• View in gallery

    Total number of mosquito avoidance strategies listed by survey respondents in Key West, FL, and Tucson, AZ. Black bars = Key West, FL; white bars = Tucson, AZ.

  • 1.

    Tolle MA, 2009. Mosquito-borne diseases. Curr Probl Pediatr Adolesc Health Care 39: 97140.

  • 2.

    Añez G, Rios M, 2013. Dengue in the United States of America: a worsening scenario? Biomed Res Int 2013: 678645.

  • 3.

    Khormi HM, Kumar L, 2011. Modeling dengue fever risk based on socioeconomic parameters, nationality and age groups: GIS and remote sensing based case study. Sci Total Environ 409: 47134719.

    • Search Google Scholar
    • Export Citation
  • 4.

    Bhatt S, Gething PW, Brady OJ, Messina JP, Farlow AW, Moyes CL, Drake JM, Brownstein JS, Hoen AG, Sankoh O, Myers MF, George DB, Jaenisch T, Wint GRW, Simmons CP, Scott TW, Farrar JJ, Hay SI, 2013. The global distribution and burden of dengue. Nature 496: 504507.

    • Search Google Scholar
    • Export Citation
  • 5.

    Brady OJ, Gething PW, Bhatt S, Messina JP, Brownstein JS, Hoen AG, Moyes CL, Farlow AW, Scott TW, Hay SI, 2012. Refining the global spatial limits of dengue virus transmission by evidence-based consensus. PLoS Negl Trop Dis 6: e1760.

    • Search Google Scholar
    • Export Citation
  • 6.

    Gubler DJ, 2004. The changing epidemiology of yellow fever and dengue, 1900 to 2003: full circle? Comp Immunol Microbiol Infect Dis 27: 319330.

    • Search Google Scholar
    • Export Citation
  • 7.

    San Martín JL, Brathwaite O, Zambrano B, Solórzano JO, Bouckenooghe A, Dayan GH, Guzmán MG, 2010. The epidemiology of dengue in the Americas over the last three decades: a worrisome reality. Am J Trop Med Hyg 82: 128135.

    • Search Google Scholar
    • Export Citation
  • 8.

    Messina JP, Brady OJ, Pigott DM, Golding N, Kraemer MUG, Scott TW, Wint GRW, Smith DL, Hay SI, 2015. The many projected futures of dengue. Nat Rev Microbiol 13: 230239.

    • Search Google Scholar
    • Export Citation
  • 9.

    Thomas SM, Tjaden NB, van den Bos S, Beierkuhnlein C, 2014. Implementing cargo movement into climate based risk assessment of vector-borne diseases. Int J Environ Res Public Health 11: 33603374.

    • Search Google Scholar
    • Export Citation
  • 10.

    de Lourdes Muñoz M, Mercado-Curiel RF, Diaz-Badillo A, Pérez Ramirez G, Black WC, 2013. Gene flow pattern among Aedes aegypti populations in Mexico. J Am Mosq Control Assoc 29: 118.

    • Search Google Scholar
    • Export Citation
  • 11.

    Khormi HM, Kumar L, 2012. Assessing the risk for dengue fever based on socioeconomic and environmental variables in a geographical information system environment. Geospat Health 6: 171176.

    • Search Google Scholar
    • Export Citation
  • 12.

    Mondini A, Chiaravalloti-Neto F, 2008. Spatial correlation of incidence of dengue with socioeconomic, demographic and environmental variables in a Brazilian city. Sci Total Environ 393: 241248.

    • Search Google Scholar
    • Export Citation
  • 13.

    Reiter P, Lathrop S, Bunning M, Biggerstaff B, Singer D, Tiwari T, Baber L, Amador M, Thirion J, Hayes J, Seca C, Mendez J, Ramirez B, Robinson J, Rawlings J, Vorndam V, Waterman S, Gubler D, Clark G, Hayes E, 2003. Texas lifestyle limits transmission of dengue virus. Emerg Infect Dis 9: 8689.

    • Search Google Scholar
    • Export Citation
  • 14.

    Van Benthem BHB, Vanwambeke SO, Khantikul N, Burghoorn-Maas C, Panart K, Oskam L, Lambin EF, Somboon P, 2005. Spatial patterns of and risk factors for seropositivity for dengue infection. Am J Trop Med Hyg 72: 201208. Available at: http://www.ncbi.nlm.nih.gov/pubmed/15741558. Accessed April 12, 2015.

    • Search Google Scholar
    • Export Citation
  • 15.

    Dantes HG, Koopman JS, Addy CL, Zarate ML, Marin MA, Longini Júnior IM, Guttierez ES, Rodriguez VA, Garcia LG, Mirelles ER, 1988. Dengue epidemics on the Pacific Coast of Mexico. Int J Epidemiol 17: 178186. Available at: http://www.ncbi.nlm.nih.gov/pubmed/3384535. Accessed April 12, 2015.

    • Search Google Scholar
    • Export Citation
  • 16.

    Gujral IB, Zielinski-Gutierrez EC, LeBailly A, Nasci R, 2007. Behavioral risks for West Nile virus disease, northern Colorado, 2003. Emerg Infect Dis 13: 419425.

    • Search Google Scholar
    • Export Citation
  • 17.

    Centers for Disease Control and Prevention (CDC), 2003. Knowledge, attitudes, and behaviors about West Nile virus—Connecticut, 2002. MMWR 52: 886888. Available at: http://www.ncbi.nlm.nih.gov/pubmed/13679793. Accessed March 23, 2015.

    • Search Google Scholar
    • Export Citation
  • 18.

    Aquino M, Fyfe M, MacDougall L, Remple V, 2004. Protective behavior survey, West Nile virus, British Columbia. Emerg Infect Dis 10: 14991501.

  • 19.

    McCarthy TA, Hadler JL, Julian K, Walsh SJ, Biggerstaff BJ, Hinten SR, Baisley C, Iton A, Brennan T, Nelson RS, Achambault G, Marfin AA, Petersen LR, 2001. West Nile virus serosurvey and assessment of personal prevention efforts in an area with intense epizootic activity: Connecticut, 2000. Ann N Y Acad Sci 951: 307316. Available at: http://www.ncbi.nlm.nih.gov/pubmed/11797787. Accessed March 23, 2015.

    • Search Google Scholar
    • Export Citation
  • 20.

    Wilson SD, Varia M, Lior LY, 2005. West Nile Virus: the buzz on Ottawa residents' awareness, attitudes and practices. Can J Public Health 96: 109113. Available at: http://www.ncbi.nlm.nih.gov/pubmed/15850029. Accessed March 23, 2015.

    • Search Google Scholar
    • Export Citation
  • 21.

    Larson A, Bryan J, Howard P, McGinn D, 2000. Queenslanders' use of personal strategies to minimise risk of mosquito-borne disease. Aust N Z J Public Health 24: 374377. Available at: http://www.ncbi.nlm.nih.gov/pubmed/11011462. Accessed March 23, 2015.

    • Search Google Scholar
    • Export Citation
  • 22.

    Vitek CJ, Gutierrez JA, Dirrigl FJ, 2014. Dengue vectors, human activity, and dengue virus transmission potential in the lower Rio Grande Valley, Texas, United States. J Med Entomol 51: 10191028. Available at: http://www.ncbi.nlm.nih.gov/pubmed/25276932. Accessed March 23, 2015.

    • Search Google Scholar
    • Export Citation
  • 23.

    National Oceanic and Atmospheric Administration, 2014. Data Tools: 1981–2010 Normals. Available at: http://www.ncdc.noaa.gov/cdo-web/datatools/normals. Accessed January 1, 2015.

