Background
Scabies is caused by the ectoparasitic mite Sarcoptes scabiei, which lives and reproduces in the epidermis. Scabies has a low mortality risk but considerable morbidity, particularly when associated with further secondary infection [Reference Chandler and Fuller1]. Clinical manifestations, which generally develop 4–6 weeks after primary exposure, include pruritis and characteristic skin lesions such as nodules and vesicles [Reference Arlian and Morgan2]. Infection is most common in cavities such as between the fingers, groin areas, armpits, and genitals, as well as the wrists and breasts [Reference Cassell3].
Though most clinical presentations are classified as ‘ordinary’ or ‘classical’ scabies, which is associated with relatively low mite burden, individuals may develop crusted scabies, characterized by hyperproliferation of the scabies mite. Crusted scabies varies in appearance but may present with deep skin fissures and hyperkeratosis, commonly on the hands and feet [Reference Cassell3, Reference Niode4]. A single infected individual may host thousands of mites, and thus index cases of crusted scabies are commonly associated with scabies outbreaks in institutional settings.
Key epidemiological studies have been undertaken to explore the disproportionately high prevalence of scabies outbreaks in aged-care facilities, where attack rates can be as high as 50% [Reference Hewitt, Nalabanda and Cassell5]. A retrospective review identified misdiagnosis of scabies as a contributing factor to 43% of outbreaks, with 83% of the index cases presenting with crusted scabies [Reference Mounsey6]. Compared to ordinary scabies, crusted scabies lesions can occur in atypical locations such as the scalp and toenails. Delayed detection and suboptimal management of scabies outbreaks may be attributable to crusted and/or atypical presentations of scabies, including in areas usually covered by clothing [Reference Cassell3]. Delayed diagnosis may also stem from the uncertainty of residential care workers in identifying scabies, who refer their residents to doctors for diagnosis before commencing mediative strategies [Reference Mounsey6, Reference Phipps7]. Finally, scabies outbreaks in aged care facilities are exacerbated by both the physical and social vulnerability of the residents [Reference Hewitt, Nalabanda and Cassell5]. Residents are more likely to be exposed to the disease, more likely to acquire it, and less able to voice their distress. Residents with co-morbidities such as dementia pose a particular risk of chronic scabies infection and challenges with treatment [Reference Utsumi8].
In this work, we explore the knowledge, attitudes, and experiences of aged-care workers towards scabies. The overall aim of this research is to improve prevention and management strategies for scabies in aged care facilities by developing informed and targeted educational resources.
Methods
Ethics
This study was approved by the Human Research Ethics Committees of the University of the Sunshine Coast (A13513) and Uniting Care Queensland (15413).
Survey design and deployment
Data was collected via an anonymous, 28-question cross-sectional survey during 2013 and 2014. Surveys were paper-based and took approximately 10–15 min to complete. Questions were developed based on an initial literature review analysing institutional scabies outbreaks from 1984 to 2013 [Reference Mounsey6]. Questions were reviewed for clarity and readability by content experts, students, and laypeople and refined after feedback. The survey was divided into four sections. Section 1 included demographics, role in aged care, experience in sector, education level, and experiences with scabies. Section 2 included multiple choice and true/false questions on scabies knowledge, including questions relating to scabies transmission, diagnosis, control, and treatment (Supplementary Table S1). Section 3 included numeric rating scale questions regarding personal attitudes towards scabies, as well as two questions regarding preferences for resources. A section for extra comments and feedback was also included. In Section 4, participants were asked to identify scabies cases from a series of images. Two representative images of skin with crusted scabies and three images of skin with ordinary scabies were included. Three additional images depicted impetigo, eczema, and dry skin (Supplementary Material S1). Participants were asked to tick images that they believed to be scabies cases.
