The rates of poststreptococcal diseases in Aboriginal Australians are among the highest in the world.1,2 The reservoir of group A streptococcal (GAS) infection in Aboriginal communities is found in high rates of pyoderma, which is frequently caused by secondary infection of scabies lesions.3, 4 Primary prevention of streptococcal and poststreptococcal diseases in GAS-endemic areas should include a focus on treatment and prevention of streptococcal skin disease and underlying scabies infestation.5 Although short term success has been reported in some Aboriginal communities with community scabies treatment programs,6 there are no published reports of medium or long term sustainability of scabies control in this population.
A successful model for community scabies control, using 5% permethrin cream, has been documented.4 This model avoided the demonstrated failure of programs in which only those with visible scabies were treated7 by treating all residents, regardless of whether or not they were infested. After one Aboriginal community approached us for assistance in dealing with its scabies problem, we chose to adapt the model of Taplin et al.4 to suit local resources.
Setting, subjects and the intervention program. The Minjilang community is located on an island just off the coast of the Northern Territory in tropical northern Australia. The population varies between 200 and 250 inhabitants and has frequent contact with residents of nearby islands and mainland Aboriginal communities. We began in October, 1994, by examining the population for scabies and pyoderma, following which all members of the community older than 2 months of age were offered treatment with 5% permethrin cream. Verbal instructions were given about the proper use of the scabicide, emphasizing the need to apply the cream from head to toe in children and from neck to toe in adults, and that the cream should be washed off after 8 to 12 h. Treatment was not directly observed so the exact proportion of the population who undertook treatment is unknown. However, the staff at the school cooperated by running a treatment program on 1 day, and the Health Workers were given lists of those families with the highest infestation rates, so that they could encourage these families to treat themselves. Children were defined as those age 15 years and younger, and adults were those age 16 years and older. Children with palmar or plantar pustules, indicative of heavy infestation,8 were retreated with 5% permethrin cream after 1 week. The diagnosis of scabies was made clinically, using accepted criteria, including papules, burrows and typical body distribution of lesions.9 Environmental measures such as clothes-washing and cleaning and airing of bedding were suggested, but only infrequently implemented.
Follow-up visits were made at 1, 4, 5, 10, 14, 18 and 25 months after the initial visit. At these visits the population was reexamined, concentrating on the children and including as many adults as could be easily located. At each visit 75 to 95% of the children present in the community at the time were seen. Any new cases of scabies were offered retreatment with 5% permethrin cream, as were their household contacts. At each visit (including the first) people with multiple purulent or crusted pyoderma lesions, whether or not scabies lesions were present, were also offered treatment with intramuscular benzathine penicillin G according to Northern Territory standard treatment protocols. The exact number of people treated with benzathine penicillin G is unknown, but this strategy was not a prominent part of the study. Between follow-up visits, the local health staff did not administer benzathine penicillin G to anyone in the community for treatment of pyoderma. Some visitors to the community, but none of the residents, were offered treatment for scabies between visits.
In addition to documenting rates of pyoderma, we also documented the severity of pyoderma affecting those with residual lesions. We developed a grading scale of pyoderma severity (the "sore score") which gave a numeric score to each person with sores. The score was calculated by adding values for sore types present anywhere on the body (purulent, 3 points; crusted, 2 points; flat/dry, 1 point; to a maximum of 6 points per person) to values for sore numbers and extent (3 points for >20 lesions, 2 points for 5 to 20 lesions and 1 point for <5 lesions, calculated for each of the following regions: upper limbs; legs/buttocks; trunk; head/neck; to a maximum of 12 points per person). The maximum possible sore score per person was 18.
Screening for scabies and pyoderma was combined with a regular school-based health screening program and was carried out by a team of one or two medical practitioners, two to four Aboriginal Health Workers and, at some visits, one to two medical students. To ensure reliability of results, a physician (JRC or CC) was present at each visit and performed or supervised the scabies and pyoderma screening and the skin and throat swabbing and the plating of swabs.
This program was instituted at the request of the community council and health staff, and results were fed back, in a culturally appropriate format, to the council and to the community as a whole. The swabbing procedures were approved by the local Institutional Ethics Committee and informed consent was obtained from participants (or parents, in the case of minors).