    • Search Google Scholar
    • Export Citation
  • 24.

    Fink TM, Hau B, Baird BL, Palmer S, Kaplan S, Ramberg FB, Mead DG, Hagedorn H, 1998. Aedes aegypti in Tucson, Arizona. Emerg Infect Dis 4: 703704.

  • 25.

    Dengue Fever Mosquito, 2015. Available at: http://keysmosquito.org/dengue-fever-mosquito/. Accessed January 1, 2016.

  • 26.

    Centers for Disease Control and Prevention, 2015. Surveillance Resources. Available at: http://www.cdc.gov/westnile/resourcepages/survresources.html. Accessed January 1, 2016.

    • Search Google Scholar
    • Export Citation
  • 27.

    Trout A, Baracco G, Rodriguez M, Barber J, Leal A, Radke E, Weis K, Stanek D, Stark L, Blackmore C, Gallagher G, Hunsperger E, Tomashek K, Gregory C, Sauber-Schatz E, 2010. Locally acquired dengue—Key West, Florida, 2009–2010. MMWR 59: 577581.

    • Search Google Scholar
    • Export Citation
  • 28.

    Radke EG, Gregory CJ, Kintziger KW, Sauber-Schatz EK, Hunsperger EA, Gallagher GR, Barber JM, Biggerstaff BJ, Stanek DR, Tomashek KM, Blackmore CGM, 2012. Dengue outbreak in Key West, Florida, USA, 2009. Emerg Infect Dis 18: 135137.

    • Search Google Scholar
    • Export Citation
  • 29.

    Bequaert J, 1946. Aedes aegypti, the yellow fever mosquito, in Arizona. Bull Brooklyn Entomol Soc 41: 157.

  • 30.

    Murphy D, 1953. Collection records of some Arizona mosquitoes. Entomol News 14: 233238.

  • 31.

    Engelthaler DM, Fink TM, Levy CE, Leslie MJ, 1997. The reemergence of Aedes aegypti in Arizona. Emerg Infect Dis 3: 241242.

  • 32.

    Dantés HG, Farfán-Ale JA, Sarti E, 2014. Epidemiological trends of dengue disease in Mexico (2000–2011): a systematic literature search and analysis. PLoS Negl Trop Dis 8: e3158.

    • Search Google Scholar
    • Export Citation
  • 33.

    Umberson D, 1987. Family status and health behaviors: social control as a dimension of social integration. J Health Soc Behav 28: 306319.

    • Search Google Scholar
    • Export Citation
  • 34.

    Lindsey NP, Staples JE, Lehman JA, Fischer M, 2010. Surveillance for human West Nile virus disease—United States, 1999–2008. MMWR Surveill Summ 59: 117. Available at: http://www.ncbi.nlm.nih.gov/pubmed/20360671. Accessed March 24, 2015.

    • Search Google Scholar
    • Export Citation
  • 35.

    Hayden MH, Cavanaugh JL, Tittel C, Butterworth M, Haenchen S, Dickinson K, Monaghan AJ, Ernst KC, 2015. Post outbreak review: dengue preparedness and response in Key West, Florida. Am J Trop Med Hyg 93: 397400.

    • Search Google Scholar
    • Export Citation

 

 

 

Mosquito Avoidance Practices and Knowledge of Arboviral Diseases in Cities with Differing Recent History of Disease

View More View Less
  • 1 University of Arizona, Tucson, Arizona.
  • 2 National Center of Atmospheric Research, Boulder, Colorado.
  • 3 University of Colorado Boulder, Boulder, Colorado.
  • 4 University of Arizona Cancer Center, Tucson, Arizona.

As the range of dengue virus (DENV) transmission expands, an understanding of community uptake of prevention and control strategies is needed both in geographic areas where the virus has recently been circulating and in areas with the potential for DENV introduction. Personal protective behaviors such as the use of mosquito repellent to limit human–vector contact and the reduction of vector density through elimination of oviposition sites are the primary control methods for Aedes aegypti, the main vector of DENV. Here, we examined personal mosquito control measures taken by individuals in Key West, FL, in 2012, which had experienced a recent outbreak of DENV, and Tucson, AZ, which has a high potential for introduction but has not yet experienced autochthonous transmission. In both cities, there was a positive association between the numbers of mosquitoes noticed outdoors and the overall number of avoidance behaviors, use of repellent, and removal of standing water. Increased awareness and perceived risk of DENV were associated with increases in one of the most effective household prevention behaviors, removal of standing water, but only in Key West.

Introduction

Dengue virus (DENV), chikungunya virus (CHIKV), and Zika virus (ZIKAV) are emerging as significant threats in communities where the primary vector, Aedes aegypti, is established.13 DENV currently affects populations in 125 countries with 100 million annual symptomatic infections and nearly 4 billion people at risk.4,5 Models project expansion of Ae. aegypti into new locations as climatic conditions become more supportive of mosquito development and survival.4,68 Although the suitable habitat range expands because of climate, geographic spread of Ae. aegypti is aided by increased globalization and transportation.9,10

The presence of competent vectors does not always lead to subsequent viral transmission.2 Although climatic shifts may make an area suitable for vector presence, human behavior modifies the incidence of disease. Socioeconomic factors influence vector–human contact through higher use of air conditioning with closed windows, screened windows and doors, and better overall housing infrastructure.3,1115 Furthermore, studies of arboviral diseases demonstrate that disease knowledge and perceived risk may drive human preventive behavior to reduce vector–human contact and mosquito density.1622

To ascertain the factors that influence mosquito avoidance strategies, we conducted a knowledge, attitudes, and practices (KAP) survey to evaluate the role that mosquito and disease knowledge, perceived risk of disease, experience with public health campaigns, and perception of mosquito control responsibility play in influencing residents' preventive behaviors. Our study areas include two cities with differing histories of arbovirus transmission: Key West, FL, and Tucson, AZ. Key West has a tropical wet and dry savanna climate with an average annual rainfall of 1,012 mm and an average annual minimum temperature of 22.8°C and maximum of 28.1°C. Tucson has a subtropical steppe climate with averages of 294 mm rainfall, 13.2°C minimum temperature, and 28.4°C maximum temperature.23 Although environmentally distinct, both have established populations of Ae. aegypti.24,25 Residents of Key West experienced an outbreak of DENV in 2009–2010. In 2015, there were 60 imported DENV cases in Arizona, but no locally acquired transmission was detected.26 Although no local DENV transmission has been reported in Tucson, residents have been exposed to mosquito control campaigns related to West Nile virus (WNV) transmission, a disease that is endemic to the city. The purpose of this study was to use household surveys to examine differences in prevention strategies undertaken within the two cities with established Ae. aegypti populations, one with a history of DENV and one without.

Materials and Methods

Site description.

Before 2009, the last known locally acquired cases of DENV in Florida occurred in 1934.27 An outbreak of DENV was detected in Key West and neighboring Stock Island, FL, in 2009–2010, with a total of 93 autochthonous cases of DENV confirmed by the Florida State Health Department.27,28 Outreach and educational activities were conducted by the Florida Mosquito Controls District and the Monroe County Health Department during the outbreak.

Aedes aegypti populations were noted in Tucson between 1931 and 1946,29,30 after which they remained undetected until 1994.24,31 Although there have been no reported autochthonous DENV cases in Tucson, DENV cases are reported annually in Hermosillo, Sonora, Mexico, just 250 miles south of Tucson.32 In 2014, over 60 cases of DENV were reported in Yuma, AZ. All cases had a history of cross-border travel, with most traveling across the border to San Luis Rio Colorado, Sonora (Arizona Department of Health Services, personal communication). Thus, although no autochthonous DENV cases have been confirmed yet in Arizona, cross-border travel from endemic areas is indicative of the risk of viral introduction.

