Introduction

Scabies is a common skin condition, caused by the ectoparasite Sarcoptes scabiei var. hominis. The Global Burden of Disease study estimates more than 200 million people are affected at any one time [1]. The direct effects of scabies are estimated to account for 0.21% of DALYs caused by all conditions [2], but the broader impact, incorporating complications of bacterial skin infection, invasive bacterial disease and auto-immune kidney and heart disease is likely to be substantially greater [3].

The World Health Organization has recently designated scabies as a neglected tropical disease (NTD) for large-scale disease control action [4]. Whilst there has been a very high prevalence reported in some population-based studies, there is a paucity of disease burden data from most regions [5]. There are numerous opportunities to integrate population-based surveys for scabies with mapping and surveillance for other NTDs and health programs [6, 7], but few successful examples thus far [8]. Therefore, standardized methods are needed for further epidemiological mapping, and to monitor the effectiveness of control interventions.

The need to improve and standardize the diagnosis of scabies was identified as a priority at the inaugural meeting of the International Alliance for the Control of Scabies (IACS) in Atlanta, 2012 [3]. Microscopy of skin samples has limited sensitivity and utility in field settings, and there are no laboratory tests available. A systematic review of diagnostic methods used in therapeutic trials for scabies found wide variation with no uniformity [9], similar to results of a previous systematic review [10]. Given the limitations in available evidence for diagnostic methods, we aimed to develop consensus criteria for the diagnosis of human scabies in a range of epidemiological and research environments, using a formal consensus process.

Methods

The Delphi method is an established, iterative, multi-stage process for developing consensus using at least two rounds of anonymous surveys [11]. The Delphi method has been used for establishment of diagnostic and/or treatment frameworks where there is a lack of scientific evidence to make recommendations, including for several skin conditions [12–14]. After identification of the research issue and formation of the panel of participants, initial survey rounds generate a range of ideas and opinions through open-ended questions. In subsequent rounds, participants are provided a summary of the responses of the whole panel, and then given the opportunity to revise their own response. The advantages of the method include the ability to involve a large group of international participants, relatively low cost, anonymity, and reducing the likelihood of dominance by certain participants. Disadvantages include arbitrary cut-offs for consensus, potential for influence of moderators, and length of time required [11, 15].

Results

Thirty-four panel participants completed at least one survey. Participants were predominantly highly experienced dermatologists with practice across a range of settings and representing all major geographic regions. The characteristics of the Round 4 participants are shown in Table 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Characteristics of Delphi panel members responding to Round 4 survey.

https://doi.org/10.1371/journal.pntd.0006549.t001

Responses were received from 30 participants for both Round 1 and Round 2. Participants expressed that criteria should only be used for diagnosis of common scabies, and not crusted or other atypical clinical variants. Based on responses, a set of proposed criteria were developed that could be used in a variety of settings, by incorporating three levels of diagnostic certainty–Confirmed Scabies, Clinical Scabies and Suspected Scabies.

Twenty-eight responses were received for Round 3. Agreement on the structure and contents of the draft criteria was high, with 23 respondents (82%) supporting the adoption of the draft criteria (Table 2). Extensive commentary was also received about content and terminology, leading to further revision of the draft criteria.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Round 3 survey responses.

https://doi.org/10.1371/journal.pntd.0006549.t002

Twenty-eight responses were received for Round 4. There was a high level of agreement on all levels of criteria and individual subcategories (Table 3). Twenty-seven participants (96.4%) supported the adoption of the proposed criteria. Further commentary from panelists resulted in minor editing of terminology to improve clarity.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Round 4 survey responses.

https://doi.org/10.1371/journal.pntd.0006549.t003

Discussion

We established consensus on criteria for the diagnosis of scabies through a four-round Delphi process (Box 1). The final product is a single set of criteria with three levels and eight subcategories, representing a spectrum of diagnostic certainty. The agreement for each level and subcategory was very high, with at least 89% agreement for all subcategories, and 96% support for the adoption of the criteria.

The advantage of the levels and categories is that the criteria may be applied in a range of situations and settings. For example, a clinical trial of new therapies may choose to only enrol participants with Confirmed Scabies (level A) as this would be the most specific and least sensitive level. Alternately, for field mapping, Clinical and Suspected Scabies (levels B and C) may be used, acknowledging these would be less specific but more sensitive. Where examination of genital areas is not feasible or appropriate, due to a lack of visual privacy or other dignity considerations, sub-category B2 could be omitted. Investigators using the criteria are advised to list the distribution of diagnoses by sub-category, which will further allow comparison between survey findings.

Whilst these criteria are not intended to replace the clinical acumen and assessment of clinicians treating individual patients, we anticipate there may be clinical applications, particularly as a reference in settings where non-expert health workers are required to make a diagnosis. Clinical practice requires consideration of other factors outside of these criteria, such as history and progression over time, and the benefits and risks of treating or not treating individuals. Therefore, the threshold for diagnosis and treatment may be higher or lower in some cases.

Although this Delphi process resulted in high agreement for the diagnostic criteria, there are several limitations. In the absence of an appropriate reference standard test, or adequate data on the accuracy of clinical criteria, this process is reliant on expert opinion, which may not correlate with diagnostic accuracy. Therefore, an important next step is the validation of the criteria in a range of settings, including different geographic regions, areas of differing prevalence, and where other important differential diagnoses may be prevalent.

These criteria should be viewed as a starting point, allowing consistency and comparison in collected data. The criteria are likely to require revision, for example, if accurate laboratory or point-of-care tests for scabies infection become available. Although the criteria allow for diagnosis in mapping scenarios using a quick skin examination, the inclusion of history features, particularly relating to contact history, is not ideal for rapid assessment. Therefore, further research into the accuracy and utility of more simplified approaches to assessment are needed. It may be that more simplified assessment (for example, a limited skin examination of the limbs) could be used in field surveys, particularly if results can be correlated back to a comprehensive skin examination including assessment of itch and contact history.

Application of the criteria, summarized in Box 1, requires reference to the evidence-based explanatory notes and definitions that are being prepared in parallel. These notes will clearly define and explain each feature and subcategory of the criteria and will include recommended techniques for examination, microscopy and visualization, as well as details of important differential diagnoses. To enable use of the criteria for epidemiological mapping, a training methodology will be required for non-expert health workers, similar to that developed for trachoma graders [17].

Strong consensus has been established for criteria to diagnose scabies in a variety of settings. These criteria will facilitate diagnosis and comparison of findings across studies. The 2018 criteria require validation in a range of epidemiological settings.

Supporting information

Acknowledgments

We acknowledge the contribution of the Delphi panel including Erin Amerson, Charlotte Bernigaud, Franck Boralevi, Raul Cabrera, Olivier Chosidow, Roberto Estrada, Guadalupe Estrada-Chavez, Hermann Feldmeier, Gioconda Gaudiano, Roderick Hay, Jorg Heukelbach, Janet Hickman, Norihisa Ishii, Richard Keller, Arum Krismi, Stefania Lanza, Kieron Leslie, Harvey Lui, Fatimata Ly, Mohamed Mahdi, Antoine Mahe, Guiseppe Micali, Michele Murdoch, Pragya Nair, Scott Norton, David Pariser, Wingfield Rehmus, Dev Tilakaratne, Stephen Walker, Shelley Walton, Karolyn Wanat, Margot Whitfeld, Junko Yoshizumi and Rie Yotsu. We also thank Melanie Thompson, Hywel Williams and Ciara Baker for their contributions and advice.