Microbiologic methods. At each visit swabs were taken of up to three pyoderma lesions per affected person. At all except for the second visit, people screened were also given the option of having a throat swab taken. There were no cases of symptomatic or clinical pharyngitis when the throat swabs were taken. At the initial visit all swabs were immediately placed into a transport medium (Todd-Hewitt broth with added colistin and nalidixic acid), frozen at −20°C and transported frozen via airplane to the laboratory located in Darwin, 200 km away. At the laboratory the swabs were kept frozen at −20°C for up to 3 days, until they were thawed and an aliquot was taken for culturing on 5% horse blood agar plates with added colistin and nalidixic acid. The plates were incubated at 35°C for 24 to 48 hours, and GAS colonies were identified using standard grouping sera.
The pyoderma culture results from the first visit gave a lower yield of GAS than we had previously experienced with similar lesions. Therefore at all subsequent visits the swabs were plated, immediately after being taken, onto 5% horse blood agar plates with added colistin and nalidixic acid. These plates were streaked within 2 h and kept refrigerated at 4°C until they reached the laboratory for incubation (within 24 h of the swab being taken). With this technique the yield of GAS became much higher. Unfortunately the different methodologies used at the first and subsequent visits precluded an assessment of the impact of the program on rates of GAS isolation.
Statistical analysis. Because the population sampled at each visit was not completely homogeneous, most comparisons were performed by treating the data as unpaired. Prevalence rates of scabies and pyoderma and the upper limb and lower limb distributions of sores at follow-up visits were compared with rates at the first visit using chi square analysis with Yates continuity correction, except where an expected count was <5, in which case Fisher's exact test was used. Because the adult population sampled at each visit was very heterogeneous and because pyoderma rates in the adults became negligible during the course of the program, most of the data for pyoderma severity were analyzed for children only. Mean sore scores at follow-up visits were compared with those at the first visit using the Wilcoxon rank sum test, because these data were nonnormally distributed. A substantial subset of children seen at the first visit was also seen at follow-up visits, and for those children the differences in sore scores between Visit 1 and follow-up visits were approximately normally distributed. These data were therefore analyzed by the paired t test. For the sake of thoroughness they were also analyzed by the Wilcoxon rank sign test; the resulting P values were almost identical with those obtained by the parametric test. To assess whether a trend in skin sore severity could be determined over time, the sore scores at each visit were classified as none (score 0), mild (score 1 to 4) and severe (score >4). These data were then fitted to a log-linear model using the Generalized Linear Interactive Modelling (GLIM4) package.10 All other statistical analyses were performed using Epi-Info Version 6.03 or SPIDA (Statistical Package for Interactive Data Analysis).
Scabies and pyoderma. The program had a significant impact on the prevalences of scabies and pyoderma in children and in the population as a whole (Fig. 1), with significant reductions in the prevalences of both of these conditions sustained during the last 16 months of the program. There was also a substantial reservoir of scabies and pyoderma among the adult population before the program started, with 25% of adults having scabies (95% confidence interval, 15 to 38%) and 29% (18 to 41%) having pyoderma. One month after the initial intervention, the scabies rate among adults was 6% (1 to 21%) (chi square, 3.90; P = 0.05) and the pyoderma rate was 9% (2 to 25%) (chi square, 3.50; P = 0.06). By 18 months there were no cases of scabies found in adults (chi square, 8.85; P = 0.003) and 3% (0 to 15%) had pyoderma (chi square, 7.95; P = 0.005). After the initial treatment program there was only one case of endemic scabies identified in the community: a young child with severe infected scabies who, initially, did not adhere to recommended regimens of retreatment with scabicide. At all visits after the 10-month visit, this child did not have scabies or pyoderma. All other cases of scabies at follow-up visits could be clearly traced to direct contact with residents of other communities.
The only reports of adverse reactions to the scabicide came from two adults who experienced a burning sensation within 4 h of application, which in each case was relieved by washing with water. There were no reports of adverse reactions in children.
Severity of pyoderma. In addition to pyoderma becoming less common during the program, the residual pyoderma lesions became less severe than at the start of the program. Table 1 shows that at all but one of the follow-up visits, the mean sore score for children with residual sores was significantly lower than at the initial visit. When the subset of children who were seen at the first visit and at follow-up visits was analyzed separately, treating the data from these children as paired, the reduction in mean sore scores compared to Visit 1 ranged from 1.8 at Visit 3 (t = 2.334, P = 0.025) to 4.3 at Visit 5 (t = 6.046, P < 0.001), and was >3.5 at each of the last four visits. At the first visit 51% of children had no sores, 22% were in the mild category and 27% were severe. There was a steady improvement during the study period, such that at Visits 5 through 8, 71 to 77% of children had no sores, 14 to 23% were mild and 6 to 8% were severe. Log-linear modeling of these data revealed a significant trend to reduced sore severity, with an overall reduced prevalence of children with severe pyoderma and an increased prevalence of children with no sores, whereas the numbers with mild pyoderma remained relatively stable (chi square, 19.8; df = 1; P < 0.001), although there was a significant excess, not completely explained by the model (residual chi square, 22.9; df = 11; P < 0.025).
