Patterns of Testing for Tick-Borne Diseases and Implications for Surveillance in the Southeastern US | Infectious Diseases | JAMA Network Open | JAMA Network
[Skip to Navigation]
Sign In
Figure.  Seasonal Trends of Testing and Positive Tests for SFGR, Ehrlichiosis, and Lyme Disease in North Carolina, 2017-2020
Seasonal Trends of Testing and Positive Tests for SFGR, Ehrlichiosis, and Lyme Disease in North Carolina, 2017-2020

Tests are aggregated by week of the year, with weekly number of tests given by the height of the bar and test result denoted by the color. Tick season (March-October) is designated by the vertical dashed lines. For Lyme disease, equivocal EIAs are included with positive results. EIA indicates enzyme immunoassay; SFGR, spotted fever group rickettsiosis.

Table 1.  Operational Use of Diagnostic Terminology and Case Classifications for SFGR and Ehrlichiosis Test Results Used Throughout the Study
Operational Use of Diagnostic Terminology and Case Classifications for SFGR and Ehrlichiosis Test Results Used Throughout the Study
Table 2.  Demographics for Individuals Tested and Testing Positive for Tick-Borne Disease Within the UNC Health System, 2017-2020
Demographics for Individuals Tested and Testing Positive for Tick-Borne Disease Within the UNC Health System, 2017-2020
Table 3.  Number of Tests and Paired Testing Completeness, SFGR, and Ehrlichiosis
Number of Tests and Paired Testing Completeness, SFGR, and Ehrlichiosis
Table 4.  Serial Testing Results for Individuals Tested for Lyme Disease, Enzyme Immunoassay (EIA), and Western Blot (WB)
Serial Testing Results for Individuals Tested for Lyme Disease, Enzyme Immunoassay (EIA), and Western Blot (WB)
1.
Rosenberg  R, Lindsey  NP, Fischer  M,  et al.  Vital signs: trends in reported vector-borne disease cases—United States and territories, 2004-2016.   MMWR Morb Mortal Wkly Rep. 2018;67(17):496-501. doi:10.15585/mmwr.mm6717e1PubMedGoogle ScholarCrossref
2.
Olano  JP, Masters  E, Hogrefe  W, Walker  DH.  Human monocytotropic ehrlichiosis, Missouri.   Emerg Infect Dis. 2003;9(12):1579-1586. doi:10.3201/eid0912.020733PubMedGoogle ScholarCrossref
3.
Hinckley  AF, Connally  NP, Meek  JI,  et al.  Lyme disease testing by large commercial laboratories in the United States.   Clin Infect Dis. 2014;59(5):676-681. doi:10.1093/cid/ciu397PubMedGoogle ScholarCrossref
4.
Nelson  CA, Saha  S, Kugeler  KJ,  et al.  Incidence of clinician-diagnosed Lyme disease, United States, 2005-2010.   Emerg Infect Dis. 2015;21(9):1625-1631. doi:10.3201/eid2109.150417PubMedGoogle ScholarCrossref
5.
Monaghan  AJ, Moore  SM, Sampson  KM, Beard  CB, Eisen  RJ.  Climate change influences on the annual onset of Lyme disease in the United States.   Ticks Tick Borne Dis. 2015;6(5):615-622. doi:10.1016/j.ttbdis.2015.05.005PubMedGoogle ScholarCrossref
6.
Caminade  C, McIntyre  KM, Jones  AE.  Impact of recent and future climate change on vector-borne diseases.   Ann N Y Acad Sci. 2019;1436(1):157-173. doi:10.1111/nyas.13950PubMedGoogle ScholarCrossref
7.
Alkishe  A, Raghavan  RK, Peterson  AT.  Likely geographic distributional shifts among medically important tick species and tick-associated diseases under climate change in North America: a review.   Insects. 2021;12(3):225. doi:10.3390/insects12030225PubMedGoogle ScholarCrossref
8.
Sickbert-Bennett  EE, Weber  DJ, Poole  C, MacDonald  PD, Maillard  JM.  Completeness of communicable disease reporting, North Carolina, USA, 1995-1997 and 2000-2006.   Emerg Infect Dis. 2011;17(1):23-29. doi:10.3201/eid1701.100660PubMedGoogle ScholarCrossref
9.
Tick-Borne Disease Working Group.  2020 Report to Congress. U.S. Department of Health and Human Services; 2020.
10.
Biggs  HM, Behravesh  CB, Bradley  KK,  et al.  Diagnosis and management of tick-borne rickettsial diseases: Rocky Mountain spotted fever and other spotted fever group rickettsioses, ehrlichioses, and anaplasmosis—United States.   MMWR Recomm Rep. 2016;65(2):1-44. doi:10.15585/mmwr.rr6502a1PubMedGoogle ScholarCrossref
11.
Centers for Disease Control and Prevention.  National Notifiable Diseases Surveillance System, 2019 Annual Tables of Infectious Disease Data. CDC Division of Health Informatics and Surveillance; 2021.
12.
Straily  A, Stuck  S, Singleton  J,  et al.  Antibody titers reactive with Rickettsia rickettsii in blood donors and implications for surveillance of spotted fever rickettsiosis in the United States.   J Infect Dis. 2020;221(8):1371-1378.PubMedGoogle Scholar
13.
Mead  P, Petersen  J, Hinckley  A.  Updated CDC recommendation for serologic diagnosis of Lyme disease.   MMWR Morb Mortal Wkly Rep. 2019;68(32):703. doi:10.15585/mmwr.mm6832a4PubMedGoogle ScholarCrossref
14.
Boyce  RM, Speight  C, Lin  JT, Farel  CE.  Errors in diagnostic test use and interpretation contribute to the high number of Lyme disease referrals in a low-incidence state.   Open Forum Infect Dis. 2020;7(1):ofaa009. doi:10.1093/ofid/ofaa009PubMedGoogle ScholarCrossref
