Assessing the Repeatability of Tick Dragging as a Method for Ixodes scapularis Surveillance

Abstract

Tick dragging is an important tool used by public health for Ixodes scapularis surveillance to identify Lyme disease risk areas in Ontario, Canada. Concerns have been raised on the repeatability of tick dragging due to fluctuations that occur in the tick population in response to micro- and macroclimatic variations. Our objective was to assess the repeatability of tick dragging over a short timescale by examining three outcome measures: presence/absence of ticks, tick abundance, and likelihood of tick establishment based on an indicator developed by Clow et al. (2018). We conducted tick dragging twice per site within a 1-month period at a total of 15 sites in eastern and southern Ontario. Ixodes scapularis were detected at 11 sites. The outcome of presence/absence was consistent at 13 of 15 sites. Abundance was highly variable, changing between each visit at sites where ticks were detected. The likelihood level was consistent at 13 of 15 sites. Based on the kappa statistic, there was substantial agreement between measurements for the presence/absence and the likelihood levels. Our results indicate that both presence/absence and likelihood levels provide more consistent outcomes for tick dragging than tick abundance alone; however, applying the dragging data to the likelihood indicator provides additional information about the potential risk associated with I. scapularis establishment in the area.

Introduction

With the ongoing northward range expansion of the blacklegged tick, Ixodes scapularis, into Ontario, Canada, active tick surveillance has become a vital component of public health efforts in the province (Sider et al. 2012, Ogden et al. 2014a). Ixodes scapularis is the vector for several pathogens of public health significance; Borrelia burgdorferi, the causative agent for Lyme disease, currently poses the highest risk to human and animal health in Ontario (Burgdorfer et al. 1982, Ogden et al. 2014b).

Tick dragging is the most common approach for active tick surveillance. In brief, this involves dragging a white flannel drag cloth over the forest floor and vegetation to collect questing ticks. Public health units can employ this method with relatively limited investment in terms of human resources, time, and supplies. It also has increased specificity when compared with passive surveillance (i.e., tick submissions from the public). However, areas can be declared falsely positive or negative for I. scapularis populations, due to adventitious ticks and low tick densities, respectively (Ogden et al. 2014a).

Concerns exist with the repeatability of tick dragging results (Dobson 2013). This is primarily due to fluctuations in temperature, humidity, and other conditions that impact the tick's activity patterns (Schulze and Jordan 2005). The distribution of ticks at the site level can also be highly variable. Public health professionals have to work within these limitations as most public health units only have resources to sample once per site each season (spring and fall). Risk assessment is, therefore, based on limited site-level data.

The objective of our study was to assess the repeatability of tick dragging results over a short timescale. The short timescale represents the sampling approach employed by public health professionals. Three outcome measures based on tick dragging results were chosen: presence/absence of ticks, tick abundance, and an assessment of the likelihood I. scapularis establishment in an area, using an indicator developed by Clow et al. (2018). The indicator takes into account tick abundance (life stage(s) collected and number) as well as past tick dragging findings.

Methods

Fifteen forested sites were selected in southern and eastern Ontario, based on the following criteria: two sites with established I. scapularis populations, seven sites with variable findings from past surveillance conducted in 2014–2016, and six sites where I. scapularis had not been detected in the previous 3 years (Clow et al. 2017). Each site was visited twice within a 1-month timeframe during the spring or fall of 2017 or spring of 2018. At each site visit, tick dragging was conducted for 3 person-hours by pulling a 1 m² flannel drag cloth across the forest floor and vegetation. The drag cloth was checked every 3 minutes and all I. scapularis were counted and removed. Tick dragging results from each visit were applied to the indicator in Clow et al. (2018) to determine the likelihood of I. scapularis establishment.

The kappa statistic was calculated to determine the level of agreement between presence/absence and likelihood levels between visits for each site. The 95% confidence interval (CI) was determined using 500 bootstrap replications. Analysis was completed using STATA version 14 (2017; STATACorp, College Station, TX).

Results

Ixodes scapularis were detected at 11 sites (Table 1), with I. scapularis collected during both visits at 9 of the 11 sites. However, abundance, in terms of number and life stage(s) collected, was different between each visit where ticks were collected. When the indicator was applied, seven sites were consistently classified as high likelihood of an established I. scapularis population, whereas three were classified as low and three as nonzero. Two sites changed in likelihood level (low to high, medium to low).

