NEJM | Social media and other digital platforms provide the opportunity to collect data on vaccine hesitancy in nearly real time70,71; they also allow new methods of analysis72 and the opportunity to investigate the effect of vaccine sentiment on actual vaccine uptake and vaccine-preventable diseases. Facebook collaborated with Carnegie Mellon University and the University of Maryland to collect survey data on a wide variety of behaviors related to the Covid-19 pandemic.73 Starting in January 2021, Facebook users who agreed to participate in the survey were asked about their attitudes toward Covid-19 vaccines and reasons underlying vaccine hesitancy.
Although data collected on social media platforms, such as Facebook, Twitter, and YouTube, may not be representative, since the users of the platform are not a random sample of the population, the data have aligned well with other, less frequently compiled survey data that are available for select topics and populations. In addition, sometimes data collected through online platforms are the only available information about vaccine hesitancy (e.g., when large-scale surveys have not been conducted). Furthermore, the large samples and the speed with which data are collected and made available make real-time analysis possible for what has become a volatile topic. As data collected through social media platforms become more widely used, we anticipate that validation studies will be conducted, with improvements made in the sampling, weighting, and interpretation of the data.
The large volume of timely data on vaccine hesitancy has provided an opportunity to develop spatially detailed estimates of vaccine hesitancy (i.e., mapping by location). For the United States, surveys administered through Facebook have been used to estimate vaccine hesitancy according to week and ZIP code. These spatial analyses show that vaccine hesitancy varies substantially within a county. For example, vaccine hesitancy ranges from 7 to 49% across ZIP codes within the rural Stearns County, Minnesota. Such widespread variation within a county is common in all U.S. states (Figure 2).
Spatially refined estimates of vaccine hesitancy have proved to be useful in local efforts to increase vaccination rates.75,76 The information has been used by community outreach programs to tailor their efforts to local areas that have the greatest need. Other groups have used local patterns to help to decide where to provide mobile vaccination clinics and where to initiate other measures for reducing barriers to vaccination. Local information can also be used to monitor the effect of local interventions, including the effect of various types of vaccination mandates.
In the future, large and complex data sets on vaccine hesitancy, often referred to as big data, can be analyzed according to spatial identifiers such as ZIP code and various individual characteristics, including race or ethnic group, age, sex, and occupation, which can help to further microtarget vaccination outreach efforts. This information is also potentially critical for monitoring progress toward vaccine equity.
One of the various challenges in taking such an approach to scale and applying it globally is the inequity in the access to and reach of digital media. As the digital revolution unfolds globally, the global health community must keep pace. The consequences of not doing so are loud and clear, as we have seen in the context of the Covid-19 pandemic with regard to the rapid spread of misinformation and consequent vaccine hesitancy.
0 comments:
Post a Comment