15.
Lantos  PM, Lipsett  SC, Nigrovic  LE.  False positive Lyme disease IgM immunoblots in children.   J Pediatr. 2016;174:267-269.e1. doi:10.1016/j.jpeds.2016.04.004PubMedGoogle ScholarCrossref
16.
Webber  BJ, Burganowski  RP, Colton  L, Escobar  JD, Pathak  SR, Gambino-Shirley  KJ.  Lyme disease overdiagnosis in a large healthcare system: a population-based, retrospective study.   Clin Microbiol Infect. 2019;25(10):1233-1238. doi:10.1016/j.cmi.2019.02.020PubMedGoogle ScholarCrossref
17.
Eisen  RJ, Paddock  CD.  Tick and tick-borne pathogen surveillance as a public health tool in the United States.   J Med Entomol. 2021;58(4):1490-1502. doi:10.1093/jme/tjaa087PubMedGoogle ScholarCrossref
18.
Boyce  RM, Sanfilippo  AM, Boulos  JM, Cleinmark  M, Schmitz  J, Meshnick  S.  Ehrlichia infections, North Carolina, USA, 2016.   Emerg Infect Dis. 2018;24(11):2087-2090. doi:10.3201/eid2411.180496PubMedGoogle ScholarCrossref
19.
Dahlgren  FS, Holman  RC, Paddock  CD, Callinan  LS, McQuiston  JH.  Fatal Rocky Mountain spotted fever in the United States, 1999-2007.   Am J Trop Med Hyg. 2012;86(4):713-719. doi:10.4269/ajtmh.2012.11-0453PubMedGoogle ScholarCrossref
20.
Kuriakose  K, Pettit  AC, Schmitz  J, Moncayo  A, Bloch  KC.  Assessment of risk factors and outcomes of severe ehrlichiosis infection.   JAMA Netw Open. 2020;3(11):e2025577. doi:10.1001/jamanetworkopen.2020.25577PubMedGoogle ScholarCrossref
21.
North Carolina Division of Public Health. 2020. Accessed August 10, 2021. https://epi.dph.ncdhhs.gov/cd/rmsf/RMSFSurveillanceSummary2019.pdf
22.
Barbarin  AM, Seagle  SW, Creede  S.  Notes from the field: four cases of Lyme disease at an outdoor wilderness camp - North Carolina, 2017 and 2019.   MMWR Morb Mortal Wkly Rep. 2020;69(4):114-115. doi:10.15585/mmwr.mm6904a5PubMedGoogle ScholarCrossref
23.
Lantos  PM, Nigrovic  LE, Auwaerter  PG,  et al.  Geographic expansion of Lyme disease in the southeastern United States, 2000-2014.   Open Forum Infect Dis. 2015;2(4):ofv143. doi:10.1093/ofid/ofv143PubMedGoogle ScholarCrossref
24.
Vaughn  MF, Delisle  J, Johnson  J,  et al.  Seroepidemiologic study of human infections with spotted fever group Rickettsiae in North Carolina.   J Clin Microbiol. 2014;52(11):3960-3966. doi:10.1128/JCM.01733-14PubMedGoogle ScholarCrossref
25.
Moncayo  A, Heitman  K, Williams  C,  et al.  Changes to Public Health Reporting and National Notification for Spotted Fever Rickettsiosis (including Rocky Mountain Spotted Fever). Council of State and Territorial Epidemiologists; 2019.
26.
Engel  J, Hull  H, Lynch  C, Parrish  R.  Public Health Reporting and National Notification for Ehrlichiosis and Anaplasmosis. Council of State and Territorial Epidemiologists; 2009.
27.
UNC PCORnet DataMart Table 1. North Carolina Translational and Clinical Science Institute. Published July 1, 2019. Accessed April 11, 2022. https://tracs.unc.edu/docs/cdwh/PCORnetTable1_July2019_Public.pdf
28.
Drexler  NA, Dahlgren  FS, Heitman  KN, Massung  RF, Paddock  CD, Behravesh  CB.  National surveillance of spotted fever group rickettsioses in the United States, 2008-2012.   Am J Trop Med Hyg. 2016;94(1):26-34. doi:10.4269/ajtmh.15-0472PubMedGoogle ScholarCrossref
29.
Nichols Heitman  K, Dahlgren  FS, Drexler  NA, Massung  RF, Behravesh  CB.  Increasing incidence of ehrlichiosis in the United States: a summary of national surveillance of Ehrlichia chaffeensis and Ehrlichia ewingii infections in the United States, 2008-2012.   Am J Trop Med Hyg. 2016;94(1):52-60. doi:10.4269/ajtmh.15-0540PubMedGoogle ScholarCrossref
30.
Wallace  JW, Nicholson  WL, Perniciaro  JL,  et al.  Incident tick-borne infections in a cohort of North Carolina outdoor workers.   Vector Borne Zoonotic Dis. 2016;16(5):302-308. doi:10.1089/vbz.2015.1887PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Original Investigation
    Infectious Diseases
    May 16, 2022

    Patterns of Testing for Tick-Borne Diseases and Implications for Surveillance in the Southeastern US

    Author Affiliations
    • 1Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina
    • 2Institute for Global Health and Infectious Diseases, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina
    • 3Department of Pathology and Laboratory Medicine, McLendon Clinical Laboratories, UNC Health, Chapel Hill, North Carolina
    • 4Division of Public Health, Communicable Disease Branch, Raleigh, North Carolina
    • 5School of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina
    • 6Carolina Population Center, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina
    JAMA Netw Open. 2022;5(5):e2212334. doi:10.1001/jamanetworkopen.2022.12334
    Key Points

    Question  What is the tick-borne disease diagnostic testing landscape in central North Carolina?