Study design and sample selection.

Key West and Stock Island (hereafter referred to as Key West).

We conducted household surveys from mid-June to mid-July 2012 to assess KAP with respect to DENV and residential mosquito control. A total of 400 residential parcels were randomly selected for inclusion in the survey using data obtained from the Monroe County Appraisers Office. Parcel data reflect distinct property boundaries and can be assigned as residential, governmental, commercial, or multiple uses. Residential parcels and multiple-use parcels that included residential use were included for selection. If a parcel had multiple residential units listed, that parcel was repeated in the database for an equivalent number of times to ensure all residential units had an equal probability of selection. Household surveys were conducted from mid-June to mid-July 2012 reflecting the period with highest mosquito activity. The sampling period was restricted to 1 month to reduce any seasonal variability in responses.

Tucson.

Residential surveys were conducted in late July and August 2012. Timing coincided with the peak of mosquito activity after the onset of the monsoon rains and was restricted to a 1-month period. Parcel data were obtained from Pima County. Because of the greater geographic size of Tucson, a random set of 20 points at least 1 km apart were generated using ArcGIS (Redlands, CA) within the city limits of Tucson. The closest residential neighborhoods, defined as two by two city blocks, were selected for participation. Within each neighborhood, 20 residential parcels were randomly selected for a total of 400 randomly selected parcels.

Surveys were conducted following the same standardized procedure in each city. To account for variability in activities and work schedules, two home visits were made at different times of day and on both weekdays and weekends. If residents were unreachable on the first attempt, informational material was left inviting them to participate and asking them to call to set up a time that would work best. If, after the second visit, the randomly selected household remained unreachable or refused to participate, a systematic approach was used to find a replacement for the home. Most of the respondents were from the replacement sampling (82% of Key West respondents and 75% of Tucson respondents). Reasons for nonparticipation included 1) direct refusal, 2) no one at home after two attempts, and 3) vacant or abandoned homes. In Tucson, replacements were not obtained for all randomly selected parcels because of time constraints (1-month sampling period), resulting in a final sample size of 375 households in this city. Environmental data were also collected, and the property was examined for the presence of potential mosquito oviposition sites and immature mosquitoes. Only the results of the KAP survey are presented in this analysis. To minimize bias, training was provided to all interviewers on ethics, maintaining a neutral attitude toward responses, and minimizing leading bias. This was of particular concern in Key West, an area that has experienced public health campaigns to educate community members about DENV and reduction of mosquito populations.

KAP survey content.

Survey topics included residential history, awareness of mosquito-borne diseases, perceived risk of mosquito-borne diseases, health-care-seeking behavior, knowledge of mosquito habitat, perceptions of mosquito control responsibility (personal and governmental), perceptions toward alternative control methods (biologic versus chemical), willingness to pay for public control of mosquitoes, and demographic information. Given the recent outbreak of DENV in Key West, questions were added to target knowledge of specific symptoms, local public health concerns, and local educational campaigns. Because of the lack of DENV transmission in Tucson, eight questions were added in Tucson to assess knowledge awareness and perceptions of local public health measures to combat WNV, which is endemic in Tucson.

Ethics statement.

A disclosure statement that was read to participating households indicated that they were under no obligation to participate and they could stop participation at any time. Written consent was not obtained as it would be an added link between study data and participants, and the survey was designed to pose minimal risks to participants. Participants were provided a copy of the disclosure statement with contact information for the principal investigator if questions arose. The study protocol, including a disclosure statement, was approved by the University of Arizona Human Subjects Research Committee and deemed exempt.

Data analysis.

Absolute and relative frequencies of responses were calculated for all survey respondents, for a total of 400 in Key West and 375 in Tucson. Three separate dichotomous outcomes were defined to examine respondents' participation in mosquito avoidance strategies: 1) the reported use of five or more mosquito avoidance practices, regardless of the established efficacy of each method (see Table 2 for full list); 2) reported frequency of emptying standing water at least once per week or after each rainfall, a frequency considered sufficient to prevent complete development of immature Ae. aegypti mosquitoes; and 3) reported use of mosquito repellent to be “often” or greater (in contrast to “sometimes” or less).

To examine correlates with these mosquito avoidance strategies, we used univariate logistic regression models to compare those who undertook prevention strategies and those who did not. Factors assessed in these models were demographic characteristics and KAP regarding mosquitoes and arboviral diseases. Multivariable models were explored for potential confounding and effect modification but are not presented, as there was no significant influence on parameter estimates or overall model goodness of fit.

Multiple imputation was performed to examine the impact of missing data in the household income question (38.5% for Key West and 18.4% for Tucson). Sensitivity analyses were performed to examine suitability of model parameters. Several cut points for the threshold of total avoidance strategies were explored. Questions that pertained only to Tucson were excluded from the calculations of total strategies used. None of these adjustments substantively changed the results from those presented here.

Results

Respondent characteristics differed between the two study sites (Table 1). In Tucson, respondents were younger, more likely to be Hispanic, and reported lower income overall (30.4%, < $35,000 annually, compared with 13.5% in Key West < $35,000 annually).

Table 1

Characteristics of survey respondents

 Key West (N = 400*)Tucson (N = 375*)
n (%)n (%)
Age group (year)
 18–3598 (25.3)108 (29.6)
 36–5077 (19.9)88 (24.1)
 51–65121 (31.3)97 (26.6)
 ≥ 6691 (23.5)72 (19.7)
Gender (female)183 (46.1)198 (53.4)
Race
 White non-Hispanic247 (63.8)203 (55.2)
 Hispanic77 (19.9)133 (36.1)
 Black36 (9.3)11 (3.0)
 American Indian/Alaska Native5 (1.3)6 (1.6)
 Asian/Pacific Islander14 (3.6)5 (1.4)
 Multiracial0 (0)4 (1.1)
 Other8 (2.1)6 (1.6)
Own home228 (61.1)273 (74.2)
Household income ($)
 < 35,00054 (13.5)114 (30.4)
 35,000–49,99931 (7.8)50 (13.3)
 50,000–74,99952 (13.0)62 (16.5)
 75,000–99,99937 (9.3)25 (6.7)
 ≥ 100,00072 (18.0)55 (14.7)
 Declined154 (38.5)69 (18.4)
Highest level of education completed
 Less than high school30 (7.7)37 (10.0)
 High school graduate93 (23.9)59 (16.0)
 Some college77 (19.8)83 (22.5)
 Associate's19 (4.9)31 (8.4)
 Bachelor's107 (27.5)88 (23.9)
 Graduate or professional63 (16.2)71 (19.2)

Not all categories sum to the total number of individuals because of missing data.

Respondents in Key West were significantly more likely to report using each of the mosquito avoidance strategies than respondents in Tucson (Table 2). Not only did Key West respondents report greater use of specific strategies, but they also reported use of a greater number of the total avoidance strategies (Figure 1). Respondents in Key West also reported checking for and removing standing water from their property more frequently than those in Tucson (P = 0.01) (Table 2).

