Abstract
Consistent and effective use of personal prevention methods for tickborne diseases, including Lyme disease (LD), is dependent on risk awareness. To improve our understanding of the general U.S. population’s experiences with tick exposure and use of personal prevention methods, we used data from ConsumerStyles, a web-based, nationally representative questionnaire on health-related topics. Questions addressed tick bites and LD diagnosis in the last year, use of personal prevention methods to prevent tick bites, and willingness to receive a theoretical LD vaccine. Of 10,551 participants surveyed over three years, 12.3 % reported a tick bite for themselves or a household member in the last year, including 15.4 % of participants in high LD incidence (LDI) states, 16.3 % in states neighboring high LDI states, and 9.4 % in low LDI states. Participants in high LDI states and neighboring states were most likely to use personal prevention methods, though 46.6 % of participants in high LDI states and 53.9 % in neighboring states reported not using any method. Participants in low LDI states, adults ≥ 75 years of age, those with higher incomes, and those living in urban housing tended to be less likely to practice personal prevention methods. Likeliness to receive a theoretical LD vaccine was high in high LDI (64.5 %), neighboring (52.5 %), and low LDI (49.7 %) states. Targeted educational efforts are needed to ensure those in high LDI and neighboring states, particularly older adults, are aware of their risk of LD and recommended personal prevention methods.
Keywords: Tick-borne disease; Lyme disease; Risk; Prevention, humans
1. Introduction
Despite underreporting and variation in surveillance practices, Lyme disease (LD) is the most commonly reported vector-borne disease in the United States, with an estimated 329,000 cases diagnosed by clinicians annually (Nelson et al., 2015). The bacterial pathogens known to cause LD in the United States are Borrelia burgdorferi sensu stricto (ss) and B. mayonii (Pritt et al., 2016; Steere et al., 2016). These are transmitted to humans through the bite of Ixodes scapularis ticks, though Ixodes pacificus ticks also transmit B. burgdorferi ss in some parts of the Pacific Coast (Eisen et al., 2016a). Human LD cases are highly concentrated in the northeastern, mid-Atlantic, and upper midwestern portions of the country (Schwartz et al., 2017). However, surveillance data indicate that the distribution of cases has been expanding over the past two decades, as factors potentially related to weather and climate increase the geographical range of I. scapularis (Eisen, Eisen, and Beard paper (2016); Kugeler et al., 2015). As a result, states that share a border with high LD incidence (LDI) states are increasingly at risk of exposure to I. scapularis ticks that may carry B. burgdorferi ss and other human-infecting pathogens (Schwartz et al., 2017).
Avoiding tick exposure is the most effective way to prevent LD (Hayes and Piesman, 2003). Recommended prevention practices include measures that are meant to prevent tick exposure by an individual (personal prevention measures) and reduce ticks in the environment around households or on properties (environmental tick control measures). Personal prevention measures include using insect repellent, performing tick checks, showering soon after spending time outdoors, and wearing insecticide-treated or protective clothing (Eisen and Dolan, 2016; Hayes and Piesman, 2003; Connally et al., 2009). These methods have proven effective in preventing LD when practiced consistently (Vazquez et al., 2008; Connally et al., 2009). Environmental tick control methods can include landscaping to reduce tick habitat, deer fencing, host-targeted acaricide treatments, and acaricide applications on properties. Another potential method of LD prevention is a vaccine. While there is no LD vaccine currently on the market, one was available in the United States from 1998 until it was withdrawn in 2002 (Poland, 2011). Development and testing of new LD vaccines is in progress (Valneva, 2019).
Despite the demonstrated efficacy of many environmental tick control methods in reducing ticks on individual properties (Hinckley et al., 2016; Pound et al., 2009; Williams et al., 2018), these methods have not been shown to prevent LD and other tickborne diseases in humans (Hinckley et al., 2016). In addition, even if an environmental tick control measure is optimized and demonstrated to be effective on residential properties, it will likely be more complicated for a homeowner to implement properly and safely. Cost and inconsistent household participation can also be significant barriers to reducing tick populations at the community level (Hayes and Piesman, 2003). For example, though treating one’s own property with an acaricide may reduce tick abundance on that property, if few other households in the community use an environmental tick control method, tick abundance in that community will likely remain high. Household members may still be exposed to ticks outside of their own properties. A survey in Connecticut, a high LDI state, found that 65 % of participants reported ever using environmental tick control methods to prevent LD, while 99 % reported using personal prevention methods at least sometimes (Gould et al., 2008). As such, it may be more worthwhile for educational campaigns to focus on the promotion of personal prevention methods in at-risk communities rather than environmental tick control methods.
A recent knowledge, attitudes, and behaviors (KAB) survey in Connecticut and Maryland found that the only factor associated with use of most recommended personal prevention methods was perceived prevalence of LD (Niesobecki et al., 2019). Thus, to effectively prevent LD in high LDI and neighboring states, individuals must be aware of their risk (Hayes et al., 1999). We analyzed 2013–2015 data from an annual, nationally representative, health-based survey to assess the general U.S. population’s experiences with tick exposure, LD, tick bite prevention practices, and willingness to get a theoretical LD vaccine. The goal of this analysis was to assess how these factors differed by LDI category and to better define target populations for educational campaigns regarding LD risk and prevention.
