Howard White | 15 October 2020
HIV and Covid-19 are different. You need more intimate contact to catch HIV, but it was far more deadly once you did so prior to the widespread availability of antiretroviral drugs. But they are also the same. Both are pandemics for which we have no vaccine yet. Behaviour change to reduce the risk of infection is the only protection we have. It therefore seems sensible to ask, what can we learn from tackling HIV for fighting Covid-19?
Unsurprisingly, I am not the only person to think this. Several papers and blogs turn up from googling ‘lessons HIV AIDS Covid’. Let’s look at a few of these lessons.
The best of the batch of papers and blogs I looked at is Lisa Eaton and Seth Kalichman’s paper ‘Social and behavioral health responses to Covid-19: lessons learned from four decades of an HIV pandemic’. Their lessons are ones which are general enough to see the applications to Covid-19, but not so vague (e.g. ‘reduce risk of infection’) as to be useless to guide action. That is precisely where CEDIL sees middle-range theory as sitting: principles which are not programme-specific, but not at so general a level as to be useless to decision-makers.
One of Eaton and Kalichman’s six lessons is ‘Multi-level community interventions yield more robust and sustainable outcomes than single-level efforts to prevent HIV transmission’. This fits with the common finding in Campbell reviews that multicomponent programmes are more effective than single component programmes. For example, one of our early reviews on teenage pregnancy found that nothing worked other than a small effect from multicomponent interventions. I used to think this was just because different things work for different people. I even wrote a blog about that. That blog is also about middle-range theory though I didn’t call it that. Do read it – here’s the link again!
But ‘different things work for different people’ is not the only reason why multicomponent programmes are more effective. When I was thinking about the application of middle-range theory to water and sanitation interventions, I realised that a likely more important reason is related to the process of behaviour change. My discussion used the transtheoretical model of behaviour change: people go through several stages when changing behaviour. First, they need be convinced that there is a problem, next they need decide to do something, then to decide what to do, then do it, and finally to keep doing it. The key insight for me is that different interventions are necessary to support different stages of this process and failure at any stage scuppers the whole process.
This insight fits with Julia Rosenbaum’s blog arguing that ‘lessons learned from HIV prevention show that using fear tactics alone is ineffective’. It didn’t work for HIV and it won’t work for Covid-19. It needs to be supported with practical information on actions people can take to mitigate the risk. Fear may convince people there is a problem. But to decide to do something and what should be done, they need to be presented with possible paths of action and, usually, positive incentives to follow them – positive incentives at every step of the way.
Unfortunately – and this is the first of Eaton and Kalichman’ s lessons – ‘sustained, individual-level, behavioral change is challenging to achieve’. That is a mid-level principle. And a well-established one at that. It took decades to bring down smoking, and it was done through price incentives, prohibitions on where you could smoke and slowly changing social norms as much as public health messages. Attempts to promote a healthy lifestyle have not helped much in fighting the rising tide of obesity and the budgets of Big Sugar. Social distancing, wearing masks and regular handwashing will be difficult behaviour changes to sustain. Even if there has been an initial change in behaviour, many will soon go back to ‘business as usual’.
Furthermore, attempts to promote new behaviours are impeded by ‘medical mistrust and conspiracy beliefs undermine data-driven public health interventions’ – Eaton and Kalichman’s second lesson. The applications of these lessons to Covid-19 are clear. Conspiracy theories and rumours have fertile ground in social media. But they are not new to Covid-19, as seen by the anti-vaxxer movement. And they were prominent at the time of HIV/AIDS. As a sign that the times aren’t a changing, a widely believed conspiracy theory that HIV was developed in labs by the US government was started by the Russians. In 1992, I read an interview in a local magazine in Dar es Salaam with a visiting US professor who said that HIV was being spread by the condoms being distributed by USAID to control the African population. Even in the early 2000s, over twenty per cent of African Americans agreed that ‘AIDS was created by the government to control the black population’.
Such conspiracy theories undermine efforts to combat the virus. A key lesson, which really should be a central element in Development Intervention Design 101, though unfortunately there is no such course, is to get community leaders on board with messaging. In Indonesia, the government ensured that leading Islamic groups endorsed their policy that people should not travel home to visit their families during Eid this year.
A middle-range theory identifies barriers to successful implementation. A common general barrier is captured by the principle that ‘if you tell people to do something then the stuff they need to do it should be available’. My favourite example is one I came across for a study I did of basic education in Ghana. As one teacher said, ‘I would like to put posters on my walls, but I have no posters. In fact, as you can see, I have no walls’. The same point applies to the widespread promotion of hand washing as a preventive measure against Covid-19. Munro and Richards-Hewat’s blog on tackling Covid-19 in Indonesia notes that ‘Only 11% of Indonesian households have piped water inside their dwelling, and even then the quality is often low. In Papua province, in the eastern part of Indonesia, only around 35 per cent of the population has access to public hand-washing facilities with piped water and soap’.
The purpose of this blog has been to illustrate that we can use evidence from other programmes to inform programme design, and that middle-range theory provides a framework for doing so. We have not attempted to present the middle-range theory approach. To learn more about that, you can read the new CEDIL paper by Nancy Cartwright and colleagues, and the accompanying CEDIL Methods Brief.
Howard White is CEDIL’s research director and CEO of the Campbell Collaboration.
For updates on CEDIL’s work, sign up to join our mailing list.
Photo credit: Ousmane Traore, World Bank