Table 2

Respondents reporting use of mosquito avoidance strategies in Key West, FL, and Tucson, AZ

 Key WestTucsonP value
N (%)N (%)
Wear protective clothing122 (30.5)112 (29.9)0.85*
Burn citronella candles140 (35.0)78 (20.8)< 0.001*
Drain stagnant water285 (71.3)176 (46.9)< 0.001*
Call mosquito control107 (26.8)12 (3.2)< 0.001*
Burn coils59 (14.8)25 (6.7)< 0.001*
Burn tiki torches72 (18.0)18 (4.8)< 0.001*
Spray insecticide163 (40.8)113 (30.1)0.002*
Stay indoors217 (54.3)113 (30.1)< 0.001*
Fix and install screensN/A80 (21.3)N/A
Clear yard of brush and long weedsN/A116 (30.9)N/A
Other5 (1.3)2 (0.5)0.45
Nothing27 (6.8)89 (23.7)< 0.001*
Reported frequency of checking for standing water0.01*
 Never78 (21.1)102 (29.1)
 Once or twice a month47 (12.7)41 (11.7)
 Once a week59 (15.9)52 (14.9)
 More than once a week53 (14.3)45 (12.9)
 Daily75 (20.3)41 (11.7)
 After every rainfall58 (15.7)69 (19.7)
Water checking frequency§0.0497*
 Ineffective frequency125 (33.8)143 (40.9)
 Effective frequency245 (66.2)207 (59.1)

Calculated by χ2 test.

Calculated by Fisher's exact test.

Some avoidance questions were not asked of residents in both cities and are therefore excluded.

An effective frequency of checking for standing water was considered to be once a week, more than once a week, daily, or after every rainfall.

Figure 1.
Figure 1.

Total number of mosquito avoidance strategies listed by survey respondents in Key West, FL, and Tucson, AZ. Black bars = Key West, FL; white bars = Tucson, AZ.

Citation: The American Society of Tropical Medicine and Hygiene 95, 4; 10.4269/ajtmh.15-0732

Logistic regression was used to explore factors associated with utilization of mosquito avoidance practices (Key West, Table 3; Tucson, Table 4). In both cities, all older age groups (> 35 years) had higher odds of reporting that they cleared standing water from their yard on an effective schedule (those older than 65 years relative to 18–35 years old, Key West odds ratio [OR] = 2.48, 95% confidence interval [CI] = 1.33–4.61; Tucson OR = 2.09, 95% CI = 1.12–3.91). In Key West, age was also positively associated with use of insect repellent: those older than 65 years had 68% decreased odds of using repellent (OR = 0.32, 95% CI = 0.12–0.86) as compared with the 18- to 35-year-old group. Key West respondents that owned their home had lower odds of reporting effective water removing practices than non-homeowners (OR = 0.54, 95% CI = 0.34–0.84). Key West residents with children living in the home had higher odds of practicing multiple mosquito avoidance strategies as compared with those without children in the household (OR = 1.76, 95% CI = 1.06–2.93).

Table 3

The association between respondent demographic characteristics and reported avoidance strategies used in Key West, FL

 OR (95% CI)
Reports practicing five or more avoidance strategies 87 (21.8%)Uses repellent often or always 61 (15.3%)Checks for standing water at least once a week or after each rainfall 245 (66.2%)
Age group (year)
 18–35Ref.Ref.Ref.
 36–501.46 (0.74, 2.89)1.25 (0.60, 2.61)1.99 (1.05, 3.79)*
 51–651.01 (0.53, 1.91)0.76 (0.37, 1.55)3.78 (2.07, 6.91)*
 ≥ 660.56 (0.26, 1.20)0.32 (0.12, 0.86)*2.48 (1.33, 4.61)*
Own home
 Yes0.91 (0.55, 1.51)1.78 (1.01, 3.14)*0.54 (0.34, 0.84)*
 NoRef.Ref.Ref.
Children live in the home
 Yes1.76 (1.06, 2.93)*1.31 (0.72, 2.37)0.79 (0.49, 1.28)
 NoRef.Ref.Ref.
Residency duration
 < 1 year0.61 (0.20, 1.88)1.00 (0.27, 3.66)0.26 (0.10, 0.67)*
 1–5 years0.81 (0.40, 1.64)1.59 (0.71, 3.55)0.31 (0.16, 0.61)*
 ≥ 5 years0.61 (0.35, 1.08)1.02 (0.52, 2.01)0.90 (0.52, 1.57)
 LifelongRef.Ref.Ref.

CI = confidence interval; OR = odds ratio.

P < 0.05 by Wald χ2 test for this level of the variable.

Table 4

The association between respondent demographic characteristics and reported avoidance strategies used in Tucson, AZ

 OR (95% CI)
Reports practicing five or more avoidance strategies 58 (15.5%)Uses repellent often or always 61 (16.4%)Checks for standing water at least once a week or after each rainfall 207 (59.1%)
Age group (year)
 18–35Ref.Ref.Ref.
 36–501.85 (0.89, 3.87)1.52 (0.71, 3.28)2.95 (1.61, 5.39)*
 51–651.73 (0.83, 3.58)1.68 (0.80, 3.52)3.16 (1.74, 5.75)*
 ≥ 660.85 (0.35, 2.06)0.77 (0.31, 1.93)2.09 (1.12, 3.91)*
Own home
 Yes0.69 (0.36, 1.33)0.79 (0.40, 1.53)0.64 (0.39, 1.04)
 NoRef.Ref.Ref.
Children live in the home
 Yes1.44 (0.85, 2.45)1.34 (0.77, 2.33)0.93 (0.60, 1.44)
 NoRef.Ref.Ref.
Residency duration
 < 1 year1.08 (0.13, 9.11)1.02 (0.12, 8.63)
 1–5 years1.88 (0.72, 4.88)0.91 (0.29, 2.84)0.60 (0.26, 1.38)
 ≥ 5 years1.31 (0.74, 2.30)1.04 (0.58, 1.85)0.84 (0.53, 1.32)
 LifelongRef.Ref.Ref.

CI = confidence interval; OR = odds ratio.

P < 0.05 by Wald χ2 test for this level of the variable.

No individuals met these criteria.

Associations were also explored between the avoidance strategies and knowledge and perceptions about mosquitoes and arboviruses (Tables 5 and 6). In general, respondents who noticed fewer mosquitoes had lower odds of participating in avoidance strategies than those noticing many mosquitoes. This relationship is noted in both the Key West (OR = 0.53, 95% CI = 0.28–1.00) and Tucson (OR = 0.39, 95% CI = 0.22–0.68) analyses. Participants in Key West were asked about their knowledge of mosquito breeding sites. Compared with those who did not know where mosquitoes lay eggs or did not report a standing water source, those who reported either standing water or a specific oviposition container had higher odds of reporting using multiple avoidance strategies (OR = 2.37, 95% CI = 1.02–5.52 and OR = 3.33, 95% CI = 1.36–8.15, respectively). Furthermore, respondents knowledgeable about mosquito oviposition sites had higher odds of checking their yards for standing water (stated standing water: OR = 2.36, 95% CI = 1.29–4.32; stated a specific container: OR = 1.97, 95% CI = 1.00–3.87). Dengue awareness was associated with increased participation in avoidance strategies to a greater extent in Key West than in Tucson. Key West respondents who had heard of DENV had higher odds of checking their yard for standing water (OR = 2.18, 95% CI = 1.31–3.65). Key West residents had higher odds of using several avoidance practices when they thought it was more likely that they or a family member would become infected with DENV (unlikely to equally likely/unlikely versus very unlikely: OR = 1.81, 95% CI = 1.01–3.24; likely to very likely versus very unlikely: OR = 2.60, 95% CI = 1.12–6.05). Among Key West respondents, recent exposure to educational messages was associated with utilization of five strategies or more (OR = 1.94, 95% CI = 1.13–3.34) but not with the specific strategies of repellent use or standing water removal. Though 85.9% of Tucson residents had heard of WNV (Table 7), overall respondents' knowledge and perceptions of the disease was not significantly associated with any mosquito avoidance behavior.