2. Materials and methods
2.1. Study design and population
ConsumerStyles is an annual, cross-sectional survey conducted through a nationally representative online research panel called KnowledgePanel. Porter Novelli, a public relations firm specializing in health and social marketing, conducts the survey in three waves per year, and the Centers for Disease Control and Prevention (CDC) annually licenses results. Participants are randomly recruited by address from a pool of over 50,000 panelists 18 years of age and older and living in the United States, and asked questions about their knowledge and experiences with a variety of health-related topics. Individuals are recruited regardless of whether they have a landline phone or internet access, and are provided with a computer, if needed. The median time to complete the survey ranged from 26 min to 40 min over the three years. To ensure representativeness, data were weighted by KnowledgePanel using a correction factor based on nine different demographic factors (gender, age, household income, race/ethnicity, household size, education, census region, metro status, and internet access) to match the U.S. Current Population Survey (CPS) proportions.
2.2. Questions
Six different questions related to tick bites and LD were asked in the fall editions of ConsumerStyles from 2013 to 2015, and some questions were asked multiple years. We report results from four of these questions here. Responses were fielded beginning in either September or October each year. In 2015, the fall survey was administered in two waves; questions regarding ticks and LD were not asked in the second wave of the survey, and no results from this wave are reported here. Questions included in our analysis covered topics such as self-reported tick bites and LD diagnosis in the previous 12 months, tick bite prevention, and willingness to get a LD vaccine if one becomes available (Table 1).
Table 1.
ConsumerStyles tick-related questions and year(s) questions were asked.
| 1. What steps do you routinely take to prevent tick bites? Select all that apply: I wear repellent; I shower soon after coming indoors; I check my body for ticks daily; I take other steps not listed here; I don’t take any steps to prevent tick bites (2013, 2014) |
| 2. In the past 12 months, has anyone in your household been bitten by a tick? Select one: Yes, I was bitten; Yes, someone else in my household; No; Don’t know (2013, 2014, 2015) |
| 3. In the past 12 months, has anyone in your household been diagnosed with Lyme disease?a Select one: Yes, I was; Yes, someone else in my household; No; Don’t know (2013, 2014) |
| 4. If a vaccine that prevented Lyme disease were available, would you get vaccinated? Select one: Very likely; Somewhat likely; Somewhat unlikely; Very unlikely (2014, 2015) |
In 2014, participants were asked, “In the past 12 months, has anyone in your household been diagnosed with Lyme disease by a health care provider?”.
2.3. Analysis
Participants were categorized as residing in a high LDI state, neighboring state, or low LDI state based on classifications from 2008 to 2015 LD surveillance data (Schwartz et al., 2017) (Fig. 1). High LDI states included those with ≥ 10 confirmed cases of LD per 100,000 population during the reporting period, and neighboring states were those sharing a border with a high LDI state. Fourteen states were classified as high LDI, and eleven states and the District of Columbia were classified as neighboring. All other states were classified as low LDI.
Fig. 1.
United States by Lyme disease incidence category 2013–2015.
Frequencies, proportions, and chi-square tests for comparisons of categorical measures were generated using SAS JMP v. 13.2.1 (Cary, NC), and logistic regression was carried out in R v. 3.5.1 (Vienna, Austria) using the survey (Lumley, 2004, 2019) and MASS (Venables and Ripley, 2002) packages. All reported frequencies are unweighted and proportions are weighted. Statistical analyses were conducted using weighted counts. The Bonferroni correction was used to adjust for multiple comparisons when comparing routine use of prevention behaviors between LDI categories. Participants who did not answer a question were excluded from analysis for that specific question. Multi-variable logistic regression with backwards stepwise selection by AIC was used to examine associations between participant characteristics and personal protective measures routinely taken. LDI category, sex, age group, education level, race/ethnicity, combined household income, urban versus non-urban residence type, and having a child were included as covariates in the regression analysis. We categorized attached and detached single-family homes and mobile homes as non-urban residences, and buildings with at least two other units as urban. Alpha was set at 0.05 for all tests of statistical significance. ConsumerStyles data are considered exempt from institutional review board requirements.
3. Results
Survey response rates were 79.2 % (3502/4420) in 2013, 76.6 % (3520/4594) in 2014, and 79.6 % (3529/4432) in 2015. Demographic characteristics of survey participants matched the CPS proportions for each year (see Appendix A). Across all three years, 51.8 % of participants were female, with a median age of 54 years (range 18–94). High LDI states represented 27.8 % of all participants, neighboring states represented 21.7 %, and low LDI states represented 50.5 %. Sex, race/ethnicity, education, combined household income, urban residence type, and having a child all differed by state LDI category (Table 2).
Table 2.
Participant demographics by state Lyme disease incidence category (2013–2015).