Aged-care facilities in the Sunshine Coast region of Southeast Queensland, Australia, were invited to participate via phone and/or email to management. Eight high-care facilities agreed to participate, and a total of 680 surveys were deployed. Surveys were distributed with the recommendation to leave in a staff-only communal area (such as a break room) for staff to complete at their leisure. Surveys were anonymous and could not be linked back to individuals or facilities. Reply-paid envelopes were included for staff to individually mail back surveys to the investigators. One centre elected to collect in a central box and return in batches.
Analysis
Survey data was collated and analysed using SPSS v.28 (IBM). For multiple choice and Likert scale questions, no-response answers were coded as missing values. Participants who did not tick any images in Section 4 were coded as missing values. Missing values were excluded from all percentage and frequency calculations.
Knowledge analysis
A knowledge construct score was developed from 11 knowledge-based multiple-choice questions (Table 2, Supplementary Material S1). The construct score was equal to the number of correct answers selected by each participant. A missing value for any of the 11 construct-score questions resulted in a construct score not being calculated for that participant. The Kruskal–Wallis test was used to determine whether the distribution of construct scores varied within levels of different demographic variables, such as age, role in aged care, time worked in the sector, and education status. The Mann–Whitney U test was used to determine whether the construct score distributions varied within binomial explanatory variables, such as whether the respondent had prior exposure to or experience with scabies. Each individual component of the construct score was then run using the same analyses to determine whether any individual questions, such as those relating to treatment or identification, were influenced by demographic factors.
Respondents were presented with eight images showing dermatological conditions, including examples of crusted and ordinary scabies, and asked to select which images depicted scabies. Responses were coded as selected (1), not selected (0), or missing value (999) when an unclear selection had been made. Respondents who did not identify any of the images were allocated missing values for all images (i.e. were deemed as not participating in this section of the survey). The proportion of participants who identified each image correctly and incorrectly was determined. Where this section of the survey was completed, the differences in proportions of correct answers between images were evaluated using Cochran’s Q test. Post hoc analyses were evaluated using Dunn’s test. McNemar’s test was used to determine if there was a statistically significant difference in the proportion of people that believed scabies could affect the scalp and the proportion that could correctly identify a scabies presentation of the scalp when provided with an image.
Attitudes analysis
Respondents were provided with numeric rating scale questions assessing how problematic the respondents believed scabies outbreaks were and how comfortable they would feel managing them. For each question, Wilcoxon’s sign rank test was used to compare participant’s median responses to a hypothesized neutral (3) response. For each question, the Mann–Whitney U test was used to compare the distribution of answers between binomial variables relating to scabies exposure. These included whether the respondent had experienced ordinary scabies, crusted scabies, or a scabies outbreak. This was used to determine whether a participant’s prior experiences with scabies influenced their perceptions of it.
Results
Demographics
Of the 680 surveys deployed, 128 responses were received, for a response rate of 18.8% (Table 1). Most respondents were over 40 years of age (72.3%), were female (84.2%), and were either patient care assistants (56.3%) or registered or enrolled nurses (15.9%). Most (71.9%) reported frequently working with patients requiring physical assistance. The level of experience within the aged-care sector was evenly distributed, and 62.9% of respondents had obtained a diploma/certificate qualification or higher.
a Missing (‘no response’) values were excluded from percentage and frequency calculations.
Prior experience and observations of scabies
Episodes of scabies were common in aged care facilities. Respondents had asked whether they had ever encountered scabies while working in aged care, with 41% encountering individuals with ordinary scabies, 13% with crusted scabies, and 22% scabies outbreaks on one or more occasions (Figure 1a). Of those answering yes to any of the above (n = 99), (32.3%) had observed scabies within the 2 years prior to the survey, while 21.2% observed it >2 years but within the last 10 years. 8.1% had observed scabies within the 6 months prior to the survey, and 38.4% of those who had observed scabies did not provide time-specific information. Reports of occupational transmission were lower, with only 6% reporting to have acquired scabies through their workplace, but a higher number (25%) reported knowing colleagues that had caught scabies through their occupation (Figure 1b).
