in the UK, the paper suggested, could peak at more than 10,000 a year in the late 1990s. That was one of only several scenarios however, with the authors insisting they were unable to say which projection was likely to be the most accurate. The actual peak was around 1800, with a sharp decline in the new millennium.

New treatments not only saved the lives of those living with the virus but reduced viral load levels – and therefore transmission by people receiving treatment – to near zero.

Steven recalls: “This was an advisory report, and every insurance company (advised by its team of actuaries) is going to respond in the way that's appropriate for them in terms of the characteristics of their portfolio of policyholders and who they're selling insurance policies to.”

While some may think that Steven’s work – and HIV/AIDS is now of only historical interest in western nations – new HIV diagnoses in the UK have been rising since 2021.