| Characteristic | High LDI (N = 2989) | Neighboring (N = 2462) | Low LDI (N = 5100) | P-value | |||
|---|---|---|---|---|---|---|---|
| Unweighted No. | Weighted % | Unweighted No. | Weighted % | Unweighted No. | Weighted % | ||
| Sex | |||||||
| Male | 1496 | 46.1 | 1279 | 49.7 | 2631 | 48.8 | 0.02 |
| Female | 1493 | 53.9 | 1183 | 50.3 | 2469 | 51.2 | |
| Age in years | |||||||
| 18–24 | 218 | 12.5 | 181 | 11.7 | 355 | 12.2 | 0.05 |
| 25–34 | 350 | 16.1 | 295 | 17.5 | 677 | 18.4 | |
| 35–44 | 393 | 16.0 | 322 | 16.3 | 678 | 17.0 | |
| 45–54 | 567 | 19.3 | 432 | 18.0 | 923 | 18.1 | |
| 55–64 | 711 | 17.0 | 616 | 18.5 | 1105 | 15.5 | |
| 65–74 | 529 | 13.1 | 450 | 12.9 | 941 | 13.1 | |
| ≥ 75 | 221 | 6.0 | 166 | 5.1 | 421 | 5.7 | |
| Race/ethnicity | |||||||
| White | 2491 | 76.7 | 2037 | 76.9 | 3514 | 56.6 | <0.0001 |
| Black or African American | 223 | 8.7 | 256 | 12.2 | 510 | 12.4 | |
| Hispanic | 152 | 7.7 | 91 | 6.2 | 761 | 21.8 | |
| Other | 74 | 5.9 | 37 | 3.4 | 196 | 7.8 | |
| Education | |||||||
| Less than HSc | 142 | 8.4 | 169 | 13.0 | 362 | 12.6 | <0.0001 |
| HS | 967 | 31.0 | 832 | 32.7 | 1462 | 28.3 | |
| Some college | 866 | 27.4 | 724 | 27.8 | 1692 | 30.5 | |
| ≥ Bachelor degree | 1014 | 33.2 | 737 | 26.5 | 1584 | 28.6 | |
| Combined household income | |||||||
| < $25,000 | 440 | 13.3 | 470 | 20.8 | 972 | 19.8 | <0.0001 |
| $25–$49,999 | 685 | 19.6 | 675 | 23.4 | 1312 | 23.5 | |
| $50–$74,999 | 582 | 20.5 | 480 | 19.8 | 1039 | 20.5 | |
| ≥ $75,000 | 1282 | 46.6 | 837 | 36.0 | 1777 | 36.2 | |
| Residence type | |||||||
| Urban | 590 | 18.6 | 309 | 30.0 | 791 | 16.3 | < 0.0001 |
| Non-urban | 2396 | 81.3 | 2145 | 87.9 | 4283 | 83.1 | |
| Has child | |||||||
| Yes | 594 | 25.5 | 516 | 29.1 | 1131 | 30.6 | <0.0001 |
| No | 2391 | 74.4 | 1938 | 70.4 | 3961 | 69.2 | |
High school.
3.1. Tick bite
Four hundred sixty (12.2 %) participants in 2013, 432 (13.3 %) participants in 2014, and 417 (12.1 %) participants in 2015 reported that they or someone in their household had been bitten by a tick in the previous 12 months. The number of participants reporting a tick bite was not significantly different between years (p = 0.66). Across all three years, participants in neighboring states more commonly reported that they or someone in their household had experienced a tick bite in the previous 12 months (16.3 %) than high LDI (15.4 %) or low LDI states (9.4 %) (p < 0.0001).
3.2. Lyme disease diagnosis
Overall, 0.9 % of participants reported that they or someone in their household were diagnosed with LD in the previous 12 months. LD diagnoses did not differ between LDI categories (p = 0.145) or between 2013 and 2014, despite the clarification of LD diagnosis “by a health care provider” in 2014 (p = 0.52) (Table 3).
Table 3.
Number of participants who reported Lyme disease diagnosis for themselves and/or member of household in last 12 months by state Lyme disease incidence category.
| State LDI category | 2013 n (%) | 2014a n (%) | Both years n (%) |
|---|---|---|---|
| Overall | 28 (0.77) | 32 (1.1) | 60 (0.9) |
| High LDI | 18 (2.0) | 15 (1.4) | 33 (1.7) |
| Neighboring | 5 (0.5) | 7 (1.6) | 12 (1.0) |
| Low LDI | 5 (0.2) | 10 (0.8) | 15 (0.5) |
In 2014, participants were asked, “In the past 12 months, has anyone in your household been diagnosed with Lyme disease by a health care provider?”.
3.3. Routine use of personal prevention practices
The most commonly reported personal prevention method differed by LDI category. In high LDI states, performing daily tick checks was most common (28.7 %), whereas using repellent was most common in neighboring and low LDI states (24.1 % and 16.6 %, respectively) (Table 4).
Table 4.
Type of personal protective measure routinely taken by Lyme disease disease incidence category (2013, 2014).
| Prevention Measure | Total (N = 7022) n (%)a | High LDI (N = 1995) n (%) | Neighboring (N = 1632) n (%) | Low LDI (N = 3395) n (%) |
|---|---|---|---|---|
| Using repellent | 1405 (20.6) | 479 (25.2) | 366 (24.1) | 560 (16.6) |
| Daily tick checks | 1368 (19.4) | 570 (28.7) | 371 (23.3) | 427 (12.7) |
| Showering soon after coming indoors | 1075 (15.5) | 320 (16.2) | 280 (17.8) | 475 (14.3) |
| Insecticide-treated clothingb | 114 (3.3) | 46 (4.3) | 24 (1.9) | 44 (1.2) |
| Other steps | 759 (10.2) | 283 (14.1) | 144 (8.1) | 332 (9.1) |
| No steps | 4004 (57.6) | 929 (46.6) | 880 (53.9) | 2195 (65.3) |
Multiple answers were allowed. Totals may exceed 100 %.
Insecticide-treated clothing was listed as an option in 2014 only.
Using repellent was the most commonly selected prevention method across all LDI categories in 2013 and 2014 (20.6 %). Participants who lived in high LDI or neighboring states (OR = 1.74, 95 % CI: 1.45, 2.09; OR = 1.54, 95 % CI: 1.26, 1.88), were Black or African American (OR = 1.47, 95 % CI = 1.03, 2.11), and had a child (OR = 1.36, 95 % CI: 1.10, 1.68) were the most likely to routinely use repellent, while participants 75 years of age or older (OR = 0.55, 95 % CI: 0.34, 0.89) and those living in urban housing (OR = 0.69, 95 % CI: 0.54, 0.89) were the least likely (Table 5).
Table 5.