Table 1.

The Results of Tick Dragging for Ixodes scapularis at 15 Sites in Southern and Eastern Ontario Visited Twice Within a 1-Month Period Based on the Outcome Measures of Presence/Absence, Abundance, and the Likelihood of I. scapularis Establishment (Clow et al. 2018)

				First visit				Second visit
Site no.	Site location (latitude, longitude)	Previous collection of I. scapularis	Date	Presence of I. scapularis	Abundance	Likelihood of establishment	Date	Presence of I. scapularis	Abundance	Likelihood of establishment
1	42.81°, −80.62°	Yes^a	05-11-2017	Yes	7 adults, 0 nymphs, 0 larvae	High	05-31-2017	Yes	2 adults, 3 nymphs, 18 larvae	High
2	43.09°, −81.82°	No	05-15-2017	No	0	Nonzero	06-05-2017	No	0	Nonzero
3	43.51°, −80.23°	No	05-16-2017	No	0	Nonzero	06-09-2017	No	0	Nonzero
4	43.94°, −80.08°	Yes^a	05-24-2017	No	0	Low	05-31-2017	No	0	Low
5	43.28°, −80.03°	Yes^b	06-14-2017	Yes	30 adults, 69 nymphs, 17 larvae	High	06-28-2017	Yes	21 adults, 90 nymphs, 20 larvae	High
6	43.52°, −79.61°	Yes^b	06-14-2017	Yes	5 adults, 105 nymphs, 23 larvae	High	06-28-2017	Yes	1 adult, 203 nymphs, 51 larvae	High
7	43.83°, −79.59°	No	06-21-2017	Yes	5 nymphs	High	07-12-2017	Yes	15 nymphs, 4 larvae	High
8	44.29°, −76.15°	No	07-04-2017	Yes	1 nymph	Low	07-19-2017	Yes	2 nymphs, 1 larvae	High
9	43.30°, −76.10°	Yes^a	10-27-2017	Yes	2 adults	Medium	11-07-2017	Yes	1 adult	Low
10	44.29°, −76.15°	Yes^a	10-31-2017	No	0	Low	11-03-2017	Yes	1 adult	Low
11	44.99°, −76.54°	Yes^a	05-14-2018	Yes	4 adults	High	05-29-2018	Yes	6 adults, 5 nymphs, 14 larvae	High
12	44.69°, −76.40°	Yes^a	05-16-2018	Yes	49 adults, 4 nymphs, 9 larvae	High	05-29-2018	Yes	20 adults, 13 nymphs, 26 larvae	High
13	44.54°, −76.77°	Yes^a	05-17-2018	Yes	20 adults	High	05-29-2018	Yes	8 adults, 1 nymph, 1 larvae	High
14	44.14°, −79.57°	No^a	05-23-2018	Yes	1 adult	Low	06-11-2018	No	0	Low
15	43.73°, −80.33°	No^a	05-28-2018	No	0	Nonzero	06-05-2018	No	0	Nonzero

Previous findings were variable (Clow et al. 2016, 2017).

Classified as an endemic area (Sider et al. 2012).

The kappa statistic for agreement between presence/absence was 0.700 (95% CI: 0.315–1.000). For agreement between likelihood levels, the kappa statistic was 0.792 (95% CI: 0.519–1.000).

Discussion

Tick dragging was an effective method to detect questing I. scapularis, but the repeatability of tick dragging results was highly dependent on the outcome measure. If the presence or absence of I. scapularis was of interest, then the results at 13 of the 15 sites were the same between both visits. Abundance as an outcome measure was highly variable. The life stage(s) collected, the number of each life stage, and the total number of ticks varied between every visit at each site, except for the four sites where I. scapularis was not detected. In terms of likelihood of tick establishment, 13 of the 15 sites were classified at the same level. Based on the kappa statistic, there was substantial agreement for comparison of both presence/absence and the likelihood levels between two visits, with likelihood level having a slightly higher value (Landis and Koch 1977).