Knowledge
Overall scabies construct score
Construct scores were calculated for participants where all knowledge questions were completed (n = 59). Respondents had a median construct score of 9 out of a possible 11, or 82% (Figure 2). The Kruskal–Wallis test found no statistically significant difference between construct scores and different roles in aged care (K = 3.086, n = 57, p = 0.544), length of time worked in aged care (K = 6.821, n = 59, p = 0.146) or education status (K = 1.631, n = 59, p = 0.803). However, there was a small but statistically significant difference found in the construct score distributions of different ages of the respondents (K = 8.142, n = 59, p = 0.017). Dunn’s post-hoc analysis revealed that the 40+ age group demonstrated statistically higher levels of knowledge than the 18–25 age group (p = 0.05, Figure 2b). Separate analysis of the 40+ age group confirmed there was no significant difference between the median construct score and the number of years worked in aged care within the (K = 0.317, n = 39, p = 0 0.989). The Mann–Whitney U test found that prior experience with scabies had no impact on the distribution of the construct scores, whether that be from participants having caught scabies themselves (U = 35.5, n = 59, p = 0.357), having colleagues that had caught scabies (U = 269.0, n = 57, p = 0.544), or having family members that had caught scabies (U = 26.0, n = 58, p = 0.879). For individual questions comprising the construct score (Table 2), participants showed high awareness of topics such as scabies transmission and prevention of transmission (64.8% and 85.2% respectively), decontamination practices (83.6%) and symptoms (91.4%). Uncertainty was evident around the most appropriate and relevant diagnostic techniques (18.5% correct) and incubation periods (44.5% correct). A relatively high proportion (34.4%) of respondents incorrectly thought that poor hygiene caused scabies.
a For these proportions, non-completion of a question was assigned as an incorrect response.
b As not all participants answered all questions, this proportion was calculated on the numbers of participants that responded to this question.
c This section was only attempted by 99 participants. Where the n total responses is less than 99, the response was ambiguous (such as a question mark instead of a tick/circle).
Clinical identification of scabies
In this section of the survey, respondents were asked to identify which clinical images shown depicted scabies. Only 99 of the 128 respondents completed this part of the survey (i.e. for values classed as ‘missing’, no part of the survey page was marked). The results of Cochran’s Q test showed that there was a statistically significant difference in the proportions of correctly identified conditions (Q = 226.132, n = 99, p < 0.001) (Figure 3). This meant that respondents were able to identify some conditions as being scabies/not scabies better than others. Post hoc analysis determined no statistically significant differences in the proportions of correctly identified conditions between images depicting dry skin vs. crusted scabies of toes (p = 0.078), dry skin vs. impetigo/boil (p = 0.240), crusted scabies of toes vs. impetigo/boil (p = 0.769), ordinary scabies on elderly skin vs. ordinary scabies of axilla with rash (p = 0.078), ordinary scabies on elderly skin vs. infected eczema (p = 0.187), infected eczema vs. crusted scabies on scalp (p = 0.056). Statistically significant differences (p < 0.05) were observed for every other pairwise comparison of proportions (Figure 3). Participants demonstrated a strong ability to correctly identify images that did not depict scabies (impetigo/boil, 84% correct, dry skin, 93% correct). The exception was an example of eczema, which only 55% correctly identified as a non-scabies case. The identification of images that depicted ordinary scabies was variable depending on clinical presentation. Participants could identify the ‘classical’ presentation of scabies (papular lesions, burrows visible on the wrist) with high confidence (90% correct). Other presentations of ordinary were identified with less confidence, including dry excoriation with generalized rash in the axillary region (68% correct) and ordinary scabies on elderly skin (55% correct). The ability of respondents to correctly identify images depicting crusted scabies was low. Crusted scabies of the toes with superimposed bacterial infection was correctly identified by only 17.2%. Although 75.8% of respondents correctly believed that scabies could affect the scalp or the face in the knowledge question, only 31.6% of participants correctly identified the crusted scabies infection of the scalp in the image. Using McNemar’s test, the difference between these proportions was determined to be statistically significant (n = 96, p < 0.01). This indicates a discordance between theoretical knowledge of crusted scabies and practical ability in scabies management.