Table 5

The association between respondent mosquito and arbovirus perceptions with reported avoidance strategies used in Key West, FL

 OR (95% CI)
Reports practicing five or more avoidance strategies 87 (21.8%)Uses repellent often or always 61 (15.3%)Checks for standing water at least once a week or after each rainfall 245 (66.2%)
Mosquitoes noticed outside
 None0.30 (0.10, 0.87)*0.26 (0.06, 1.13)0.53 (0.28, 1.00)*
 Very few to moderateRef.Ref.Ref.
 Quite a few to many1.59 (0.90, 2.78)2.38 (1.29, 4.37)*1.52 (0.82, 2.81)
Stated mosquito breeding sites
 Do not know or not standing waterRef.Ref.Ref.
 Salt marshes and puddles0.63 (0.07, 5.54)0.63 (0.07, 5.54)2.50 (0.70, 8.92)
 Just standing water2.37 (1.02, 5.52)*1.74 (0.74, 4.09)2.36 (1.29, 4.32)*
 At least one container stated3.33 (1.36, 8.15)*1.33 (0.50, 3.49)1.97 (1.00, 3.87)*
Believes control of mosquitoes is their own responsibility
 Strongly disagree0.85 (0.08, 9.30)1.92 (0.27, 13.6)0.50 (0.10, 2.48)
 Disagree1.28 (0.23, 7.10)1.28 (0.23, 7.10)0.55 (0.15, 1.96)
 Neither agree nor disagreeRef.Ref.Ref.
 Agree1.96 (0.54, 7.04)1.76 (0.49, 6.37)1.63 (0.67, 3.95)
 Strongly agree2.65 (0.77, 9.18)1.25 (0.35, 4.42)2.81 (1.19, 6.64)*
Would take preventative action if they had a mosquito problem
 Yes0.98 (0.60, 1.61)1.49 (0.82, 2.72)2.01 (1.29, 3.14)*
 NoRef.Ref.Ref.
Heard of dengue
 Yes1.91 (1.00, 3.63)0.89 (0.47, 1.69)2.18 (1.31, 3.65)*
 NoRef.Ref.Ref.
Dengue seriousness
 Not heard of0.55 (0.28, 1.08)1.08 (0.55, 2.11)0.49 (0.28, 0.84)*
 Not at all0.73 (0.34, 1.58)0.71 (0.28, 1.77)1.04 (0.53, 2.03)
 Slight to veryRef.Ref.Ref.
 Extremely1.77 (0.89, 3.54)1.13 (0.48, 2.65)1.49 (0.69, 3.22)
Perceived likelihood of dengue infection
 Not heard of0.76 (0.37, 1.58)1.32 (0.63, 2.78)0.50 (0.28, 0.90)*
 Very unlikelyRef.Ref.Ref.
 Unlikely to equally likely/unlikely1.81 (1.01, 3.24)*1.63 (0.84, 3.17)1.24 (0.72, 2.13)
 Likely to very likely2.60 (1.12, 6.05)*0.84 (0.24, 2.91)1.38 (0.54, 3.50)
Level of dengue awareness
 Never heard ofRef.Ref.Ref.
 Heard of1.70 (0.41, 7.17)0.76 (0.20, 2.84)1.09 (0.39, 3.02)
 Heard of outbreak1.68 (0.41, 6.78)0.76 (0.22, 2.58)1.90 (0.71, 5.11)
 Knows someone who had dengue2.60 (0.67, 10.13)1.53 (0.48, 4.91)3.64 (1.33, 9.93)*
Knows of the Key West mosquito control website
 Yes2.69 (1.20, 6.03)*2.07 (0.83, 5.13)1.86 (0.73, 4.73)
 NoRef.Ref.Ref.
Heard of any mosquito control and prevention information in preceding 2 months
 Yes1.94 (1.13, 3.34)*1.18 (0.61, 2.30)1.55 (0.88, 2.71)
 NoRef.Ref.Ref.

CI = confidence interval; OR = odds ratio.

P < 0.05 by Wald χ2 test for this level of the variable.

Table 6

The association between respondent mosquito and arbovirus perceptions with reported avoidance strategies used in Tucson, AZ

 OR (95% CI)
Reports practicing five or more avoidance strategies 58 (15.5%)Uses repellent often or always 61 (16.4%)Checks for standing water at least once a week or after each rainfall 207 (59.1%)
Mosquitoes noticed outside
 None0.20 (0.06, 0.66)*0.66 (0.27, 1.64)0.39 (0.22, 0.68)*
 Very few to moderateRef.Ref.Ref.
 Quite a few to many1.48 (0.80, 2.74)3.37 (1.81, 6.30)*1.02 (0.58, 1.80)
Believes control of mosquitoes is their own responsibility
 Strongly disagree1.83 (0.49, 6.86)4.81 (1.07, 21.75)*1.16 (0.52, 2.56)
 Disagree4.84 (1.37, 17.05)*2.52 (0.53, 11.92)0.52 (0.24, 1.11)
 Neither agree nor disagreeRef.Ref.Ref.
 Agree2.83 (0.77, 10.42)3.97 (0.86, 18.41)0.80 (0.36, 1.77)
 Strongly agree3.23 (0.85, 12.29)5.91 (1.26, 27.66)*0.70 (0.30, 1.61)
Would take preventative action if they had a mosquito problem
 Yes1.17 (0.63, 2.16)0.81 (0.44, 1.49)2.11 (1.29, 3.44)*
 NoRef.Ref.Ref.
Heard of dengue
 Yes1.61 (0.95, 2.74)1.10 (0.63, 1.91)1.33 (0.86, 2.05)
 NoRef.Ref.Ref.
Dengue seriousness
 Not heard of0.61 (0.32, 1.19)1.02 (0.49, 2.10)1.17 (0.68, 2.02)
 Not at all0.96 (0.44, 2.10)0.96 (0.39, 2.36)2.95 (1.42, 6.15)*
 Slight to veryRef.Ref.Ref.
 Extremely1.23 (0.22, 7.06)5.72 (1.10, 29.70)*0.71 (0.15, 3.46)
Perceived likelihood of dengue infection
 Not heard of0.66 (0.36, 1.19)1.08 (0.56, 2.05)0.65 (0.40, 1.07)
 Very unlikelyRef.Ref.Ref.
 Unlikely to equally likely/unlikely1.07 (0.45, 2.57)1.32 (0.50, 3.44)0.58 (0.27, 1.24)
 Likely to very likely2.10 (0.43, 10.35)3.46 (0.71, 16.93)1.23 (0.24, 6.46)
Level of dengue awareness
 Never heard ofRef.Ref.Ref.
 Heard of1.43 (0.82, 2.51)1.16 (0.66, 2.04)1.43 (0.91, 2.26)
 Knows someone who had dengue3.29 (1.20, 8.98)*0.66 (0.15, 3.02)0.78 (0.30, 2.05)
Heard of WNV
 Yes2.34 (0.89, 6.11)0.93 (0.43, 2.02)1.54 (0.85, 2.81)
 NoRef.Ref.Ref.
WNV seriousness
 Not heard of0.39 (0.15, 1.03)1.16 (0.52, 2.60)0.68 (0.37, 1.26)
 Not at all0.57 (0.25, 1.27)1.10 (0.52, 2.34)0.99 (0.55, 1.79)
 Slight to veryRef.Ref.Ref.
 Extremely0.99 (0.40, 2.43)1.55 (0.62, 3.88)1.63 (0.71, 3.72)
Perceived likelihood of WNV infection
 Not heard of0.50 (0.18, 1.39)1.14 (0.48, 2.72)0.60 (0.31, 1.17)
 Very unlikelyRef.Ref.Ref.
 Unlikely to 50/501.19 (0.65, 2.18)1.01 (0.53, 1.93)0.76 (0.46, 1.25)
 Likely to very likely1.70 (0.66, 4.36)1.59 (0.59, 4.27)1.80 (0.69, 4.67)
Level of WNV awareness
 Never heard ofRef.Ref.Ref.
 Heard of2.20 (0.84, 5.79)0.94 (0.43, 2.05)1.51 (0.82, 2.75)
 Know someone with WNV4.48 (1.24, 16.2)0.75 (0.18, 3.10)2.14 (0.75, 6.15)
Heard of Fight the Bite
 Yes0.93 (0.38, 2.25)1.04 (0.41, 2.60)1.02 (0.49, 2.12)
 NoRef.Ref.Ref.
Heard of any mosquito control and prevention information in preceding 2 months
 Yes1.44 (0.80, 2.61)1.67 (0.91, 3.10)1.16 (0.69, 1.94)
 NoRef.Ref.Ref.

