[Skip to Navigation]
Sign In
July 7, 2021

Lay Epidemiology and Vaccine Acceptance

Author Affiliations
  • 1Department of Medicine, Massachusetts General Hospital, Boston
JAMA. 2021;326(4):301-302. doi:10.1001/jama.2021.11130

As vaccination rates against SARS-CoV-2 slowed across the US, increasing vaccine uptake became a national priority. Concerns about lack of confidence in the vaccine dominated lay and professional headlines and hundreds of programs were created to increase vaccine confidence, particularly among vulnerable populations who were most affected by the COVID-19 pandemic. In general, these programs focused on providing more information about the vaccine to communities thought to be at increased risk for vaccine refusal. This approach assumed that the decision to avoid or delay vaccination was based on inadequate understanding or information, perhaps overlaid by distrust of those involved in creating or delivering the vaccine. In that model, more information delivered by “trusted messengers,” including community leaders and local physicians, is the solution.

Reality, however, is more complicated. Behavioral science has long demonstrated that knowledge of the risks and benefits of a given intervention has a surprisingly limited relationship with health behaviors. For instance, across dozens of studies, perception of the risk or severity of the disease has only a small to moderate correlation with the decision to undergo cancer screening, attempt to stop smoking, or get a vaccination.1-3 A similar pattern is seen with perceptions of the benefit of an intervention and with studies of interventions to increase understanding of risk and benefits. These factors matter, but they generally explain much less than half of the variance in why someone does or does not follow a recommendation.

Why does this gap exist? Sometimes other factors determine whether an individual can or will want to follow the recommendation. For example, if a patient cannot afford out-of-pocket costs, information about risks and benefits is irrelevant. If the social norms in a person’s group are not aligned with the recommendation, they may struggle to override that pressure even when they understand the risks and benefits. If people do not believe the source of the information, they are unlikely to follow it. But sometimes it occurs because there is a difference between the average, population-based information that is provided and the assessment of the likelihood that an individual would experience a positive or negative outcome from the intervention.

How could that likelihood be judged? Across cultures, people try to make sense of the world around them, including how likely it is that a negative outcome will occur for them and what will increase or reduce that risk. This concept has been referred to as lay epidemiology.4 Lay epidemiology is how inferences are drawn from patterns of disease in small groups like friends and family, larger groups from social media or other sources, and even entire populations from public information or news stories.

An example is the concern about the possible link between vaccines and autism. Many parents heard news stories linking autism with vaccination, believed there could be an increased risk from vaccination, and were hesitant to give their children vaccines even when given vaccine information by a trusted pediatrician. The same phenomenon occurs when parents draw inferences about what exposure led to a newborn having a birth defect, women attribute their breast cancer to a breast injury, or people believe that they will not get cancer from smoking because of all the smokers they know who do not have cancer.5

What happens when people must make a new decision like whether to get the COVID-19 vaccine? In the same way, they extrapolate from what they know and have heard, past and present. For disadvantaged groups in the US especially, the message is clear. Outcomes for these groups are worse than others. The average life expectancy in the US is 78 years, but life expectancy for men in the lowest strata of socioeconomic status is less than 73 years and is 71 years or less for men in many areas of the rural South.6 Individuals who are poor or from a racial or ethnic minority group are more likely to develop a disease, less likely to access needed treatment, and more likely to experience morbidity and mortality. For some groups, the government has conducted medical experiments without their consent in the past. The evidence is consistent across media, public data, and lived experience. Furthermore, these facts have been known for a long time without much done to address them. The greater burden of COVID-19 among disadvantaged communities has further reinforced this lay epidemiology over the last year, a critical period for influencing current decisions about vaccination.

From the lay epidemiologic perspective, it makes sense that groups with clear evidence of experiencing worse outcomes from most aspects of the US health care system should be skeptical of the information about the average risks and benefits of vaccination. In fact, lack of confidence is a rational response to these experiences. What can be done?

First, tailored messaging and data are needed. People adjust their perceptions of their expected health outcomes based on data that they find relevant to themselves and often ignore information that does not seem relevant. Information that is tailored to individual patient characteristics has been demonstrated to improve the use of a wide range of preventive services.7 For COVID-19 vaccination, information about risk of infection, effectiveness of vaccination, and the chance of vaccine-related adverse effects could be shown by age, sex, race and ethnicity, geography, job type, and even socioeconomic background. Such information would be most effective when shared through images and stories, rather than numbers alone. By providing information about lay groups, health professions can use lay epidemiology in their favor.