    Findings  In this cross-sectional study of 11 367 individuals, there was a higher incidence of spotted fever group rickettsiosis (SFGR) (4.7%) and ehrlichiosis (7.1%) relative to Lyme disease (0.7%), which had more than twice the testing volume. Most of those tested for SFGR and ehrlichiosis lacked convalescent testing, and the test volume for SFGR was twice that for ehrlichiosis despite a similar clinical spectrum.

    Meaning  These findings suggest that ehrlichiosis remains underrecognized and Lyme disease test ordering is overrepresented given its low incidence.

    Abstract

    Importance  Tick-borne diseases (TBD), including spotted fever group rickettsiosis (SFGR), ehrlichiosis, and, increasingly, Lyme disease, represent a substantial public health concern throughout much of the southeastern United States. Yet, there is uncertainty about the epidemiology of these diseases because of pitfalls in existing diagnostic test methods.

    Objective  To examine patterns of diagnostic testing and incidence of TBD in a large, academic health care system.

    Design, Setting, and Participants  This cross-sectional study included diagnostic test results for TBD at UNC Health, a large academic health care system with inpatient and outpatient facilities, from January 1, 2017, to November 30, 2020. Participants included all individuals seeking routine care at UNC Health facilities who had testing for SFGR, ehrlichiosis, or Lyme disease performed during the study period.

    Main Outcomes and Measures  Rates of test positivity, testing completeness, and incidence of TBD.

    Results  During the 4-year study period, 11 367 individuals (6633 [58.4%] female; 10 793 [95%] non-Hispanic individuals and 8850 [77.9%] White individuals; median [IQR] age, 53 [37-66] years) were tested for TBD. Among the 20 528 diagnostic tests performed, 47 laboratory-confirmed, incident cases of SFGR, 27 cases of ehrlichiosis, and 76 cases of Lyme were confirmed, representing incidence rates of 4.7%, 7.1%, and 0.7%, respectively. However, 3984 of SFGR tests (79.3%) and 3606 of Ehrlichia tests (74.3%) lacked a paired convalescent sample. Of 20 528 tests, there were 11 977 tests (58.3%) for Lyme disease from 10 208 individuals, 5448 tests (26.5%) for SFGR from 4520 individuals, and 3103 tests (15.1%) for ehrlichiosis from 2507 individuals. Most striking, testing for ehrlichiosis was performed in only 55% of patients in whom SFGR was ordered, suggesting that ehrlichiosis remains underrecognized. An estimated 187 incident cases of SFGR and 309 of ehrlichiosis were potentially unidentified because of incomplete testing.

    Conclusions and Relevance  In this cross-sectional study, most of the patients suspected of having TBD did not have testing performed in accordance with established guidelines, which substantially limits understanding of TBD epidemiology. Furthermore, the data revealed a large discrepancy between the local burden of disease and the testing performed. These findings underscore the need to pursue more robust, active surveillance strategies to estimate the burden of TBD and distribution of causative pathogens.

    Introduction

    From 2004 to 2016, the US Centers for Disease Control and Prevention (CDC) received nearly 500 000 reports of tick-borne diseases (TBD), comprising more than three-quarters of all vector-borne disease reports in the continental United States.1 Notably, reports of TBD doubled during this same time period, with consistent annual increases in Lyme disease, spotted fever group rickettsiosis (SFGR), and ehrlichiosis. These estimates, based on reports from state health departments, likely capture only a fraction of total infections and clinical illness.2-4 These trends will likely be exacerbated by global climate change, which may affect not only the geographic distribution of ticks, but also the tick and pathogen life cycles.5-7

    While TBD represents an emerging public health crisis, timely and accurate surveillance is severely constrained by both the passive nature of reporting8 and limitations of existing diagnostic methods.9 Indirect immunofluorescence assays (IFA) are widely employed laboratory tests for TBD which rely on detection of host antibodies to the infecting pathogen.10 Because antibodies are often not present at high levels during the first week of illness—the period when most patients initially seek care—a negative test cannot rule out disease. To confirm the diagnosis of SFGR or ehrlichiosis, a second, or convalescent, test is required 2 to 10 weeks after the initial, or acute, test. Unfortunately, few patients undergo both acute and convalescent testing with less than 3% of SFGR cases reported to the CDC being classified as confirmed.11 In the absence of paired acute and convalescent results, interpretation of a single titer is challenging, especially in areas where background rates of seropositivity are high.12

    In contrast, testing for Lyme disease employs a 2-tier algorithm on a single serum sample, consisting of an enzyme immunoassay (EIA) or IFA, followed by a Western blot if the EIA is positive or equivocal.13 While this approach does not require multiple samples collected at different time points, clinical practice, especially in historically lower-incidence states, is fraught with ordering errors (ie, Western blot without a preceding EIA) and misinterpretation of results.14 Furthermore, incorporation of the much less specific IgM-based assay, intended to increase sensitivity early in the disease course, often increases false positivity rates.15,16