Attitudes and perceptions towards scabies
Most respondents agreed that scabies outbreaks were a problem in aged care facilities (W = 2075.5, n = 110, p = 0.003), that scabies outbreaks caused disruption to facilities (W = 3,271.0, n = 108, p < 0.001), and felt uncomfortable at the prospect of catching scabies (W = 3,376.0, n = 110, p < 0.001) (Table 3). Although most participants expressed discomfort at the idea of catching scabies (median 4.00), they felt neutral interacting with scabies-infected residents (median 3.00) (Table 3). The distribution of all rating scale responses demonstrated that participants who had been exposed to ordinary scabies, crusted scabies, or scabies outbreaks had no significant difference in attitudes to those that had not (p > 0.05). We asked respondents which resources they would find useful in learning more about scabies and its control. In-person workshops and training were the most selected resource (n = 73, 57%), followed by online resources (n = 69, 53.9%) and the provision of formal written guidelines or standard operating protocols (n = 67, 52.3%). Online training modules were the least popular resource option (n = 31, 27.9%).
Discussion
Worldwide, scabies prevention and treatment within aged care facilities is inadequate, and outbreaks remain a leading cause of morbidity within these settings [Reference Morrison9]. When strategizing interventions for the improvement of scabies management, it is important to explore the driving factors of outbreaks. It has been previously demonstrated that misdiagnosis or delayed diagnosis of index cases of scabies underlies most outbreaks [Reference Mounsey6]. Under resourcing of workers and workload constraints may also be a driving factor in scabies outbreaks. The relative influence of each exacerbating factor in this multifaceted issue remains understudied. This study examined 128 mail-in surveys completed by aged care facility workers, examining the experience, knowledge, and attitudes of workers in relation to scabies prevention and treatment.
To investigate the theoretical knowledge of workers, a construct score was developed using 11 knowledge-based multiple-choice questions from the survey. While construct scores demonstrated a reasonably high level of knowledge, score distributions were not significantly different between experience – both in regard to years worked in the sector and previous exposure to scabies patients and/or outbreaks. This was somewhat unexpected and could suggest a lack of learning within the facilities, with new residential care workers demonstrating similar levels of knowledge compared to more experienced workers. Construct score distributions were however affected by participant age, implying more experience and knowledge on scabies occurring throughout life. Other demographic details, such as the role within the facility and education status, did not influence knowledge construct scores. It could be speculated that overall changes in community scabies exposure and declining general awareness/knowledge in recent years could explain why knowledge was associated with age but not specific aged-care experience.
To our knowledge, this represents the first survey of scabies knowledge, attitudes, and experiences of practitioners in aged-care facilities, despite the relatively common occurrence and substantial health burden imposed by scabies in this setting. Compared to practitioner surveys in other settings internationally, our participants displayed similar or higher levels of general scabies knowledge, although this may have been inflated by limitations of our survey design. A survey of physicians in Saudi Arabia (n = 216) showed inadequate levels of knowledge regarding scabies diagnosis and management, with an overall median knowledge score of 67.5% [Reference Alsaidan10]. Similarly, a Belgian study identified knowledge scores of 59% amongst general practitioners, with dermatologists demonstrating higher scores, as expected (79%) [Reference Lapeere11]. A mixed-methods survey in Guinea-Bissau, consisting of community members and healthcare workers, showed high levels of general awareness but poor diagnostic capability (42% positive identification of scabies lesions) and misconceptions around scabies aetiology and transmission [Reference Lopes12].