CI = confidence interval; OR = odds ratio; WNV = West Nile virus.

P < 0.05 by Wald χ2 test for this level of the variable.

Table 7

Mosquito and disease knowledge among respondents in Key West, FL, and Tucson, AZ

 Key West (N = 400*)Tucson (N = 375*)
n (%)n (%)
Mosquitoes noticed outside
 None271 (67.8)230 (61.3)
 Very few to moderate52 (13.0)73 (19.5)
 Quite a few to many77 (19.3)72 (19.2)
Stated mosquito breeding sites
 Do not know or no standing water64 (16.0)
 Salt marshes and puddles14 (3.5)
 Just standing water222 (55.5)
 At least one container stated100 (25.0)
Mosquitoes come from nearby sources
 Strongly disagree84 (21.6)34 (10.6)
 Disagree108 (27.8)72 (22.5)
 Neither agree nor disagree64 (16.5)54 (16.9)
 Agree95 (24.5)89 (27.8)
 Strongly agree37 (9.5)71 (22.2)
Believes control of mosquitoes is their own responsibility
 Strongly disagree10 (2.5)95 (25.6)
 Disagree21 (5.3)94 (25.3)
 Neither agree nor disagree26 (6.6)41 (11.1)
 Agree123 (31.2)82 (22.1)
 Strongly agree214 (54.3)59 (15.9)
Would take preventative action if they had a mosquito problem
 Yes259 (64.8)278 (74.1)
Heard of dengue
 Yes308 (77.0)158 (42.5)
Dengue seriousness
 Not heard of91 (23.5)214 (60.4)
 Not at all52 (13.4)73 (20.6)
 Slight to very199 (51.4)60 (16.9)
 Extremely45 (11.6)7 (2.0)
Perceived likelihood of DENV infection
 Not heard of91 (23.6)214 (58.8)
 Very unlikely140 (36.4)108 (29.7)
 Unlikely to equally likely/unlikely124 (32.2)35 (9.6)
 Likely to very likely30 (7.8)7 (1.9)
Level of dengue awareness
 Never heard of15 (3.8)214 (57.5)
 Heard of60 (15.0)139 (37.4)
 Heard of outbreak231 (57.8)
 Knows someone who had dengue94 (23.4)19 (5.1)
Knows of the Key West mosquito control website
 Yes27 (6.8)
Heard of any mosquito control and prevention information in preceding 2 months
 Yes98 (32.3)91 (27.7)
Heard of WNV
 Yes322 (85.9)
WNV seriousness
 Not heard of53 (14.6)
 Not at all62 (17.0)
 Slight to very216 (59.3)
 Extremely33 (9.1)
Perceived likelihood of WNV infection
 Not heard of53 (14.9)
 Very unlikely160 (45.1)
 Unlikely to 50/50113 (31.8)
 Likely to very likely29 (8.2)
Level of WNV awareness
 Never heard of53 (14.1)
 Heard of300 (80.0)
 Know someone with WNV22 (5.9)
Heard of Fight the Bite
 Yes27 (8.4)

DENV = dengue virus; WNV = West Nile virus.

Not all categories sum to the total number of individuals because of missing data.

Some avoidance questions were not asked of residents in both cities and are therefore excluded.

The number of individuals classified as “Never heard of dengue” is smaller than previous questions because of individuals that responded they had never heard of dengue but subsequently stated they had heard of the outbreak or knew someone with dengue.

Discussion

With recent DENV activity in Key West and the potential for introduction of DENV into Tucson, it is important to assess the use of mosquito avoidance strategies among community members to limit the spread of mosquito-borne viruses. Greater use of mosquito avoidance strategies was evident in Key West compared with Tucson (Table 2). Similar to previous literature, factors associated with use of avoidance strategies varied by strategy.1622

Several patterns of behavior were consistent between the two cities. In both cities, older individuals had higher odds of checking their yard and empty standing water. This finding may be more reflective of available time, leisure activity, or home maintenance practices between age groups. Mosquito avoidance behaviors in both cities were associated with the number of mosquitoes noticed when outdoors. Interestingly, in both cities > 60% of participants indicated that they noticed no mosquitoes outdoors (Table 7). There were consistently lower odds of avoidance behaviors among those who never noticed mosquitoes outdoors and higher odds of avoidance behavior if participants noticed “quite a few” or “very many” mosquitoes. Rather than for disease prevention, these behavioral differences may be a reaction to the nuisance aspect of mosquito biting as has been demonstrated in previous literature.19,21

The residential status of individuals in Key West is associated with mosquito avoidance practices. Homeowners in Key West had higher odds of clearing standing water from their yard than renters. This could be a result of the highly visible presence of public mosquito control in the area. Homeowners in Key West may depend upon mosquito control to perform these tasks. Renters may be unaware of existing mosquito control efforts and, therefore, perform breeding site reduction activities on their own. Indeed, an independent association existed in which individuals who felt mosquito control was their responsibility being more likely to remove standing water. However, Key West homeowners were more likely to report that they agreed or strongly agreed that mosquito control was their own responsibility (93.7% among owners, 71.5% among renters, P < 0.001, χ2). Key West residents were also more likely to participate in more than five avoidance strategies if they have children living in the home; evidence suggests that having children in the home may actually minimize the practice of negative health behaviors.33

Although there are significant differences in overall use of avoidance strategies between the two cities, it is important to note that all the strategies listed are not of equal effectiveness. A summary score placing equal weight to each activity is intended to measure the respondent's total perceived mosquito avoidance efforts, and is not intended to indicate their level of risk reduction. Of the commonly recommended avoidance strategies, Key West residents more frequently reported staying indoors and removing standing water than Tucson residents, but there was no difference in the reported frequency of wearing protective clothing, which was quite low for both cities. It is also important to note that although there is a statistically significant difference in what is considered to be an effective rate of checking for standing water between the cities, the magnitude of the difference (66.2% in Key West versus 59.1% in Tucson) may be a reflection of more annual rainfall in Key West compared with Tucson.23