    Diagnosis and prevention of TBD depends greatly on an accurate understanding of local TBD epidemiology, much of which is derived from routine clinical practice, thus improving clinician adherence to diagnostic testing guidelines remains a key priority.17 Without new testing methods or rigorous protocols to ensure high levels of testing completeness, estimates of disease incidence will remain susceptible to both under- and overrepresentation, the magnitude of which is ill-defined. Perhaps, more importantly, misunderstanding of the predictive value of test results, especially a negative acute result, may contribute to delays in diagnosis and treatment, which is one of the main risk factors for severe disease and death.18-20

    To date, there has been a relative paucity of research examining the patterns of diagnostic testing for tick-borne disease, particularly for SFGR and ehrlichiosis. Moreover, previous studies have largely focused on data derived from routine surveillance, which does not include individuals who test negative. Including individuals who test negative provides important information about the underlying population assumed to be at risk and can yield complementary information regarding disease incidence and routine clinical practice. Therefore, the overarching objective of this study was to explore patterns in diagnostic testing for TBD in a large academic health system with the goals of (1) describing the demographic characteristics of patients tested for TBD, (2) determining rates of adherence to testing guidelines, and (3) identifying potential gaps in clinician ordering behaviors that may be improved with systems-based interventions (eg, panels, reflex testing).

    Methods

    This cross-sectional study was approved by the institutional review board at the University of North Carolina at Chapel Hill. Written informed consent or a waiver of Health Insurance Portability and Accountability Act authorization was not required because the data were classified as a limited data set under the Code of Federal Regulations 45, Part 164.514 (e). This study follows the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

    Study Site

    North Carolina experiences some of the highest rates of SFGR and ehrlichiosis in the United States, often accounting for more than of 10% and 5% of the national totals reported to the CDC, respectively.11 In 2019, the incidence rate of SFGR was 6.6 cases per 100 000 residents with most cases being geographically clustered in the central and eastern parts of the state.21 Ehrlichiosis is less frequently reported, but is an underrecognized cause of tick-borne illness, despite evidence that infection rates may be similar to those of SFGR.18 Cases of Lyme disease are increasing in the state and are more concentrated in northwestern parts of the state in proximity to the Appalachian Mountains.22,23

    We obtained diagnostic test results from the Carolina Data Warehouse for Health, a central repository containing clinical, research, and administrative data sourced from the UNC Health, the largest academic health system in North Carolina, comprising 12 hospitals and 350 outpatient clinics located across the state. In 2018, UNC Health reported approximately 3.5 million clinical visits, including approximately 500 000 emergency department visits.

    Data Sources and Management

    We requested the results of diagnostic testing performed by UNC’s McLendon Clinical Laboratories for SFGR, ehrlichiosis, and Lyme disease from January 1, 2017, to November 30, 2020. Specifically, we abstracted results of IFA testing for immunoglobulin G (IgG) antibodies against R. rickettsia and E. chaffeensis, recognizing that cross-reactivity between species is well-known.24 Immunoglobulin M (IgM) results for Rickettsia or Ehrlichia, when ordered, were excluded from the analysis given that these tests are not included in the current testing guidelines. For Lyme disease, we abstracted the results of EIA, Western blots and polymerase chain reaction (PCR). Western blot results were included in the analysis regardless of the preceding EIA result and even if the EIA was not ordered or not available.

    For each result, we documented the date of the test and age, sex, race, and ethnicity of the patient. Sex, race, and ethnicity data originated from medical records. Race and ethnicity were self-identified by the patients at intake and were considered to compare the TBD-tested population with the general patient population at UNC Health. Race categories included American Indian or Alaskan Native, Asian, Black or African American, White, and other, with other race representing individuals who did not identify with a provided category. Ethnicity categories included Hispanic or Latino, not Hispanic or Latino, and unknown.

    Statistical Analysis

    Data were cleaned to remove duplicate entries. All analysis was performed using R version 4.0.2 (R Project for Statistical Computing). We compared test positivity for each disease across age and sex using Wilcoxon rank sum tests and Pearson χ2 tests, respectively, with a significance level of P < .05. χ2 tests were performed for differences in test volume by season, defined as high (March to October) and low season (November to February) based on historical tick activity,21 with a null assumption of equal testing throughout the year.

    We established operational case classifications and laboratory criteria (Table 1) modeled after the CDC’s 2020 Surveillance Case Definitions used for SFGR and ehrlichiosis.25,26 We excluded clinical criteria from these definitions due to lack of clinical information. Tests for SFGR and ehrlichiosis were considered paired if the acute and convalescent samples were performed on the same individual between 14 and 70 days of each other.24 Paired tests were subsequently defined to represent an incident case if either (1) the result of the acute test was less than or equal to 1:64 and the result of the convalescent test was greater than or equal to 1:64 for ehrlichiosis or greater than or equal to 1:128 for SFGR or (2) the results of both acute and convalescent test were greater than or equal to 1:64, and the convalescent result represented at least a 4-fold increase in titer compared with the acute result. Paired tests were defined as a prevalent case if the convalescent titer was less than 4-fold that of the acute and (1) both tests were greater than or equal to 1:64 for ehrlichiosis or (2) both tests were greater than or equal to 1:64 with at least 1 test greater than or equal to 1:128 for SFGR. In addition, we defined an individual as a probable case if at least 1 dilution was greater than or equal to 1:128 or a suspected case if at least 1 test was positive at 1:64. A patient was considered as having a Lyme disease case if (1) their EIA test result was indeterminate or positive and the Western blot satisfied criteria for either IgG or IgM positivity, (2) they lacked an antibody test performed and their Western blot was positive for IgG, or (3) they had a positive PCR result from cerebrospinal fluid (CSF).