Although our respondents demonstrated a good theoretical knowledge of scabies, their ability to identify scabies cases in a practical scenario was inconsistent. When provided with eight images of dermatological skin conditions, participants were asked to identify which images represented scabies infections. The proportion of correct answers ranged from 17% (crusted scabies) to 93% (non-scabies, dry skin). Participants demonstrated a higher level of specificity than they did sensitivity – non-scabies cases were generally associated with a higher proportion of correct identification than scabies cases. The non-scabies presentation with the lowest diagnostic confidence (55%) depicted an eczema flare. This is notable because eczema (atopic dermatitis) is a frequent differential diagnosis for crusted scabies, and the initial incorrect application of corticosteroids is a key contributor to the development and exacerbation of crusted scabies [Reference Mounsey6, Reference Frazier and Bhardwaj13]. These survey results highlight the difficulty in differentiating atopic dermatitis and scabies and should be an area of focus for future education. One caveat to our interpretation of diagnostic specificity is that the survey design meant that a designation of ‘non-scabies’ by the respondent was presumed (by the absence of tick or circle) unless otherwise specified. This meant that uncertain respondents may have been more likely to identify a condition as ‘non-scabies’ than ‘scabies’. There was also a lower response rate for this component of the survey, with only 99/128 participants completing this element of the survey – this could indicate a lack of confidence or clarity with the task. Future survey design should have more explicit selection options.
The image depicting pathognomonic signs of scabies on the wrist was associated with a high proportion of correct answers (90%), confirming that respondents had familiarity with scabies within specific contexts. However, other images that also showed ordinary scabies infections had lower proportions of correct answers (55% and 68%). Interestingly, one image showed ordinary scabies affecting elderly skin with numerous scratch marks, which could be a common clinical manifestation in an aged care facility, noting that elderly patients may present with atypical signs of scabies [Reference Cassell3]. However, this case was largely misidentified. Previous literature has noted that such signs of scabies may be seen as attributable to generalized dry skin or ‘senile itching’, which is not uncommon in this setting [Reference Chung14]. In addition to the reduced capacity of residents to voice discomfort [Reference Hewitt, Nalabanda and Cassell5], it is likely that cases such as these facilitate scabies outbreaks. As well as allowing transmission, missed scabies diagnoses are associated with poor clinical outcomes for the individual, especially if crusted scabies develops from iatrogenic corticosteroid use.
Crusted scabies cases were associated with the lowest proportions of correct identification by respondents. The image depicting infected crusted scabies of the feet and toenails had only 17% correct identification, a concerning finding given that infections from crusted scabies can become life-threatening [Reference May15]. The second image of crusted scabies impacting the scalp was also not readily identified (32% correct), with differential diagnoses commonly including psoriasis or seborrheic dermatitis [Reference Skayem16]. A persisting misconception of scabies is that areas of the scalp and face are rarely affected, which was supported by 33% of our respondents. Indeed, some current guidelines for topical treatment of scabies only recommend treating the skin from the neck down [17, 18]. While this may be the case for most presentations of ordinary scabies, crusted scabies in atypical locations commonly present in the elderly and immunocompromised index cases [Reference Wilson, Philpott and Breer19].
Delayed diagnosis is a common feature of scabies outbreaks in residential facilities. In a UK study, Phipps and colleagues [Reference Phipps7] reported a median time of 6 weeks between initial symptom onset and outbreak notification. During this time, scabies outbreaks continue to spread, assisted by the characteristic delay in the appearance of symptoms post-exposure in a primary infestation of scabies (usually 4–6 weeks) [Reference Arlian and Morgan2]. The lower proportion of correct responses to this survey question suggests limited awareness of this delayed onset, which may contribute to spread and impact decision-making about who receives treatment when scabies is identified [Reference Morrison9]. A similar lack of knowledge regarding incubation periods has been reported in other practitioner surveys [Reference Alsaidan10].
