Mobile strategies have become a cost-effective, timely way to learn more about health behaviors.
In 2014, a measles outbreak occurred at Disneyland in Anaheim, California. What was remarkable about the outbreak wasn’t the just the damage it caused, but the fact that it happened at all. Thanks to the invention of a measles vaccine in the 1960s, the once-devastating illness was officially declared eliminated in 2000.
Those of us who study public health issues like vaccination need to understand which messages are getting through and whether misinformation is prompting people to adopt unhealthy behaviors, like avoiding important preventive health measures. To this end, it’s important to be able to conduct health surveys that are fast to plan and implement, and that help us understand at-risk populations.
Mobile strategies have become a cost-effective, timely way to collect public health data, particularly for at-risk or rare subgroups. We recently set out to assess the feasibility of using a mobile panel survey, Child Health Immunization Mobile Panel Survey (ChIMPS), to examine attitudes towards vaccination among those with young children (a low incidence population). The survey also assessed respondents’ trust in vaccine safety, public health information provided by the government, and whether they delayed vaccination.
Breaking Down the Study
In order to gauge whether the data we collected were consistent with national benchmarks, we patterned ChiMPS on the National Immunization Survey (NIS), which CDC uses to monitor vaccination coverage among children 19-35 months. An eligibility screener identified panel households with children ages 19-35 months. About 1,100 children were represented by the parents and other adult household members who responded. The survey lasted about 10 minutes. The respondents sample is 37% male, 27% with high school education or less, 65% white, and 85% of age 40 or younger. The respondents represented 1,115 eligible children 19-35 months old in their households. The characteristics of the parents and children in the survey sample were generally comparable to those of the NIS.
The mobile approach afforded us a few benefits. First, we could collect data faster. More than 1,000 respondents completed an interview within a week, a considerable reduction compared to similar phone or in-person surveys. The rapid collection of these data allowed for an assessment of attitudes and beliefs, and the impact on vaccination choices. Second, our findings — like the survey parameters — were generally consistent with NIS data.
Better Understanding of Attitudes Towards Vaccines
The good news? According to ChIMPS, the overwhelming majority of children — despite rampant vaccine misinformation — are being vaccinated. Most respondents (96.7%) reported intending to fully vaccinate their children, and the same number (96.7%) reported that their child has received at least one immunization. These numbers imply high levels of support and trust in CDC and vaccination, but a deeper dive in the data shows that even among parents who are vaccinating, misconceptions and fears about vaccine safety are impacting their beliefs—and behaviors.
Looking at the figure below, we can see that more than half of people who agree strongly with the statement “Vaccines may cause learning disabilities, such as autism” have delayed a vaccine for a non-medical reason. Among those who disagree strongly with the statement (which is, notably, the largest portion of respondents), only a small percentage reported vaccine delays.
In short, the children living in households where parents have significant safety concerns about vaccines may not have the same level of protection from vaccination, especially at younger ages.
Figure 2 provides even more context. Children in households where respondents strongly agreed with the statement about autism represented the largest portion of children who had not received all recommended vaccines, even though parents who agreed strongly with this statement were the minority overall. Most children in this group experienced vaccine delay for nonmedical reasons, while the majority of children in all other groups (respondents who somewhat agreed, disagreed, or were not sure about the statement on vaccines causing autism) received vaccines without unnecessary delays. Among those who disagree strongly with the statement about autism, the overwhelming majority have received all recommended vaccines and vaccine delay is less commonly reported.
This methodology allowed for a fast and cost effective way to collect vaccine attitudes and beliefs, but could be applied in many areas of public health research, including before and after surveys to gauge the effectiveness of public health interventions. This method can be used for collecting local level data or national samples. At ICF, we’ve also used similar methods to look at the public's use of cancer-related information and cancer risk behaviors in 6 cities, and also found these methods to be cost effective and swift.
How else do you think mobile panel surveys can help support public health research? Share your ideas with us on Facebook, Twitter, or LinkedIn.