    Results

    During the study, a total of 20 528 diagnostic tests for TBD were performed on 11 367 unique individuals, including 11 977 tests for Lyme disease from 10 208 individuals, 5448 tests for SFGR from 4520 individuals, and 3103 tests for ehrlichiosis from 2507 individuals (Table 2).

    The median (IQR) age of individuals tested was 53 (37-66) years. Children younger than 18 years represented 671 (5.9%) of those tested, much lower than the UNC Health systemwide under 20 patient population in 2017 (195 971 of 1 027 521 [19%]).27 Women represented 6633 (58.4%) of individuals tested, similar to the demographic profile of patients systemwide (587 768 of 1 011 507 [58.1%]), while 10 288 of 11 367 (90.5%) identified as non-Hispanic individuals (833 171 of 1 011 649 [82.9%] systemwide) and 8850 of 11 367 (77.9%) as White individuals (628 131 of 1 011 649 [62.1%] systemwide). Racial and ethnic breakdowns did not vary substantially between those tested for each disease.

    A total of 4144 of 11 367 individuals (36.5%) were tested for more than 1 TBD with combinations Lyme, SFGR, and ehrlichiosis (1724 [15.2%]) and Lyme and SFGR (1686 [14.8%]), being most frequent. Testing for SFGR and ehrlichiosis was significantly higher during the high season (March-October) with 4785 (87.9%; P < .001) and 2683 (86.5%; P < .001) of tests occurring during this period, respectively (Figure). In contrast, testing for Lyme was more equally distributed throughout the year. When stratified by the year testing was performed (eTable 1 in the Supplement), we observed a large decline in total tests performed in 2020, even when accounting for missing December 2020 data.

    The overall test positivity rate (TPR), defined as the proportion of all tests that were positive at greater than or equal to 1:128, was 27.1% for SFGR and 38.8% for ehrlichiosis, reflecting positive results of greater than or equal to 1:64. The TPR for Lyme disease, defined as the proportion of positive EIA, PCR, and Western blot results, was 4.2%. TPR increased substantially during the analysis period for SFGR and for ehrlichiosis, which more than doubled from 16.7% to 53.2%. In contrast, TPR for Lyme disease remained relatively stable. Assuming clinical criteria were met, among individuals with paired tests, there were 25 (4.7%) incident cases of SFGR and 27 (7.1%) incident cases of ehrlichiosis (Table 3). A total of 76 (0.7%) individuals satisfied the 2-tier testing criteria for Lyme disease, among which 11 met both IgM and IgG criteria, 12 met only IgG criteria, and 53 met only IgM criteria (Table 4; eTable 2 in the Supplement). None of the 224 Lyme disease PCR tests conducted on CSF were positive.

    Additionally, testing identified 287 prevalent SFGR cases, 626 probable SFGR cases, and 288 suspected SFGR cases and 239 prevalent ehrlichiosis cases, 384 probable ehrlichiosis cases, and 233 suspected ehrlichiosis cases. In total, 28.8% of individuals tested for SFGR and 35.2% of individuals tested for ehrlichiosis met the laboratory case definition of a suspected or probable case.

    TPR was similar between seasonal periods for Lyme disease (4.1% during the high season and 4.8% during the low season), but it was greater during the high vs low season for SFGR (44.6% vs 39.2%; P = .01) and during the low vs high season for ehrlichiosis (37.9% vs 45.7%; P < .01). Compared with those without a positive test, patients testing positive for Lyme disease were younger with a median (IQR) age of 53 (30-62) years vs 50 (38-66) years (P = .04), whereas those testing positive for ehrlichiosis trended older with a median (IQR) age of 57 (41-69) years vs 53 (37-57) years (P = .03). Age was not significantly associated with SFGR positivity.

    Of the 4577 individuals who were tested for SFGR, only 536 (11.8%) had paired acute and convalescent testing performed. A higher rate of paired testing was observed with ehrlichiosis (378 of 2507 [15.1%]) (Table 3). Individuals with paired testing tended to be older than those without paired testing for both SFGR (median [IQR] age, 57 [41-68] years vs 52 [35-66] years; P < .001) and ehrlichiosis (median [IQR] age, 56 [41-69] years vs 53 [37-67] years; P = .007). Additionally, individuals with an acute test result of greater than or equal to 1:64 were more than twice as likely to have a convalescent test ordered. Adherence to standard testing protocol was higher for Lyme disease, with only 295 individuals (2.9%) undergoing a Western blot without a preceding indeterminate or positive antibody test. Of these, 12 (4.1%) had positive IgM results and 1 (0.3%) had a positive IgG result (Table 4).

    If individuals who only received an acute test experienced the same incidence rate as those with paired testing, we estimate that approximately 187 incident cases of SFGR and 151 incident cases of ehrlichiosis were not identified as a result of incomplete testing. Furthermore, because SFGR and ehrlichiosis have overlapping clinical disease spectra—and thus individuals presenting with typical symptoms (eg, fever, headache, myalgia) or suspected tick exposure should be tested for both diseases—we estimate that there were 2256 missing ehrlichiosis tests. Again, assuming similar incidence rates among individuals in whom SFGR was considered, but testing for ehrlichiosis was not performed, there would be an additional 158 missing incident cases, raising the estimate of total missed ehrlichiosis cases to 309.