The inconsistencies observed in outcome measures reflect numerous challenges associated with field sampling of ticks and are not unexpected. As previously mentioned, tick activity varies throughout the day due to microclimatic conditions, which can impact the abundance of ticks collected. The seasonal activity patterns of each life stage also contribute to variability in abundance. Although our repeat site visits were within 1 month of each other, this may not have been a sufficiently short time period to capture similar activity of a life stage (Lindsay et al. 1999). For the two sites where there was a change in the presence and absence outcome, either no I. scapularis were detected at the first visit, while one adult was collected at the second visit, or vice versa. It is possible that these were bird-borne adventitious ticks, or early reproducing populations at low density (Ogden et al. 2008). The discrepancy in likelihood levels between sites was observed at two sites that had low tick densities. When the tick density is low, it can be difficult to collect questing I. scapularis, and stochastic fade-out of new populations of I. scapularis may occur (Ogden et al. 2014a).

When considering our findings, applying the indicator to the tick dragging results to assess the likelihood of I. scapularis establishment may be the most suitable outcome to enhance the repeatability of tick dragging results, and its value to public health surveillance. Using abundance as an outcome measure, especially if only a single drag session is planned, raises concerns in the reliability of the data in terms of identifying Lyme disease risk areas and the subsequent use of those results. Although presence/absence was reliable, it is a binary outcome and does not further quantify risk. Quantification of risk may be desirable for public health units with limited resources because they can target preventative measures to the areas of greatest risk.

Moving forward, we recommend continued application and evaluation of the indicator. Our study was preliminary, with a small sample size and limited timeframe. Greater information can be gained on the utility of the indicator to enhance the overall value of tick dragging to public health by applying it to more sites over longer time periods.

Footnotes

Acknowledgments

We thank the Public Health Agency of Canada (PHAC) and the Natural Sciences and Engineering Research Council (NSERC) for financial support. K.C. was supported by an NSERC scholarship and the Blake Graham Fellowship. R.F. was supported by an Andrea Leger Dunbar Summer Research Studentship. We would like to thank Parks Canada, the Ontario Ministry of National Resources and Forestry and conservation authorities in Ontario for providing access to their parks and conservations areas.

Author Disclosure Statement

No competing financial interests exist.

References

Burgdorfer

, Barbour

, Hayes

, Benach

, et al. Lyme disease—A tick-borne spirochetosis?. Science, 1982; 216:1317–1319.

Clow

, Ogden

, Lindsay

, Michel

, et al. The influence of abiotic and biotic factors on the invasion of Ixodes scapularis in Ontario, Canada. Ticks Tick Borne Dis, 2017; 8:554–563.

Clow

, Ogden

, Lindsay

, Russell

, et al. A field-based indicator to determine the likelihood of Ixodes scapularis population establishment at sites in Ontario, Canada. PLoS One, 2018; 13:e0193524.

Dobson

ADM

. Ticks in the wrong boxes: Assessing error in blanket-drag studies due to occasional sampling. Parasit Vectors, 2013; 6:344.

Landis

, Koch

. The measurement of observer agreement for categorical data. Biometrics, 1977; 33:159–174.

Lindsay

, Mathison

, Barker

, McEwen

, et al. Abundance of Ixodes scapularis (Acari: Ixodidae) larvae and nymphs in relation to host density and habitat on Long Point, Ontario. J Med Entomol, 1999; 36:243–254.

Ogden

, Koffi

, Lindsay

. Assessment of a screening test to identify Lyme disease risk. Canada Commun Dis Rep, 2014a; 40:83–87.

Ogden

, Koffi

, Pelcat

, Lindsay

. Environmental risk from Lyme disease in central and eastern Canada: a summary of recent surveillance information. Canada Commun Dis Rep, 2014b; 40:74–82.

Ogden

, Lindsay

, Hanincová

, Barker

, et al. Role of migratory birds in introduction and range expansion of Ixodes scapularis ticks and of Borrelia burgdorferi and Anaplasma phagocytophilum in Canada. Appl Environ Microbiol, 2008; 74:1780–1790.

10.

Schulze

, Jordan

. Influence of meso- and microscale habitat structure on focal distribution of sympatric Ixodes scapularis and Amblyomma americanum (Acari: Ixodidae). J Med Entomol, 2005; 42:285–294.

11.

Sider

, Patel

, Russell

, Jain-Sheehan

, et al. Technical report: Update on Lyme disease prevention and control. Toronto: Public Health Ontario, 2012.