Disease awareness and perception of risk appear to play a bigger role in Key West than in Tucson. In Key West, there was a positive association between both knowledge of and personal experience with DENV, which was associated with higher frequency of removing of standing water. Further, there was a positive association between heightened perception of disease risk and using at least five avoidance strategies. These relationships are similar to those seen in other studies of mosquito avoidance and arbovirus prevention.16,17,19,21,22 However, there are notably very few significant associations between DENV knowledge and mosquito avoidance in Tucson. There are two possible explanations for this discrepancy: 1) in the absence of community experience with the disease, these associations may be attenuated; or 2) there were so few people with knowledge of DENV that the CIs were too large to detect a difference. However, there was also no association between perceptions of WNV and mosquito avoidance in Tucson, which contrasts with previous studies during the emergence period of WNV.34 The first case of WNV in Arizona occurred in 2003, and residents may have become complacent and less likely to exhibit concern and carry out the associated avoidance behaviors. Despite similar levels of participants having heard mosquito control and prevention messages (32.3% and 27.7%, Key West and Tucson, respectively; Table 7), reported exposure to such educational messages was only associated with increased mosquito avoidance behaviors in Key West. However, only reporting using five or more strategies was associated with having heard these messages. The most frequently recommended prevention strategies of removing standing water and using mosquito repellant were not associated despite a survey of Key West stakeholders rating the communication of educational messages to the public as being one of the most effective methods of reducing dengue risk.35

To our knowledge, this study is the first to examine mosquito-related behavioral differences between residents in cities, both with Ae. aegypti populations, but one with and one without DENV transmission. We explored both the total number of reported mosquito avoidance behaviors and those behaviors previously identified as effective in reducing disease transmission. Although surveys were standardized and interviewers were trained to minimize leading bias, participants were aware that the survey was asking about viruses transmitted by mosquitoes and may have self-reported responses they felt were expected by the interviewers. This could have resulted in an overestimation of the frequency of these practices taking place. However, the lower reporting of several of the strategies, such as repellent use, outdoor clothing, and so forth, indicates that individuals were not just answering yes to every strategy they had heard. We attempted to minimize selection bias by randomly selecting our participants and making multiple recruitment attempts on different days and at different times of day. Sampling with replacement in a systematic manner increased the sample size and was associated with randomly selected geographic locations (the households initially selected). However, response bias is still possible. It is unclear in what direction this would drive the overall estimates of reported practices, but we anticipate that it would have less impact on the results of the regression analysis where associations are examined between these practices and other characteristics of responders. Prevalence rates of practices are likely to be the most biased. Finally, the generalizability of these results may be limited to other communities with Ae. aegypti populations, with no transmission or recent DENV introduction.

Understanding the avoidance behaviors undertaken is critical given recent arboviral outbreaks of CHIKV and ZIKAV in the western hemisphere. Just as both of our study sites have the potential for DENV transmission, they are also at risk of local transmission of CHIKV and ZIKAV. The results herein should inform stakeholders that although communication of effective control strategies to the community is an important part of controlling mosquito populations, education alone may not be enough to ensure sustained effective mosquito avoidance behaviors.

This research suggests that additional work is needed to expand and evaluate the effects of public health campaigns on mosquito avoidance behaviors. The results suggest that increased awareness of the disease and knowledge about the mosquito vector are associated with participation in effective prevention practices. However, results exploring WNV in Tucson suggest that public health educators should be alerted that this relationship may decline if the disease becomes endemic. Future research efforts will be needed to see if the associations noted in Key West diminish over time, or if additional/sustained public health campaigns may be effective in maintaining and expanding the use of effective avoidance strategies.

ACKNOWLEDGMENTS

We thank Michael Doyle of the Florida Keys Mosquito Control District and Carina Blackmore of the Florida Department of Health. We would like to acknowledge Amit Arora, Mindy Butterworth, Pablo Reyes-Castro, Brent Cook, Linsay Edinger, Donna Feng, Etse Gebru, Alisa Jeffry, Eileen Jeffrey-Guttierez, Kevin Hayden, Andrea Mannell, Andy Monaghan Jaclyn Pierson, Syed Ali Raza, Hilary Rees, Adam Resnick, Melissa Roberts, and Laura Tippit for data collection and Carina Blackmore, Lisa Conti, Robert Eadie, Danielle Stanek, Christopher Tittel, and Emily Zielinski-Gutierrez for protocol and survey development and community engagement.

  • 1.

    Tolle MA, 2009. Mosquito-borne diseases. Curr Probl Pediatr Adolesc Health Care 39: 97140.

  • 2.

    Añez G, Rios M, 2013. Dengue in the United States of America: a worsening scenario? Biomed Res Int 2013: 678645.

  • 3.

    Khormi HM, Kumar L, 2011. Modeling dengue fever risk based on socioeconomic parameters, nationality and age groups: GIS and remote sensing based case study. Sci Total Environ 409: 47134719.

    • Search Google Scholar
    • Export Citation
  • 4.

    Bhatt S, Gething PW, Brady OJ, Messina JP, Farlow AW, Moyes CL, Drake JM, Brownstein JS, Hoen AG, Sankoh O, Myers MF, George DB, Jaenisch T, Wint GRW, Simmons CP, Scott TW, Farrar JJ, Hay SI, 2013. The global distribution and burden of dengue. Nature 496: 504507.

    • Search Google Scholar
    • Export Citation
  • 5.

    Brady OJ, Gething PW, Bhatt S, Messina JP, Brownstein JS, Hoen AG, Moyes CL, Farlow AW, Scott TW, Hay SI, 2012. Refining the global spatial limits of dengue virus transmission by evidence-based consensus. PLoS Negl Trop Dis 6: e1760.

    • Search Google Scholar
    • Export Citation
  • 6.

    Gubler DJ, 2004. The changing epidemiology of yellow fever and dengue, 1900 to 2003: full circle? Comp Immunol Microbiol Infect Dis 27: 319330.

    • Search Google Scholar
    • Export Citation
  • 7.

    San Martín JL, Brathwaite O, Zambrano B, Solórzano JO, Bouckenooghe A, Dayan GH, Guzmán MG, 2010. The epidemiology of dengue in the Americas over the last three decades: a worrisome reality. Am J Trop Med Hyg 82: 128135.

    • Search Google Scholar
    • Export Citation
  • 8.

    Messina JP, Brady OJ, Pigott DM, Golding N, Kraemer MUG, Scott TW, Wint GRW, Smith DL, Hay SI, 2015. The many projected futures of dengue. Nat Rev Microbiol 13: 230239.

    • Search Google Scholar
    • Export Citation
  • 9.

    Thomas SM, Tjaden NB, van den Bos S, Beierkuhnlein C, 2014. Implementing cargo movement into climate based risk assessment of vector-borne diseases. Int J Environ Res Public Health 11: 33603374.

    • Search Google Scholar
    • Export Citation
  • 10.

    de Lourdes Muñoz M, Mercado-Curiel RF, Diaz-Badillo A, Pérez Ramirez G, Black WC, 2013. Gene flow pattern among Aedes aegypti populations in Mexico. J Am Mosq Control Assoc 29: 118.

    • Search Google Scholar
    • Export Citation
  • 11.

    Khormi HM, Kumar L, 2012. Assessing the risk for dengue fever based on socioeconomic and environmental variables in a geographical information system environment. Geospat Health 6: 171176.

    • Search Google Scholar
    • Export Citation
  • 12.

    Mondini A, Chiaravalloti-Neto F, 2008. Spatial correlation of incidence of dengue with socioeconomic, demographic and environmental variables in a Brazilian city. Sci Total Environ 393: 241248.

    • Search Google Scholar
    • Export Citation
  • 13.

    Reiter P, Lathrop S, Bunning M, Biggerstaff B, Singer D, Tiwari T, Baber L, Amador M, Thirion J, Hayes J, Seca C, Mendez J, Ramirez B, Robinson J, Rawlings J, Vorndam V, Waterman S, Gubler D, Clark G, Hayes E, 2003. Texas lifestyle limits transmission of dengue virus. Emerg Infect Dis 9: 8689.

    • Search Google Scholar
    • Export Citation
  • 14.