Selected characteristics of users of the most common personal protection measures (2013, 2014).
| Characteristic | Using repellent (N = 1405) | Daily tick checks (N = 1368) | Showering soon after coming indoors (N = 1075) | |||
|---|---|---|---|---|---|---|
| StateLDIcategory | ||||||
| High | 479 (34.1) | 1.74 (1.45–2.09) | 570 (41.7) | 2.72 (2.14–3.44) | 320 (29.8) | 1.22 (1.01–1.49) |
| Neighboring | 366 (26.0) | 1.54 (1.26–1.88) | 371 (27.1) | 1.83 (1.40–2.41) | 280 (26.0) | 1.26 (1.02–1.56) |
| Low | 560 (39.9) | reference | 427 (31.2) | reference | 475 (44.2) | reference |
| Sex | ||||||
| Male | 660 (47.0) | N.S.a | 737 (53.9) | 1.19 (0.97–1.45) | 586 (54.5) | 1.32 (1.11–1.56) |
| Female | 745 (53.0) | N.S. | 631 (46.1) | reference | 489 (45.5) | reference |
| Age in years | ||||||
| 18–24 | 96 (6.8) | reference | 90 (6.6) | reference | 84 (7.8) | reference |
| 25–34 | 192 (13.7) | 1.08 (0.78–1.50) | 169 (12.4) | 1.35 (0.95–1.92) | 167 (15.5) | 1.10 (0.78–1.56) |
| 35–44 | 213 (15.2) | 0.95 (0.67–1.34) | 180 (13.2) | 1.02 (0.71–1.47) | 126 (11.7) | 0.71 (0.50–1.02) |
| 45–54 | 273 (19.4) | 0.95 (0.69–1.31) | 286 (20.9) | 1.23 (0.89–1.71) | 202 (18.8) | 0.80 (0.58–1.12) |
| 55–64 | 313 (22.3) | 1.05 (0.77–1.44) | 315 (23.0) | 1.01 (0.73–1.39) | 235 (21.9) | 0.84 (0.60–1.18) |
| 65–74 | 253 (18.0) | 1.00 (0.73–1.37) | 241 (17.6) | 0.81 (0.58–1.13) | 206 (19.2) | 0.86 (0.61–1.22) |
| ≥ 75 | 65 (4.6) | 0.55 (0.34–0.89) | 87 (6.4) | 0.86 (0.57–1.31) | 55 (5.1) | 0.43 (0.28–0.68) |
| Race/ethnicity | ||||||
| White | 1094 (77.9) | 1.07 (0.81–1.43) | 1176 (86.0) | 2.11 (1.51–2.95) | 839 (78.0) | 0.96 (0.72–1.29) |
| Black or African American | 136 (9.7) | 1.47 (1.03–2.11) | 73 (5.3) | 1.25 (0.80–1.95) | 101 (9.4) | 1.23 (0.84–1.80) |
| Hispanic | 112 (8.0) | reference | 67 (4.9) | reference | 94 (8.7) | reference |
| Other | 63 (4.5) | 0.77 (0.48–1.24) | 52 (3.8) | 1.24 (0.72–2.13) | 41 (3.8) | 0.54 (0.32–0.92) |
| Education | ||||||
| HSb or less | 527 (37.5) | N.S. | 562 (41.1) | reference | 390 (36.3) | N.S. |
| Some college | 423 (30.1) | N.S. | 391 (28.6) | 0.80 (0.66–0.97) | 334 (31.1) | N.S. |
| ≥ Bachelor degree | 455 (32.4) | N.S. | 415 (30.3) | 0.86 (0.71–1.05) | 351 (32.7) | N.S. |
| Combined household income | ||||||
| < $25,000 | 230 (16.4) | reference | 240 (17.5) | reference | 195 (18.1) | reference |
| $25–$49,999 | 331 (23.6) | 0.80 (0.62–1.03) | 334 (24.4) | 0.64 (0.50–0.83) | 257 (23.9) | 0.83 (0.63–1.08) |
| $50–$74,999 | 292 (20.8) | 0.95 (0.73–1.23) | 278 (20.3) | 0.71 (0.55–0.91) | 235 (21.9) | 1.05 (0.80–1.38) |
| ≥ $75,000 | 552 (39.3) | 0.92 (0.73–1.17) | 516 (37.7) | 0.62 (0.48–0.79) | 388 (36.1) | 0.78 (0.61–0.99) |
| Residence type c | ||||||
| Urban | 175 (12.5) | 0.69 (0.54–0.89) | 124 (9.1) | 0.50 (0.38–0.66) | 123 (11.4) | 0.71 (0.54–0.92) |
| Non-urban | 1229 (87.5) | reference | 1239 (90.6) | reference | 949 (88.3) | reference |
| Has child | ||||||
| Yes | 366 (26.0) | 1.36 (1.10–1.68) | 331 (24.2) | 1.14 (0.92–1.41) | 245 (22.8) | N.S. |
| No | 1037 (73.8) | reference | 1035 (75.7) | reference | 829 (77.1) | reference |
N.S. denotes that this characteristic was not significant after performing backwards stepwise logistic regression and was therefore not included in the final model.
Residence types classified as “other” were excluded from regression analyses due to a small number of participants falling into this category (n = 3).
High school.
“Checking body for ticks daily” was the second most commonly selected prevention method (19.4 %). Participants who lived in high LDI and neighboring states (OR = 2.72, 95 % CI: 2.14, 3.44; OR = 1.83, 95 % CI: 1.40, 2.41), were White (OR = 2.11, 95 % CI: 1.51, 2.95), and had an income less than $25,000 (OR = 1.68, 95 % CI: 1.32, 2.13) were more likely to perform daily tick checks. Those who attended some college (OR = 0.80, 95 % CI: 0.66, 0.97) and lived in urban housing (OR = 0.50, 95 % CI: 0.38, 0.66) were the least likely to report routine use of this method.