    Discussion

    Despite the increasing awareness of TBD as a growing public health threat, our study suggests that critical gaps remain in the routine surveillance systems underlying our knowledge of the spatial and clinical epidemiology of these diseases. The low rate of testing completeness, with only 1 in 10 individuals tested for SFGR having both acute and convalescent testing performed, severely limits the ability to distinguish incident infection from prior exposure. The paucity of confirmed diagnoses, both positive and negative, confounds prevention and control efforts, and introduces substantial misclassification bias into epidemiological and clinical research. Of particular concern, the apparent failure to consider ehrlichiosis as a potential cause of acute illness in approximately half of patients tested for SFGR raises serious doubts about our understanding of the burden of this disease. Improving the quality of routine surveillance data, through targeted education, systems-based interventions, and new testing modalities that do not require multiple visits, remains a key priority.

    There are multiple potential explanations for the low level of testing completeness, including: (1) the resolution of clinical symptoms, either because of treatment or self-limited disease, which might reduce a patient’s motivation to return for a convalescent test, especially if incurring out-of-pocket costs; (2) the confirmation of alternative causes of illness over the interval between acute and convalescent testing; and (3) a general lack of frontline clinician knowledge regarding testing algorithms. One key to overcoming these issues is the development and more widespread adoption of acute stage diagnostics, such as PCR. We note that the US Food and Drug Administration recently approved a pan-Rickettsia real-time PCR assay and laboratory developed assays are available for Ehrlichia, but these tests are often inaccessible to frontline clinicians and even when available remain underused for reasons that merit further investigation.

    While rates of testing completeness were relatively low—although still higher than similar data at the state or national level—our results identified notable trends in both testing and infection. First, the overall cohort in this study was relatively older and more female than previous national surveillance studies,28,29 although these characteristics are generally similar to the larger population seeking care at UNC Health. In the study sample, sex was not significantly associated with any disease positivity; however, we did observe that those positive for Lyme disease and ehrlichiosis were older than those testing negative. Whether this is because of differences in exposure risk, age-related clinical manifestations, or care-seeking behaviors is unclear. It is plausible that given the high levels of mild and even asymptomatic seroconversion previously seen in relatively young outdoor workers,24,30 we may be observing differential care-seeking patterns.

    Another crucial finding is the high rate of testing for Lyme disease in contrast to the relative neglect of testing for ehrlichiosis, even among those tested for SFGR. The low rate of testing completeness and lack of consideration of ehrlichiosis likely resulted in a large number of unidentified cases, which would greatly impact surveillance data. While North Carolina reported a total of 1704 cases of SFGR and 372 cases of ehrlichiosis statewide during the study, our estimates suggest there might be an additional 187 SFGR and 309 ehrlichiosis cases from the UNC Health system alone. However, our estimates assume that individuals without paired testing experienced the same incidence rate as those with paired testing. Although, these groups differ significantly in ways that may affect the likelihood of a positive convalescent test and may ultimately overestimate the number of missed cases. In contrast, the high rate of Lyme disease testing may artificially inflate incidence rates in the area because of false-positive results, especially with the IgM assay, which represented most (69.7%) of the positive Western blot results.

    Our study has several strengths including the large sample size abstracted from both outpatient and inpatient settings distributed throughout a relatively large geographic area. Furthermore, unlike previous analyses of notifiable disease reports, our study includes information about individuals who tested negative, which is information not normally available in these reports. Lastly, with high rates of SFGR and ehrlichiosis, and expanding Lyme in the western regions, North Carolina is a compelling setting to monitor trends in patterns of TBD testing.

    Limitations

    This study had limitations. First, there was a lack of clinical data, such as the nature, onset, and duration of symptoms, which would have allowed us to more fully apply case definitions. Particularly, we were unable to identify cases in which initial testing was done within the first 7 days of illness onset, which is important because antibodies to many TBD may not yet be detectable. We were also unable to determine the frequency of alternative causes of febrile illness may have been present. For example, in some circumstances, individuals may have had a reactive SFGR titer, perhaps caused by prior exposure to R. amblyommatis and asymptomatic seroconversion, but ultimately have clinical disease caused by other bacterial or viral pathogens. Previous studies have demonstrated a high background rate among asymptomatic patients in the southeastern United States.12

    Second, the absence of longitudinal data limited our ability to assess treatment decisions and, most importantly, clinical outcomes. Third, our study population may not be wholly representative of all patients seeking care for TBD in the area. Some patients may have been empirically treated without formal diagnostic testing based on the presence of typical symptoms and exposure history alone. Our sample may have been biased if individuals with standard disease presentation were more likely to have been empirically treated and less likely to be tested.

    Conclusions

    These findings suggest that most patients suspected of having TBD did not have testing performed in accordance with established guidelines, which substantially limits our understanding of TBD epidemiology. Furthermore, there appears to be a large discrepancy between the local burden of disease and the testing that is performed. These findings underscore the need for better diagnostics and active surveillance programs, particularly for SFGR and ehrlichiosis, to more accurately identify the spatial distribution and risk factors for infection.

    Back to top
    Article Information

    Accepted for Publication: March 30, 2022.

    Published: May 16, 2022. doi:10.1001/jamanetworkopen.2022.12334

    Correction: This article was corrected on June 13, 2022, to fix errors in the title and byline.

    Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2022 Brown Marusiak A et al. JAMA Network Open.