    Van Benthem BHB, Vanwambeke SO, Khantikul N, Burghoorn-Maas C, Panart K, Oskam L, Lambin EF, Somboon P, 2005. Spatial patterns of and risk factors for seropositivity for dengue infection. Am J Trop Med Hyg 72: 201208. Available at: http://www.ncbi.nlm.nih.gov/pubmed/15741558. Accessed April 12, 2015.

    • Search Google Scholar
    • Export Citation
  • 15.

    Dantes HG, Koopman JS, Addy CL, Zarate ML, Marin MA, Longini Júnior IM, Guttierez ES, Rodriguez VA, Garcia LG, Mirelles ER, 1988. Dengue epidemics on the Pacific Coast of Mexico. Int J Epidemiol 17: 178186. Available at: http://www.ncbi.nlm.nih.gov/pubmed/3384535. Accessed April 12, 2015.

    • Search Google Scholar
    • Export Citation
  • 16.

    Gujral IB, Zielinski-Gutierrez EC, LeBailly A, Nasci R, 2007. Behavioral risks for West Nile virus disease, northern Colorado, 2003. Emerg Infect Dis 13: 419425.

    • Search Google Scholar
    • Export Citation
  • 17.

    Centers for Disease Control and Prevention (CDC), 2003. Knowledge, attitudes, and behaviors about West Nile virus—Connecticut, 2002. MMWR 52: 886888. Available at: http://www.ncbi.nlm.nih.gov/pubmed/13679793. Accessed March 23, 2015.

    • Search Google Scholar
    • Export Citation
  • 18.

    Aquino M, Fyfe M, MacDougall L, Remple V, 2004. Protective behavior survey, West Nile virus, British Columbia. Emerg Infect Dis 10: 14991501.

  • 19.

    McCarthy TA, Hadler JL, Julian K, Walsh SJ, Biggerstaff BJ, Hinten SR, Baisley C, Iton A, Brennan T, Nelson RS, Achambault G, Marfin AA, Petersen LR, 2001. West Nile virus serosurvey and assessment of personal prevention efforts in an area with intense epizootic activity: Connecticut, 2000. Ann N Y Acad Sci 951: 307316. Available at: http://www.ncbi.nlm.nih.gov/pubmed/11797787. Accessed March 23, 2015.

    • Search Google Scholar
    • Export Citation
  • 20.

    Wilson SD, Varia M, Lior LY, 2005. West Nile Virus: the buzz on Ottawa residents' awareness, attitudes and practices. Can J Public Health 96: 109113. Available at: http://www.ncbi.nlm.nih.gov/pubmed/15850029. Accessed March 23, 2015.

    • Search Google Scholar
    • Export Citation
  • 21.

    Larson A, Bryan J, Howard P, McGinn D, 2000. Queenslanders' use of personal strategies to minimise risk of mosquito-borne disease. Aust N Z J Public Health 24: 374377. Available at: http://www.ncbi.nlm.nih.gov/pubmed/11011462. Accessed March 23, 2015.

    • Search Google Scholar
    • Export Citation
  • 22.

    Vitek CJ, Gutierrez JA, Dirrigl FJ, 2014. Dengue vectors, human activity, and dengue virus transmission potential in the lower Rio Grande Valley, Texas, United States. J Med Entomol 51: 10191028. Available at: http://www.ncbi.nlm.nih.gov/pubmed/25276932. Accessed March 23, 2015.

    • Search Google Scholar
    • Export Citation
  • 23.

    National Oceanic and Atmospheric Administration, 2014. Data Tools: 1981–2010 Normals. Available at: http://www.ncdc.noaa.gov/cdo-web/datatools/normals. Accessed January 1, 2015.

    • Search Google Scholar
    • Export Citation
  • 24.

    Fink TM, Hau B, Baird BL, Palmer S, Kaplan S, Ramberg FB, Mead DG, Hagedorn H, 1998. Aedes aegypti in Tucson, Arizona. Emerg Infect Dis 4: 703704.

  • 25.

    Dengue Fever Mosquito, 2015. Available at: http://keysmosquito.org/dengue-fever-mosquito/. Accessed January 1, 2016.

  • 26.

    Centers for Disease Control and Prevention, 2015. Surveillance Resources. Available at: http://www.cdc.gov/westnile/resourcepages/survresources.html. Accessed January 1, 2016.

    • Search Google Scholar
    • Export Citation
  • 27.

    Trout A, Baracco G, Rodriguez M, Barber J, Leal A, Radke E, Weis K, Stanek D, Stark L, Blackmore C, Gallagher G, Hunsperger E, Tomashek K, Gregory C, Sauber-Schatz E, 2010. Locally acquired dengue—Key West, Florida, 2009–2010. MMWR 59: 577581.

    • Search Google Scholar
    • Export Citation
  • 28.

    Radke EG, Gregory CJ, Kintziger KW, Sauber-Schatz EK, Hunsperger EA, Gallagher GR, Barber JM, Biggerstaff BJ, Stanek DR, Tomashek KM, Blackmore CGM, 2012. Dengue outbreak in Key West, Florida, USA, 2009. Emerg Infect Dis 18: 135137.

    • Search Google Scholar
    • Export Citation
  • 29.

    Bequaert J, 1946. Aedes aegypti, the yellow fever mosquito, in Arizona. Bull Brooklyn Entomol Soc 41: 157.

  • 30.

    Murphy D, 1953. Collection records of some Arizona mosquitoes. Entomol News 14: 233238.

  • 31.

    Engelthaler DM, Fink TM, Levy CE, Leslie MJ, 1997. The reemergence of Aedes aegypti in Arizona. Emerg Infect Dis 3: 241242.

  • 32.

    Dantés HG, Farfán-Ale JA, Sarti E, 2014. Epidemiological trends of dengue disease in Mexico (2000–2011): a systematic literature search and analysis. PLoS Negl Trop Dis 8: e3158.

    • Search Google Scholar
    • Export Citation
  • 33.

    Umberson D, 1987. Family status and health behaviors: social control as a dimension of social integration. J Health Soc Behav 28: 306319.

    • Search Google Scholar
    • Export Citation
  • 34.

    Lindsey NP, Staples JE, Lehman JA, Fischer M, 2010. Surveillance for human West Nile virus disease—United States, 1999–2008. MMWR Surveill Summ 59: 117. Available at: http://www.ncbi.nlm.nih.gov/pubmed/20360671. Accessed March 24, 2015.

    • Search Google Scholar
    • Export Citation
  • 35.

    Hayden MH, Cavanaugh JL, Tittel C, Butterworth M, Haenchen S, Dickinson K, Monaghan AJ, Ernst KC, 2015. Post outbreak review: dengue preparedness and response in Key West, Florida. Am J Trop Med Hyg 93: 397400.

    • Search Google Scholar
    • Export Citation

Author Notes

* Address correspondence to Mary H. Hayden, National Center for Atmospheric Research, P.O. Box 3000, Boulder, CO 80301. E-mail: mhayden@ucar.edu

Financial support: This work was supported by NIH-NIAID grants R56AI091843 and R01AI091843.

Authors' addresses: Steven D. Haenchen, Kathleen Walker, Elizabeth E. Jacobs, Heidi E. Brown, Jayleen K. L. Gunn, Lindsay N. Kohler, and Kacey C. Ernst, University of Arizona, Tucson, AZ, E-mails: shaenchen@email.arizona.edu, krwalker@cals.arizona.edu, krwalker@cals.arizona.edu, heidibrown@email.arizona.edu, gunnj@email.arizona.edu, lschulz@email.arizona.edu, and kernst@email.arizona.edu. Mary H. Hayden and Katherine L. Dickinson, National Center of Atmospheric Research, Boulder, CO, E-mails: mhayden@ucar.edu and katied@ucar.edu.

Save