Second, it is essential to engage local leadership who understand local beliefs, know the local data, and can address the lay epidemiology in their communities. The marked geographic variation in vaccine use in the US highlights the reality that information provided by government sources is unlikely to be effective in driving lay perceptions of risk and benefit in many communities across the South and West in particular. While the politicization of vaccination may contribute to low uptake in some areas, there are also positive stories of collaborative efforts to address local concerns leading to high levels of vaccination. Perhaps the most striking example is the high levels of vaccination in many tribal communities (estimated at 88% vaccine uptake in Navajo Nation as of May 2021) where local leaders and epidemiologists collaborated to address lay concerns and provide salient information to community members.8

In addition, medical professionals need to move from a focus on distrust to a focus on being trustworthy. Thus far, much of the discussion has centered on why certain groups are not taking the vaccine. This places the blame, directly or indirectly, on the people affected, often disadvantaged groups. From one perspective, vaccine hesitancy is a symptom of a system that has created (and largely ignored) wide differences in health in the US. Addressing such inequities is an important step toward ensuring that all patients believe that the health care system will bring them just as much benefit at just as little risk as it does to the most advantaged in US society.

While lay epidemiology is far from the only factor driving vaccine hesitancy, it is too often overlooked. In many communities, vaccination rates are increasing as people see their friends, colleagues, and neighbors get vaccinated without adverse events. But in some areas, vaccine uptake is lagging and efforts to address lay epidemiology may be an important factor to ensure that the country is able to reach vaccination goals over the months ahead.

Back to top
Article Information

Corresponding Author: Katrina Armstrong, MD, MSCE, Department of Medicine, Massachusetts General Hospital, 55 Fruit St, Gray 730, Boston, MA 02114 (KARMSTRONG6@mgh.harvard.edu).

Published Online: July 7, 2021. doi:10.1001/jama.2021.11130

Conflict of Interest Disclosures: None reported.

Atkinson  TM, Salz  T, Touza  KK, Li  Y, Hay  JL.  Does colorectal cancer risk perception predict screening behavior? a systematic review and meta-analysis.   J Behav Med. 2015;38(6):837-850. doi:10.1007/s10865-015-9668-8PubMedGoogle ScholarCrossref
Brewer  NT, Chapman  GB, Gibbons  FX, Gerrard  M, McCaul  KD, Weinstein  ND.  Meta-analysis of the relationship between risk perception and health behavior: the example of vaccination.   Health Psychol. 2007;26(2):136-145. doi:10.1037/0278-6133.26.2.136PubMedGoogle ScholarCrossref
Costello  MJ, Logel  C, Fong  GT, Zanna  MP, McDonald  PW.  Perceived risk and quitting behaviors: results from the ITC 4-country survey.   Am J Health Behav. 2012;36(5):681-692. doi:10.5993/AJHB.36.5.10PubMedGoogle ScholarCrossref
Frankel  S, Davison  C, Smith  GD.  Lay epidemiology and the rationality of responses to health education.   Br J Gen Pract. 1991;41(351):428-430.PubMedGoogle Scholar
Lawlor  DA, Frankel  S, Shaw  M, Ebrahim  S, Smith  GD.  Smoking and ill health: does lay epidemiology explain the failure of smoking cessation programs among deprived populations?   Am J Public Health. 2003;93(2):266-270. doi:10.2105/AJPH.93.2.266PubMedGoogle ScholarCrossref
Chetty  R, Stepner  M, Abraham  S,  et al.  The association between income and life expectancy in the United States, 2001-2014.   JAMA. 2016;315(16):1750-1766. doi:10.1001/jama.2016.4226PubMedGoogle ScholarCrossref
Hawkins  RP, Kreuter  M, Resnicow  K, Fishbein  M, Dijkstra  A.  Understanding tailoring in communicating about health.   Health Educ Res. 2008;23(3):454-466. doi:10.1093/her/cyn004PubMedGoogle ScholarCrossref
Navajo Nation reports 88% vaccination rate, close to herd (community) immunity. Navajo-Hopi Observer. Published May 4, 2021. Accessed June 29, 2021. https://www.nhonews.com/news/2021/may/04/navajo-nation-reports-88-vaccination-rate-close-he/
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    1 Comment for this article
    Risk vs Outrage
    Wayne Maksylewich, MSc, MEng. | Retired Engineer, Industrial Hygienist
    The authors ask 'Why does this gap exist?'  Dr. Peter Sandman identifies 2 components of risk - the risk professionals calculate, and the risk perceived by people who are not professionally capable in the risk calculation (1). His equation is Risk = Hazard + Outrage.

    He identified a number of factors that make any issue outrageous. Sandman states: "... trust, control, voluntariness, dread, and familiarity (now widely called “the outrage factors”) are as important as mortality or morbidity in what we mean by risk." 

    Throughout my career, success in implementing successful risk perception and abatement programs
    has required addressing both hazard and outrage. I have been professionally active in both SARS and the COVID pandemic and have seen both the medical profession and politicians fail at risk communication. Above all they fail to win trust.


    1. http://www.psandman.com/index-OM.htm