There was also uncertainty amongst our respondents about appropriate diagnostic methods. While most participants correctly understood that a negative skin scraping does not preclude a diagnosis of scabies, they still believed that skin scrapings were likely to be the most sensitive method for detecting scabies. In practice, the diagnostic sensitivity of skin scrapings and microscopy is low (<50%) [3], and clinical presentation and history is currently considered the most sensitive method for diagnosis despite its lack of specificity [Reference Engelman20]. The development of more sensitive, specific, and non-invasive diagnostic methods is a priority area for scabies control [Reference Micali21], and debate exists around diagnostic methods in residential care settings [Reference Petit22, Reference Walker, Middleton and Cassell23].
Scabies, and especially crusted scabies, is an often-stigmatized condition. This was reflected in our survey, with 40% of respondents believing that the statement ‘poor hygiene causes people to develop scabies’ was true. This was further demonstrated by responses to the Likert-style question ‘How would you feel if you caught scabies?’ Respondents had a median response of 4 on a scale of 1 to 5, which demonstrated that they felt uncomfortable with the idea of contracting scabies. The interpretation here could be limited to the way respondents interpreted the question (e.g., physical discomfort vs. emotional distress/stigma). This potential stigma aligns with findings reported from other studies [Reference Phipps7, Reference Lopes12, Reference Widaty24] and likely contributes to underreporting at both the individual and facility level, a consideration when planning interventions.
Although rich insights have been obtained through this work, our survey design was not without its limitations. Participants were not supervised during the completion of the survey, meaning that some of the responses may not represent individuals but rather have been discussed amongst participants. This could have led to an overinflation of knowledge construct scores and proportion of correct responses to individual questions. Similarly, the voluntary nature of the survey may have led to an over exaggeration of scabies knowledge, as workers with higher levels of knowledge may have been more likely to undertake the survey. Conversely, the above noted stigma surrounding scabies may have led to underreporting in particular questions, despite the anonymity of the survey. As with many cross-sectional surveys, questions were routinely skipped by respondents, and some surveys were returned with missing pages, especially for the final section on clinical identification. The number of missed values limits our interpretation of some results. Specifically, knowledge construct scores may be overinflated as they were selected for participants giving complete responses, thus likely having higher confidence and knowledge about scabies. However, similar trends were shown when non-respondents were included in the calculated proportions of individual questions. These issues could have partially been alleviated by using an electronic survey design, but the paper format was selected for increased visibility, engagement, and ease of completion.
Conclusions
Although there was a suitable level of theoretical knowledge amongst aged care workers, particularly around ordinary scabies, workers demonstrated minimal ability to identify atypical presentations of scabies and particularly crusted scabies. This limited knowledge is likely a key factor in the development of outbreaks from misdiagnosed or undiagnosed index cases of crusted scabies. The dissemination of these results to facilities will now form the basis for the co-design of educational resources. This highlights potential for before and after surveys to test effectiveness for such resources. A comprehensive control program that focusses on staff education about the diagnosis and treatment of scabies may significantly lower the incidence of disease and help prevent future outbreaks. Control guidelines should be relevant for managing scabies in elderly patients and readily accessible to target workers.
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1017/S0950268824001377.
Data availability statement
All relevant data are presented in the manuscript and supplementary material. Images used in the survey are not published due to copyright restrictions but can be disseminated for educational purposes upon request to the corresponding author.
Author contributions
Conceptualization: F.O., C.P., K.M.; Funding acquisition: F.O., C.P., K.M.; Investigation: F.O., C.P., I.L., S.C., K.M.; Methodology: F.O., C.P., I.L., S.C., K.M.; Supervision: F.O., K.M.; Writing – review & editing: F.O., C.P., I.L., S.C., K.M.; Data curation: I.L., S.C., K.M.; Formal analysis: I.L., S.C.; Software: I.L.; Visualization: I.L., K.M.; Writing – original draft: I.L.; Project administration: S.C., K.M.; Validation: K.M.
Funding statement
This research was supported by a Research Fellows seed grant from the Sunshine Coast Regional Council (KM, CP, FO).
Competing interest
The authors declare none.