    Corresponding Author: Ross M. Boyce, MD, MSc, Division of Infectious Diseases, University of North Carolina at Chapel Hill, 123 W Franklin St, Ste 2151, Chapel Hill, NC 27516 (roboyce@med.unc.edu).

    Author Contributions: Ms Brown Marusiak and Dr Hollingsworth had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Ms Brown Marusiak and Dr Hollingsworth contributed equally to the manuscript.

    Concept and design: Brown Marusiak, Hollingsworth, Alejo, Arahirwa, Boyce.

    Acquisition, analysis, or interpretation of data: All authors.

    Drafting of the manuscript: Brown Marusiak, Hollingsworth, Abernathy, Alejo, Arahirwa, Mansour, Boyce.

    Critical revision of the manuscript for important intellectual content: Brown Marusiak, Hollingsworth, Arahirwa, Giandomenico, Schmitz, Williams, Barbarin, Boyce.

    Statistical analysis: Brown Marusiak, Hollingsworth, Boyce.

    Obtained funding: Alejo, Boyce.

    Administrative, technical, or material support: Arahirwa, Mansour, Giandomenico, Schmitz, Barbarin.

    Supervision: Boyce.

    Conflict of Interest Disclosures: None reported.

    Funding/Support: Dr Boyce was supported by the Creativity Hub Award from the University of North Carolina Office of the Vice Chancellor for Research and a Caregivers at Carolina award 2015213 from the Doris Duke Charitable Foundation. The study was also supported by grant UL1TR002489 from the North Carolina Translational and Clinical Sciences.

    Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    Additional Contributions: We wish to thank the regulatory team at the University of North Carolina Institute of Global Health and Infectious Diseases and the data management team at the Carolina Data Warehouse for their efforts.

