Assumptions Matter: Structural Inequality and The Relationship Between Behaviour and Communicable Disease Outbreaks
Communicable diseases, by definition, are spread by people interacting with each other. Respiratory infections have a particular risk of causing major outbreaks like the current COVID-19 pandemic. If they can transmit efficiently from person to person, then they inevitably spread through normal life — going to work, speaking to other people, being close to loved ones, and sometimes even just breathing.
Since people spread infections to one another, there is a troubling tendency to blame the worse affected communities — who are often historically marginalised groups — for causing or continuing outbreaks through their behaviour. Western discourse during the 2014–2016 Ebola outbreak, for example, featured stigmatization of individuals from Africa — often from unaffected regions — and their perceived cultural practices. Sexual stigma interacting with anti-LGBT+ prejudice, racism, and sexism has been rife and interferes with progress against the ongoing HIV-AIDS pandemic.
While the COVID-19 pandemic has often been referred to as a ‘great equalizer’, it has rapidly become apparent that the infection does not strike equally across the population. Ethnic and socioeconomic inequalities in infection and mortality rates are stark, with people of colour and people living in deprived circumstances disproportionately affected. Disconcertingly, the pandemic has already produced an explosion of xenophobic and racial discrimination sufficient to warrant its own Wikipedia page.
While targeting infection-related behaviour is essential to outbreak control, individually focused explanations and interventions can easily contribute to stigma and discrimination if structural factors influencing health and behaviour are not considered carefully. Behavioural and biological risk is underpinned by the social, political and physical world in which individuals live. For example, lack of access to safe, affordable, uncrowded housing makes shielding vulnerable people or self-isolation extremely difficult. Not everyone is equally able to work from home, and public-facing or indoor work without adequate protection can dramatically increase exposure to the virus. A history of discrimination in health and social systems can impact trust in services and guidance. All too often, quality of healthcare and access to testing can become a postcode lottery.
There are many such pathways, and without adequate consideration into how they impact the distribution and spread of an infection it can appear that heavily affected groups are deliberately or carelessly driving an outbreak through their behaviour. This problem can be reinforced when policies targeting individual behaviour are applied to whole populations without addressing how to implement these suggestions in challenging circumstances.
As scientists, we may feel like we would never make such an error of oversimplification and conflating correlation and causation. Unfortunately, it is easy to perpetuate by presenting social characteristics, such an ethnicity or socioeconomic status, as ‘risk factors’ without appropriately contextualising the structural influences on risk of infection and/or poor outcomes. Big Data and predictive modelling may be particularly at risk of entrenching these inequalities without deliberate action to represent diverse groups in datasets, and have a strong theoretical understanding of structural determinants of health and disease built into research design an interpretation.
COVID-19 is not over, nor is it the last major communicable disease threat that we will face. If we combine the huge scientific and computational advancement of recent decades with a commitment to health equity, we have the opportunity to make great progress in combating major communicable disease outbreaks for the long-term.
