RESUMO
Web-based survey data collection has become increasingly popular, and limitations on in-person data collection during the COVID-19 pandemic have fueled this growth. However, the anonymity of the online environment increases the risk of fraud, which pose major risks to data integrity. As part of a study of COVID-19 and the return to in-person school, we implemented a web-based survey of parents in Maryland, USA, between December 2021 and July 2022. Recruitment relied, in part, on social media advertisements. Despite implementing many existing best practices, the survey was challenged by sophisticated fraudsters. In this paper, we describe efforts to identify and prevent fraudulent online survey responses and provide specific, actionable recommendations for identifying and preventing online survey fraud. Some strategies can be deployed within the web-based data collection platform such as Internet Protocol address logging to identify duplicate responses and comparison of client-side and server-side time stamps to identify responses that may have been completed by respondents outside of the surveys target geography. Additional approaches include the use of a 2-stage survey design, repeated within-survey and cross-survey validation questions, the addition of "speed bump" questions to thwart careless or computerized responders, and the use of optional open-ended survey responses to identify irrelevant responses. We describe best practices for ongoing survey data review and verification, including algorithms to simplify aspects of this review.
RESUMO
ImportanceReal-world data are needed to assess incidence and factors associated with breakthrough SARS-CoV-2 infections following vaccination. ObjectiveEstimate incidence of breakthrough infections and assess associations with risk factors using self-reported data from a large NC population sample. DesignProspective observational cohort study utilizing daily online survey data to capture information about COVID-19 symptoms, testing, and vaccination status. SettingSix health care systems in North Carolina with data collected between January 15, 2021 and September 24, 2021. ParticipantsAdult study participants who reported full vaccination with a COVID-19 mRNA or J&J non-replicating viral vector vaccine (n =16,020). ExposuresPotential community exposure to SARS-CoV-2. Main Outcome and MeasuresSelf-reported breakthrough infection. ResultsSARS-CoV-2 infection after vaccination was self-reported in 1.9% of participants, with an incidence rate of 7.3 per 100,000 person-years. Younger age (45-64 vs. 18-44: HR (95% CI) = 0.65 (0.51 - 0.82); 65+ vs. 18-44: HR (95% CI) = 0.59 (0.39 - 0.90)), and vaccination with J&J Ad26.COV2.S were associated with a higher risk of breakthrough infection compared to vaccination with Pfizer BNT162b2 (Ad26.COV2.S vs. BNT162b2: HR (95% CI) = 2.23 (1.40 - 3.56)), while participants vaccinated with mRNA-1273 (mRNA-1273 vs. BNT162b2: HR (95% CI) = 0.69 (0.50 - 0.96) and those residing in urban counties experienced a lower rate of SARS-CoV-2 breakthrough infection compared with those from suburban (HR (95% CI) = 1.39 (1.01 - 1.90) or rural (HR (95% CI) = 1.57 (1.16 - 2.11) counties. There was no significant association between breakthrough infection and participant sex, race, healthcare worker status, prior COVID-19 infection, routine mask use, or overall vaccination rate in the county of residence. Conclusions and RelevanceThis NC community-based observational study showed that the proportion of the cohort who self-report breakthrough SARS-CoV-2 infections was 7.3 events per 100,000 person-years. Younger adults, those vaccinated with J&J Ad26.COV2.S, and those residing in suburban or rural counties were at higher risk of breakthrough infections and should be targeted for additional risk mitigation strategies to decrease community transmission. Trial RegistrationThe COVID-19 Community Research Partnership is listed in clinicaltrials.gov (NCT04342884). Key PointsO_ST_ABSQuestionC_ST_ABSWhat are the characteristics of those with breakthrough infections after SARS-CoV-2 vaccination in North Carolinaa FindingsIn this NC-based observational study of 16,020 participants, 1.9% self-reported a positive SARS-CoV-2 viral test at least 2 weeks following full vaccination, reflecting an event rate of 7.3 infections per 100,000 person years. Rates were higher among younger participants, participants from more rural areas in North Carolina, and those vaccinated with J&J Ad26.COV2.S. MeaningOur results show a relatively low rate of COVID-19 infection following full vaccination. Younger adults and those vaccinated with J&J Ad26.COV2.S should be targeted for additional risk mitigation strategies.