Showering soon after coming indoors was the next most commonly selected prevention method across both years (15.5 %). Living in a high LDI or neighboring state (OR = 1.22, 95 % CI: 1.01, 1.49; OR = 1.26, 95 % CI: 1.02, 1.56) and being male (OR = 1.32, 95 % CI: 1.11, 1.56) were associated with this measure. Participants 75 years of age or older (OR = 0.43, 95 % CI: 0.28, 0.68), those with a race or ethnicity other than White, Black or African American, or Hispanic (OR = 0.54, 95 % CI: 0.32, 0.92), those with a combined household income of $75,000 or more (OR = 0.78, 95 % CI: 0.61, 0.99), and those who lived in urban housing (OR = 0.71, 95 % CI: 0.54, 0.92) were the least likely to report showering soon after coming indoors.
When included as an option in the 2014 survey, wearing insecticide-treated clothing was selected by 3.3 % of participants. Participants living in high LDI states were more likely to report wearing insecticide-treated clothing compared to participants from low LDI states (OR = 1.83, 95 % CI: 1.11, 3.02). Being male was also associated with routine use of this prevention practice (OR = 1.60, 95 % CI: 1.02, 2.52). Participants who had attended at least some college were the least likely to wear insecticide-treated clothing (OR = 0.57, 95 % CI: 0.34, 0.98).
Of the 10.2 % of participants who reported routinely taking other steps not listed to prevent tick bites, those who lived in high LDI states (OR = 1.87, 95 % CI: 1.43, 2.45) and had a bachelor’s degree or higher (OR = 1.42, 95 % CI: 1.11, 1.82) were the most likely to use other prevention methods; those 75 years of age and older were the least likely (0.47, 95 % CI: 0.29, 0.75).
Overall, more than half of all participants reported taking no routine steps to prevent tick bites (57.6 %), with those in neighboring and low LDI states being more likely than those in high LDI states to take no steps (OR = 1.44, 95 % CI: 1.21, 1.71; OR = 2.18, 95 % CI: 1.87, 2.53). Other characteristics associated with not practicing personal prevention included being over 75 years of age (OR = 1.68, 95 % CI: 1.19, 2.38), living in a residence classified as urban (OR = 1.54, 95 % CI: 1.26, 1.88), and having a combined income greater than $50,000 (OR = 1.37, 95 % CI: 1.14–1.64). Participants who were White were the least likely to report taking no steps (OR = 0.66, 95 % CI: 0.48, 0.90).
3.4. Willingness to vaccinate
The majority of participants across 2014 and 2015 reported being at least somewhat likely to receive a LD vaccine if one were to become available (54.5 %). Likeliness differed by LDI category; high LDI states had the highest proportion of participants at least somewhat likely to get the vaccine (64.5 %), followed by neighboring states (52.5 %) and low LDI states (49.7 %) (p < 0.0001).
4. Discussion
Results from the 2013–2015 fall ConsumerStyles surveys indicate tick exposure, use of personal prevention methods, and likeliness to receive a LD vaccine are more common in high LDI and neighboring states than low LDI states. Still, nearly half of all participants from high LDI states and more than half of participants from neighboring states reported not taking personal prevention steps for tick bites, suggesting that much of the population in at-risk areas may not be aware of their risk for LD and other tickborne diseases. While trends in demographic factors associated with using personal prevention practices aligned with previous studies, adults 75 years of age and older stood out as a group consistently less likely than other age groups to practice personal prevention methods.
In 2013 and 2014, more participants in neighboring states reported that they or a household member were bitten by a tick in the last year than those in high LDI states. Several neighboring states fall within geographic hotspots of Amblyomma americanum, a species of tick that does not transmit LD but is known to exhibit greater mobility and more aggressive host-seeking behavior than I. scapularis, leading to more frequent human-tick encounters (Schulze et al., 2006, 2005; Stromdahl and Hickling, 2012). Therefore, it is possible that those in neighboring states who reported that they or a household member were bitten by a tick in the past year either recognized tick bites more often or were bitten more often than those in high LDI states, due to the greater abundance of A. americanum in their state. Additionally, participants were not asked which state the tick bite occurred in, and some bites may have been acquired outside of the individual’s state of residence.
A small percentage (1.7 %) of participants in high LDI states reported that they or someone in their household were diagnosed with LD in the previous year, which did not differ significantly from participants in neighboring and low LDI states. This proportion is low in comparison to a KAB survey regarding LD prevention conducted in endemic areas of Connecticut, in which nearly 5% of participants reported a LD diagnosis in the past year (Gould et al., 2008). It should be noted, that despite a large overall sample size in our study, numbers of participants reporting LD diagnosis in the past year were very small when broken down by LDI category. Thus, it is difficult to draw conclusions from these data.
Results regarding routine use of personal prevention methods were fairly consistent with results from a similar question asked in a 2011 ConsumerStyles survey (Hook, et al., 2015). Overall proportions of participants reporting routine repellent use were similar in our analysis (20.6 %) and in Hook et al. (21.1 %), but tick checks were more frequently reported in 2011, and the proportion of participants taking no steps to prevent tick bites was slightly higher in our study (Hook et al., 2015). It is possible that routine use of tick checks was higher in 2011 than in 2013 and 2014 due to slight differences in wording. In 2011, the question specifically asked about routine steps taken to prevent tick bites “when the weather is warm in your area,” while in 2013 and 2014 this specification was not made (Hook et al., 2015). Additionally, the option provided in 2013 and 2014 was “I check my body for ticks daily”, and in 2011 it was “I check my body for ticks when I come in.” As such, it is possible that participants may practice tick checks when they come in from outside during tick season, but this does not necessarily equate to checking for ticks daily. Repellent use may not have differed between 2011 and 2013–2014 because it is also used to prevent bites from other arthropods, such as mosquitos.
