Awareness and Support of Release of Genetically Modified “Sterile” Mosquitoes, Key West, Florida, USA

After a dengue outbreak in Key West, Florida, during 2009–2010, authorities, considered conducting the first US release of male Aedes aegypti mosquitoes genetically modified to prevent reproduction. Despite outreach and media attention, only half of the community was aware of the proposal; half of those were supportive. Novel public health strategies require community engagement.

attitudes, and practices survey about mosquito control and dengue virus.
We collected information on demographics, perception of dengue risk, mosquito knowledge and prevention activities, and health care-seeking behavior, among other topics. Support was determined on a scale of 1 (strongly oppose) to 5 (strongly support). We requested reasons for participants' level of support; themes raised by ≥9 respondents were coded into study categories by 2 investigators (K.C.E. and M.H.H.).
In this study, the use of GM male mosquitoes results in death of offspring in the larval or pupal stage of gestation; because of this outcome, outreach activities in the area preceding the survey referred to the mosquitoes as "sterile." The survey we used included "sterile" because this term had been used in community awareness activities and should have been familiar to those who had heard of the proposed release, and we added "genetically modified" as a descriptor of the mosquitoes.
We divided participant groups into participants into those who had or had not heard of the release plans. We used logistic regression to assess associations between hearing of the release and possible explanatory factors. Missing values for household income were imputed. Distribution of levels of support of the release among those who had heard of the plan was stratified and tested for differences by demographic factors and participation in dengue and mosquito awareness and prevention activities. We used the Mantel-Haenszel test for trend for ordinal variables (e.g., education, income) and the χ 2 test of heterogeneity for categorical variables. We used ANOVA, a nested analysis of variance approach, for continuous variables.
Of the 400 participants (Table 1), 75 (18.8%) were from the originally selected households. Of the 386 participants who responded to the question of whether they had heard of the proposed release before the survey, 195 (51.1%) answered "yes." Prior awareness was more common in white non-Hispanics, residents with income levels >$50,000 per year, older adults, those who resided on Key West Island, and residents with knowledge of the local Action to Break the Cycle of Dengue public health campaign (Table 1). Among the 195 who were aware of the release, the distribution of support was: 9.7% strongly opposed, 8.2% opposed, 25.1% neutral, 22.1% supportive, and 34.9% strongly supportive. Men, less educated persons, and those willing to pay $100 or more for mosquito control were more likely to be strongly supportive ( Table 2). The most common reasons for opposing the release were disturbance of nature and that it was an unproven technology. Most supporters of the release expressed a desire to do anything to get rid of mosquitoes or preferred the method to chemicals and spraying (Figure). On the basis of effectiveness, safety, and/or lack of unintended consequences, 22 of the 195 indicated that their support was conditional.

Conclusions
For community acceptance of the release of GM mosquitoes, several issues must be addressed. Release of GM mosquitoes into the community should be transparent; therefore, the Florida Keys Mosquito Control District has begun to disseminate information through public events, articles, and presentations. Identification of solutions to reduce risk for vector-borne disease should involve stakeholders from the public, and community leaders in public health, vector control, and municipal administrators. Open communication with community members and stakeholders through a health advisory board was instrumental in quelling a 1989 invasion of Mediterranean fruit flies in California that had become a crisis event (10). In Key West and Stock Island, public awareness and communication campaigns had limited success. Awareness of the release varied across sections of the city and by demographic group. At the time of the survey, the release was planned for Key West; in Stock Island, awareness was much lower. Adjacent areas should be included in communications because residents and Ae. aegypti are mobile. (11). Knowledge of current events has been associated with gender, education level, race and ethnicity, and age (12). Outreach should target groups with a tendency towards lower awareness of public health measures. Support was more commonly reported than opposition among those aware of the release; a large portion was neutral. Most neutral respondents reported they did not know enough to make a decision, and many supporters wanted more information or had concerns. To progress from awareness to knowledge, to understanding, and then to decision-making would require considerable effort and improvement in overall scientific literacy (13). The scientific community is divided about the amount of information that should be provided to community members on highly technical vector control strategies such as the release of the OX513A mosquito (14). Benchmarks for acceptable engagement and support should be set by public health organizations before GM vector releases are planned, which will require input from scientists, stakeholders, and the community.
Strongly opposed participants most commonly reported unintended consequences or disturbing natural ecosystems as their reason for opposition. Conversely, some supporters considered the mosquito release a more natural way of controlling mosquito populations than insecticides. This was substantiated in a follow-up study (15).  This study has several limitations. Participants may not have fully represented the community because of seasonal housing closures and inaccessibility of some gated communities. A systematic replacement strategy was used to minimize bias. To obtain information on support, we provided a short statement about the release, modeled after earlier community outreach efforts and that used the term "sterile mosquito" instead of "genetically modified mosquito." We excluded responses of participants without prior awareness from our analysis because our informational statement was cursory. Follow-up studies in Key West that provided more extensive information yielded the same 9% strong opposition rate (15).
Introduction of GM mosquitoes has the potential to reduce mosquito-borne disease; however, little data exist on the type and extent of outreach required or community support needed to reduce opposition. As of December 2014, a short-term release of Oxitec OX513A mosquitoes is proposed on Key Haven, a peninsula adjacent to Key West. This is part of an application by Oxitec: Regulatory Clearance for Investigational Use of a New Animal Drug. This release is proposed before broader implementation in Key West or elsewhere in the Florida Keys (M.S. Doyle, unpub. data). If approved, this release could serve as a model of best practices for establishing community relations and engagement before implementing vector control strategies.

Figure.
Proportion of respondents reporting different themes for their level of support of plans to release genetically modified, referred to as "sterile," male mosquitoes on Key West, Florida, USA.