    References
    1.
    Rosenberg  R, Lindsey  NP, Fischer  M,  et al.  Vital signs: trends in reported vector-borne disease cases—United States and territories, 2004-2016.   MMWR Morb Mortal Wkly Rep. 2018;67(17):496-501. doi:10.15585/mmwr.mm6717e1PubMedGoogle ScholarCrossref
    2.
    Olano  JP, Masters  E, Hogrefe  W, Walker  DH.  Human monocytotropic ehrlichiosis, Missouri.   Emerg Infect Dis. 2003;9(12):1579-1586. doi:10.3201/eid0912.020733PubMedGoogle ScholarCrossref
    3.
    Hinckley  AF, Connally  NP, Meek  JI,  et al.  Lyme disease testing by large commercial laboratories in the United States.   Clin Infect Dis. 2014;59(5):676-681. doi:10.1093/cid/ciu397PubMedGoogle ScholarCrossref
    4.
    Nelson  CA, Saha  S, Kugeler  KJ,  et al.  Incidence of clinician-diagnosed Lyme disease, United States, 2005-2010.   Emerg Infect Dis. 2015;21(9):1625-1631. doi:10.3201/eid2109.150417PubMedGoogle ScholarCrossref
    5.
    Monaghan  AJ, Moore  SM, Sampson  KM, Beard  CB, Eisen  RJ.  Climate change influences on the annual onset of Lyme disease in the United States.   Ticks Tick Borne Dis. 2015;6(5):615-622. doi:10.1016/j.ttbdis.2015.05.005PubMedGoogle ScholarCrossref
    6.
    Caminade  C, McIntyre  KM, Jones  AE.  Impact of recent and future climate change on vector-borne diseases.   Ann N Y Acad Sci. 2019;1436(1):157-173. doi:10.1111/nyas.13950PubMedGoogle ScholarCrossref
    7.
    Alkishe  A, Raghavan  RK, Peterson  AT.  Likely geographic distributional shifts among medically important tick species and tick-associated diseases under climate change in North America: a review.   Insects. 2021;12(3):225. doi:10.3390/insects12030225PubMedGoogle ScholarCrossref
    8.
    Sickbert-Bennett  EE, Weber  DJ, Poole  C, MacDonald  PD, Maillard  JM.  Completeness of communicable disease reporting, North Carolina, USA, 1995-1997 and 2000-2006.   Emerg Infect Dis. 2011;17(1):23-29. doi:10.3201/eid1701.100660PubMedGoogle ScholarCrossref
    9.
    Tick-Borne Disease Working Group.  2020 Report to Congress. U.S. Department of Health and Human Services; 2020.
    10.
    Biggs  HM, Behravesh  CB, Bradley  KK,  et al.  Diagnosis and management of tick-borne rickettsial diseases: Rocky Mountain spotted fever and other spotted fever group rickettsioses, ehrlichioses, and anaplasmosis—United States.   MMWR Recomm Rep. 2016;65(2):1-44. doi:10.15585/mmwr.rr6502a1PubMedGoogle ScholarCrossref
    11.
    Centers for Disease Control and Prevention.  National Notifiable Diseases Surveillance System, 2019 Annual Tables of Infectious Disease Data. CDC Division of Health Informatics and Surveillance; 2021.
    12.
    Straily  A, Stuck  S, Singleton  J,  et al.  Antibody titers reactive with Rickettsia rickettsii in blood donors and implications for surveillance of spotted fever rickettsiosis in the United States.   J Infect Dis. 2020;221(8):1371-1378.PubMedGoogle Scholar
    13.
    Mead  P, Petersen  J, Hinckley  A.  Updated CDC recommendation for serologic diagnosis of Lyme disease.   MMWR Morb Mortal Wkly Rep. 2019;68(32):703. doi:10.15585/mmwr.mm6832a4PubMedGoogle ScholarCrossref
    14.
    Boyce  RM, Speight  C, Lin  JT, Farel  CE.  Errors in diagnostic test use and interpretation contribute to the high number of Lyme disease referrals in a low-incidence state.   Open Forum Infect Dis. 2020;7(1):ofaa009. doi:10.1093/ofid/ofaa009PubMedGoogle ScholarCrossref
    15.
    Lantos  PM, Lipsett  SC, Nigrovic  LE.  False positive Lyme disease IgM immunoblots in children.   J Pediatr. 2016;174:267-269.e1. doi:10.1016/j.jpeds.2016.04.004PubMedGoogle ScholarCrossref
    16.
    Webber  BJ, Burganowski  RP, Colton  L, Escobar  JD, Pathak  SR, Gambino-Shirley  KJ.  Lyme disease overdiagnosis in a large healthcare system: a population-based, retrospective study.   Clin Microbiol Infect. 2019;25(10):1233-1238. doi:10.1016/j.cmi.2019.02.020PubMedGoogle ScholarCrossref
    17.
    Eisen  RJ, Paddock  CD.  Tick and tick-borne pathogen surveillance as a public health tool in the United States.   J Med Entomol. 2021;58(4):1490-1502. doi:10.1093/jme/tjaa087PubMedGoogle ScholarCrossref
    18.
    Boyce  RM, Sanfilippo  AM, Boulos  JM, Cleinmark  M, Schmitz  J, Meshnick  S.  Ehrlichia infections, North Carolina, USA, 2016.   Emerg Infect Dis. 2018;24(11):2087-2090. doi:10.3201/eid2411.180496PubMedGoogle ScholarCrossref
    19.
    Dahlgren  FS, Holman  RC, Paddock  CD, Callinan  LS, McQuiston  JH.  Fatal Rocky Mountain spotted fever in the United States, 1999-2007.   Am J Trop Med Hyg. 2012;86(4):713-719. doi:10.4269/ajtmh.2012.11-0453PubMedGoogle ScholarCrossref
    20.
    Kuriakose  K, Pettit  AC, Schmitz  J, Moncayo  A, Bloch  KC.  Assessment of risk factors and outcomes of severe ehrlichiosis infection.   JAMA Netw Open. 2020;3(11):e2025577. doi:10.1001/jamanetworkopen.2020.25577PubMedGoogle ScholarCrossref
    21.
    North Carolina Division of Public Health. 2020. Accessed August 10, 2021. https://epi.dph.ncdhhs.gov/cd/rmsf/RMSFSurveillanceSummary2019.pdf
    22.
    Barbarin  AM, Seagle  SW, Creede  S.  Notes from the field: four cases of Lyme disease at an outdoor wilderness camp - North Carolina, 2017 and 2019.   MMWR Morb Mortal Wkly Rep. 2020;69(4):114-115. doi:10.15585/mmwr.mm6904a5PubMedGoogle ScholarCrossref
    23.
    Lantos  PM, Nigrovic  LE, Auwaerter  PG,  et al.  Geographic expansion of Lyme disease in the southeastern United States, 2000-2014.   Open Forum Infect Dis. 2015;2(4):ofv143. doi:10.1093/ofid/ofv143PubMedGoogle ScholarCrossref
    24.
    Vaughn  MF, Delisle  J, Johnson  J,  et al.  Seroepidemiologic study of human infections with spotted fever group Rickettsiae in North Carolina.   J Clin Microbiol. 2014;52(11):3960-3966. doi:10.1128/JCM.01733-14PubMedGoogle ScholarCrossref
    25.
    Moncayo  A, Heitman  K, Williams  C,  et al.  Changes to Public Health Reporting and National Notification for Spotted Fever Rickettsiosis (including Rocky Mountain Spotted Fever). Council of State and Territorial Epidemiologists; 2019.
    26.
    Engel  J, Hull  H, Lynch  C, Parrish  R.  Public Health Reporting and National Notification for Ehrlichiosis and Anaplasmosis. Council of State and Territorial Epidemiologists; 2009.
    27.
    UNC PCORnet DataMart Table 1. North Carolina Translational and Clinical Science Institute. Published July 1, 2019. Accessed April 11, 2022. https://tracs.unc.edu/docs/cdwh/PCORnetTable1_July2019_Public.pdf
    28.
    Drexler  NA, Dahlgren  FS, Heitman  KN, Massung  RF, Paddock  CD, Behravesh  CB.  National surveillance of spotted fever group rickettsioses in the United States, 2008-2012.   Am J Trop Med Hyg. 2016;94(1):26-34. doi:10.4269/ajtmh.15-0472PubMedGoogle ScholarCrossref
    29.
    Nichols Heitman  K, Dahlgren  FS, Drexler  NA, Massung  RF, Behravesh  CB.  Increasing incidence of ehrlichiosis in the United States: a summary of national surveillance of Ehrlichia chaffeensis and Ehrlichia ewingii infections in the United States, 2008-2012.   Am J Trop Med Hyg. 2016;94(1):52-60. doi:10.4269/ajtmh.15-0540PubMedGoogle ScholarCrossref
    30.
    Wallace  JW, Nicholson  WL, Perniciaro  JL,  et al.  Incident tick-borne infections in a cohort of North Carolina outdoor workers.   Vector Borne Zoonotic Dis. 2016;16(5):302-308. doi:10.1089/vbz.2015.1887PubMedGoogle ScholarCrossref
    ×