Niesobecki et al. reported much higher proportions for routine use of personal protective measures in Connecticut and Maryland than we report for participants from high LDI states (Niesobecki et al., 2019), though a recent KAB study in Delaware reported results similar to ours for the same measures (Gupta et al., 2018). The population surveyed in Niesobecki et al. was recruited specifically to participate in a survey regarding tickborne disease and may have been more knowledgeable or interested in tickborne disease prevention than the convenience sample surveyed in the Delaware KAB and the representative sample surveyed here (Niesobecki et al., 2019).
Themes in demographic characteristics of participants reporting routine use of personal prevention methods in our study mostly aligned with themes in the literature. As expected, high LDI and neighboring states were generally more likely to practice personal prevention methods than those in low LDI states, likely due to more exposure to LD. Similar findings are reported in Hook et al., in which geographic regions that corresponded with high LDI and neighboring states had lower proportions of participants who did not take personal prevention steps (Hook et al., 2015). Living in urban housing was negatively associated with nearly every prevention practice; this is likely because participants living in urban settings may have less surrounding tick habitat, and subsequently, less risk of tick exposure than those in more rural areas. Daily tick checks were more commonly selected by participants of lower income and who attended some college. Niesobecki et al. reported similar results and proposed that individuals with lower income and education levels may be more likely to have occupations that require time outdoors, and less likely to practice personal prevention methods that are more costly (e.g., using repellent, wearing insecticide-treated clothing) (Niesobecki et al., 2019).
Adults 75 years of age and older were often the least likely to routinely use personal prevention methods. Schwartz et al. presented surveillance data that showed an increase in LD in older adults, with a peak among those 50–55 years of age (Schwartz et al., 2017). It is possible that LD is less easily recognized in older adults due to the sometimes non-specific symptoms of early LD and an increased potential for similar conditions, such as arthritis, that can be due to other causes. Studies have also shown that older adults may be more likely to have comorbidities at the time of LD onset and take longer to recover following treatment (Borsic et al., 2018; Weitzner et al., 2017). The increased potential for unrecognized LD in older adults, longer recovery period, and under-utilization of personal prevention methods makes this an important population to target in future educational campaigns.
Likeliness to get a LD vaccine if one were to become available was generally high among participants. Notably, half of all participants in low LDI states reported being at least somewhat likely to get a hypothetical LD vaccine—a surprising finding given LD risk is generally low in these states. While it is possible that some of these participants may spend time in higher risk areas, including other states, during the months when LD transmission commonly occurs, we hypothesize that this result may reflect members of the U.S. population who are generally in favor of vaccination as a primary disease prevention method. One study found that 94 % of parents surveyed by ConsumerStyles in 2010 had either already vaccinated or planned to vaccinate their children with all recommended childhood vaccines, indicating vaccine confidence in the United States is generally high (Kennedy et al., 2011). Thus, reasons for indicating willingness to get a LD vaccine may be related to confidence in vaccines in general rather than perceived risk of LD. Future research should further explore willingness to get a LD vaccine among various populations and factors that influence willingness.
4.1. Limitations
Results reported here are subject to limitations. First, all data from the ConsumerStyles surveys were self-reported and therefore subject to recall bias. Second, we report weighted proportions. Though weighted proportions increase the representativeness of responses to the general U.S. population, the un-weighted proportion of participants from neighboring states who reported regularly wearing insecticide-treated clothing was notably higher than the weighted proportion (3.1 % vs. 1.9 %, respectively) due to a small number of responses. These results are also subject to limitations regarding the phrasing of questions. For example, the question regarding LD diagnosis in the past year was changed in 2014 to specify diagnosis “by a provider.” As a result, responses in 2013 may have included diagnoses that were self-diagnoses or given by someone other than a provider. However, no significant difference in responses was found between the two years, indicating that this specification did not affect the results. Similarly, we did not explicitly define the term “household member” in the questions regarding tick bite and LD diagnoses by any household member in the past year. As such, some participants may consider their pets members of their household and may have answered “yes” to these questions based on their pets’ experiences. Misclassification may have occurred when assigning participants to an urban or non-urban residence based on housing type, causing us to over- or under-estimate the use of prevention methods in these strata. We also were not able to distinguish suburban households from urban or non-urban households. Thus, we were not able to characterize tick bite prevention behaviors for this specific demographic group. It is also possible that participants who used environmental or yard-based methods to control ticks on their properties would not have considered this a routine step taken to prevent tick bites, due to the wording of the question. As a result, these individuals may be included among those who selected they do not take any steps to prevent tick bites. Alternatively, these participants may have selected that they routinely take steps other than those listed to prevent tick bites. However, we did not collect further information from individuals who selected this option.
These findings are also limited by classifications made for stratifying survey participants. For current surveillance purposes in the United States, states are classified as either high LDI (>10 cases/100,000 population) or low LDI (< 10,000 cases/100,000 population). The states Schwartz et al. classified as neighboring fall into the low LDI category, and therefore our results for low LDI states may not be generalizable to all states that are considered low LDI for surveillance purposes (Schwartz et al., 2017). However, our results pertaining to neighboring states represent information usually not captured in surveillance data: tick exposure and personal prevention method use in states that are approaching, but not yet considered high LDI. Our results indicate that neighboring states are perhaps more similar to high LDI states than low LDI states in terms of tick exposure and use of personal prevention methods. Lastly, not all who reside within a particular region have the same risk of LD due to a number of environmental and individual factors that we were not able to capture here. Future studies assessing personal prevention behaviors for LD should include a measure of participants’ perceived risk of contracting LD and an assessment of individual and household behaviors that may contribute to overall risk.
5. Conclusions
Although tick exposure and use of personal protective measures are highest in high LDI and neighboring states, nearly half of all participants in these regions report not taking steps to prevent tick bite. More educational efforts are needed to ensure populations in these regions are aware of local incidence of LD. In particular, educational campaigns targeted to high risk groups who are less likely to routinely practice personal prevention, such as older adults, are needed to increase awareness of LD risk and knowledge of prevention methods. Lastly, a LD vaccine would likely be well-received by the U.S. population as a primary prevention method.
Acknowledgments
We wish to thank Anna Perea, CDC, for creating Fig. 1 and Brad Biggerstaff, CDC, for guidance on statistical analysis.
Appendix A
Participant demographics for the fall waves of the 2013, 2014, and 2015 ConsumerStyles surveys.
| ConsumerStyles 2013 (N = 3502) | ConsumerStyles 2014 (N = 3520) | ConsumerStyles 2015 (N = 3529) | ||||
|---|---|---|---|---|---|---|
| Sex | ||||||
| Male | 1762 | 47.7 | 1829 | 48.4 | 1815 | 48.6 |
| Female | 1740 | 52.3 | 1691 | 51.6 | 1714 | 51.4 |
| Age in years | ||||||
| 18–29 | 452 | 18.8 | 444 | 20.9 | 508 | 21.7 |
| 30–44 | 714 | 27.2 | 678 | 25.8 | 673 | 24.7 |
| 45–59 | 1079 | 27.6 | 1060 | 26.6 | 1028 | 26.3 |
| ≥ 60 | 1257 | 26.4 | 1338 | 26.7 | 1320 | 27.3 |
| Race/ethnicity | ||||||
| White | 2694 | 67.1 | 2682 | 66.4 | 2666 | 66.3 |
| Black or African American | 312 | 11.3 | 346 | 11.5 | 331 | 11.2 |
| Hispanic | 304 | 14.0 | 324 | 14.7 | 376 | 14.9 |
| Other | 192 | 7.6 | 168 | 7.5 | 156 | 7.7 |
| Education | ||||||
| Less than HS | 229 | 10.6 | 242 | 12.2 | 202 | 11.8 |
| HS | 1018 | 30.7 | 1151 | 29.7 | 1092 | 29.7 |
| Some college | 1127 | 29.2 | 1044 | 29.0 | 1111 | 29.0 |
| ≥ Bachelor | 1128 | 29.5 | 1083 | 29.1 | 1124 | 29.6 |
| Income | ||||||
| < $25,000 | 569 | 18.4 | 659 | 18.5 | 654 | 17.7 |
| $25–$49,999 | 922 | 23.4 | 875 | 22.0 | 875 | 21.9 |
| $50–$74,999 | 702 | 20.8 | 685 | 19.4 | 714 | 20.8 |
| ≥ $75,000 | 1309 | 37.5 | 1301 | 40.1 | 1286 | 39.6 |
| Employment status | ||||||
| Employed | 1890 | 55.0 | 1877 | 55.5 | 1901 | 57.8 |
| Not employed | 1612 | 45.0 | 1643 | 44.5 | 1628 | 42.2 |
Footnotes
CRediT authorship contribution statement
C.C. Nawrocki: Formal analysis, Writing - original draft, Writing - review & editing. A.F. Hinckley: Conceptualization, Methodology, Writing - review & editing, Supervision.
References
- Borsic K, Blagus R, Cerar T, Strle F, Stupica D, 2018. Clinical course, serologic response, and long-term outcome in elderly patients with early Lyme borreliosis. J. Clin. Med 7 (12), 506. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Connally NP, Durante AJ, Yousey-Hindes KM, Meek JI, Nelson RS, Heimer R, 2009. Peridomestic Lyme disease prevention: results of a population-based case-control study. Am. J. Prev. Med 37 (3), 201–206. [DOI] [PubMed] [Google Scholar]
- Eisen L, Dolan MC, 2016. Evidence for personal protective measures to reduce human contact with blacklegged ticks and for environmentally based control methods to suppress host-seeking blacklegged ticks and reduce infection with Lyme disease spirochetes in tick vectors and rodent reservoirs. J. Med. Entomol 53 (5), 1063–1092. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Eisen RJ, Eisen L, Beard CB, 2016b. County-scale distribution of Ixodes scapularis and Ixodes pacificus (Acari: ixodidae) in the continental United States. J. Med. Entomol 53 (2), 349–386. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Eisen RJ, Eisen L, Ogden NH, Beard CB, 2016a. Linkages of weather and climate with Ixodes scapularis and Ixodes pacificus (Acari: ixodidae), enzootic transmission of Borrelia burgdorferi, and Lyme disease in North America. J. Med. Entomol 53 (2), 250–261. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gould LH, Nelson RS, Griffith KS, Hayes EB, Piesman J, Mead PS, Cartter ML, 2008. Knowledge, attitudes, and behaviors regarding Lyme disease prevention among Connecticut residents, 1999–2004. Vector Borne Zoonotic Dis. 8 (6), 769–776. [DOI] [PubMed] [Google Scholar]
- Gupta S, Eggers P, Arana A, Kresse B, Rios K, Brown L, Sampson L, Kploanyi M, 2018. Knowledge and preventive behaviors towards tick-borne diseases in Delaware. Ticks Tick. Dis 9 (3), 615–622. [DOI] [PubMed] [Google Scholar]
- Hayes EB, Piesman J, 2003. How can we prevent Lyme disease? N. Engl. J. Med 348 (24), 2424–2430. [DOI] [PubMed] [Google Scholar]
- Hayes EB, Maupin GO, Mount GA, Piesman J, 1999. Assessing the prevention effectiveness of local Lyme disease control. J. Public Health Manag. Pract 5 (3), 84–92. [DOI] [PubMed] [Google Scholar]
- Hinckley AF, Meek JI, Ray JA, Niesobecki SA, Connally NP, Feldman KA, Jones EH, Backenson PB, White JL, Lukacik G, Kay AB, Miranda WP, Mead PS, 2016. Effectiveness of residential acaricides to prevent Lyme and other tick-borne diseases in humans. J. Infect. Dis 214 (2), 182–188. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hook SA, Nelson CA, Mead PS, 2015. U.S. public’s experience with ticks and tick-borne diseases: results from national HealthStyles surveys. Ticks Tick. Dis 6 (4), 483–488. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kennedy A, Lavail K, Nowak G, Basket M, Landry S, 2011. Confidence about vaccines in the United States: understanding parents’ perceptions. Health Aff. (Millwood) 30 (6), 1151–1159. [DOI] [PubMed] [Google Scholar]
- Kugeler KJ, Farley GM, Forrester JD, Mead PS, 2015. Geographic distribution and expansion of human Lyme disease, United States. Emerg. Infect. Dis 21 (8), 1455–1457. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lumley T, 2004. Analysis of complex survey samples. J. Stat. Softw 9 (1), 1–19. [Google Scholar]
- Lumley T, 2019. Survey: analysis of complex survey samples. R Package Version 3.35–1. [Google Scholar]
- Nelson CA, Saha S, Kugeler KJ, Delorey MJ, Shankar MB, Hinckley AF, Mead PS, 2015. Incidence of clinician-diagnosed Lyme disease, United States, 2005–2010. Emerg. Infect. Dis 21 (9), 1625–1631. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Niesobecki S, Hansen A, Rutz H, Mehta S, Feldman K, Meek J, Niccolai L, Hook S, Hinckley A, 2019. Knowledge, attitudes, and behaviors regarding tick-borne disease prevention in endemic areas. Ticks Tick. Dis 10 (6), 101264. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Poland GA, 2011. Vaccines against Lyme disease: what happened and what lessons can we learn? Clin. Infect. Dis 52 (3), s253–s258. [DOI] [PubMed] [Google Scholar]
- Pound JM, Miller JA, George JE, Fish D, Carroll JF, Schulze TL, Daniels TJ, Falco RC, Stafford KC, Mather TN, 2009. The United States Department of Agriculture’s northeast area-wide tick control project: summary and conclusions. Vector Borne Zoonotic Dis. 9 (4), 439–448. [DOI] [PubMed] [Google Scholar]
- Pritt BS, Mead PS, Johnson DKH, Neitzel DF, Respicio-Kingry LB, Davis JP, Schiffman E, Sloan LM, Schriefer ME, Replogle AJ, Paskewitz SM, Ray JA, Bjork J, Steward CR, Deedon A, Lee X, Kingry LC, Miller TK, Feist MA, Theel ES, Patel R, Irish CL, Petersen JM, 2016. Identification of a novel pathogenic Borrelia species causing Lyme borreliosis with unusually high spirochaetaemia: a descriptive study. Lancet Infect. Dis 16 (5), 556–564. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Schulze TL, Jordan RA, Schulze CJ, Mixson T, Papero M, 2005. Relative encounter frequencies and prevalence of selected Borrelia, Ehrlichia, and Anaplasma infections in Amblyomma americanum and Ixodes scapularis (Acari: ixodidae) ticks from central New Jersey. J. Med. Entomol 42 (3), 450–456. [DOI] [PubMed] [Google Scholar]
- Schulze TL, Jordan RA, Healy SP, Roegner VE, Meddis M, Jahn MB, Guthrie DL Sr., 2006. Relative abundance and prevalence of selected Borrelia infections in Ixodes scapularis and Amblyomma americanum (Acari: ixodidae) from publicly owned lands in Monmouth County, New Jersey. J. Med. Entomol 43 (6), 1269–1275. [DOI] [PubMed] [Google Scholar]
- Schwartz AM, Hinckley AF, Mead PS, Hook SA, Kugeler KJ, 2017. Surveillance for Lyme disease - United States, 2008–2015. Surveill. Summ 66 (22), 1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Steere AC, Strle F, Wormser GP, Hu LT, Branda JA, Hovius JW, Li X, Mead PS, 2016. Lyme borreliosis. Nat. Rev. Dis. Primers 2, 16090. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Stromdahl EY, Hickling GJ, 2012. Beyond Lyme: aetiology of tick-borne human diseases with emphasis on the south-eastern United States. Zoonoses Public Health 59 (2), 48–64. [DOI] [PubMed] [Google Scholar]
- Valneva, 2019. Lyme Disease - VLA15 (accessed 16 September 2020). https://valneva.com/research-development/lyme-disease.
- Vazquez M, Muehlenbein C, Cartter M, Hayes EB, Ertel S, Shapiro ED, 2008. Effectiveness of personal protective measures to prevent Lyme disease. Emerg. Infect. Dis 14 (2), 210–216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Venables WN, Ripley BD, 2002. Modern Applied Statistics With S, fourth ed. Springer, New York. [Google Scholar]
- Weitzner E, Visintainer P, Wormser GP, 2017. Impact of patient age on clinical features, serologic test reactivity and long-term outcome of culture-confirmed early Lyme disease. Diagn. Microbiol. Infect. Dis 89 (4), 300–302. [DOI] [PubMed] [Google Scholar]
- Williams SC, Stafford KC 3rd, Molaei G, Linske MA, 2018. Integrated control of nymphal Ixodes scapularis: effectiveness of white-tailed deer reduction, the entomopathogenic fungus Metarhizium anisopliae, and fipronil-based rodent bait boxes. Vector Borne Zoonotic Dis. 18 (1), 55–64. [DOI] [PubMed] [Google Scholar]