Distribution of pyoderma. The distribution on the body of pyoderma lesions also changed during the course of the program. At the first visit 77% of children with pyoderma had lesions on their upper limbs, and 86% had lesions on their lower limbs. The proportion of children with pyoderma who had lesions on the upper limbs reduced to between 18% and 48% at Visits 3 through 8 (P < 0.025 compared with Visit 1 for all except Visit 6), whereas the proportion with lower limb lesions remained between 72 and 100% at all visits (P = 0.18 to 1.0). This represented a substantial reduction in the number of scabies-related infections, particularly around the hands and wrists, but a persistence of lower limb lesions which were related to insect bites and minor trauma.
Bacteriology.Table 2 shows the pharyngeal carriage rates of GAS during the program which, except for Visit 6, were uniformly low (<5%). The rates of GAS isolation from pyoderma swabs varied from 30 to 82% at each visit, depending on the method of culturing and the selection of types of lesions to be cultured. Group A streptococci were isolated from 97% of purulent or crusted lesions and 44% of flat, dry (healing) lesions, when cultured by the direct plating method. Because of the inconsistent choice of lesions to be cultured at each visit and the variation in culture technique between Visit 1 and subsequent visits, these data were not analyzed further.
We have demonstrated that a modification of the model of scabies control proposed by Taplin et al.4 can be successfully implemented in Australian Aboriginal communities. By concentrating on scabies control a substantial impact was also seen in the reduced prevalence and severity of streptococcal pyoderma. The program described here required less resources than the model of Taplin et al. and could be instituted largely by personnel with limited training, using simple clinical criteria for diagnosis of scabies and for grading of pyoderma severity. The use of clinical criteria meant that the bacterial culture results did not provide information that was essential to the success of the program, and such procedures could be eliminated from future models of scabies and pyoderma control.
Environmental measures such as the washing and airing of linen and clothing are usually recommended to avoid transmission and reinfestation with scabies when individual cases occur.11 These measures did not figure prominently in our program. Nor did we include measures to treat dogs for scabies, despite the belief of some health staff working with Aboriginal communities that dogs are likely to be important contributors to human scabies. Human scabies infestations acquired from infested dogs have been reported, but such infestations are considered to be short lived and to cure spontaneously.12 Our study suggests that communitybased scabies control programs can succeed by concentrating on the use of scabicides, without the need for widespread adherence to measures aimed at reducing nonhuman reservoirs of scabies.
Although our clinic-based program suggests that home-based treatment is not always a prerequisite for the success of community scabies control programs, as suggested by Taplin et al.,4 we agree that other keys to the success of similar programs are the involvement and education of the entire community, and the availability of permethrin at no (or low) cost.4 We did not institute any formal scabies education procedures but found that regular feedback, in a culturally appropriate way, of the results of the program helped to maintain the enthusiasm of the community. In poor countries who could not afford scabicide for the entire population, programs targeted at the highest risk groups (e.g. focusing on children and families with recurrent scabies at follow-up) might prove useful.13
Another option that may soon become available for public health use is ivermectin, which is effective against scabies and has further advantages of being inexpensive and able to be administered in a single, oral dose.14, 15 Although it appears to be safe in humans, clarification of its use in small children and pregnancy will be required before its use in community-based scabies programs.16 In addition the role of individuals with crusted ("Norwegian") scabies as "core-transmitters" will have to be addressed. Although one child in the community studied had severe and persistent scabies, there were no cases of true crusted scabies present. Hyperinfested people remain an important source of reinfestation in some Aboriginal communities, despite attempts at household cleaning, insecticide use, and multiple doses of ivermectin (BJ Currie, unpublished data).
We report low pharyngeal carriage rates of GAS. This is commonly found in rural Aboriginal Australian communities.17 The reasons for this, and for the apparent paradox of high rates of acute rheumatic fever1 in a population with low pharyngeal GAS isolation rates (and apparently low rates of GAS pharyngitis), are unknown. One possibility is that widespread GAS skin infection may provide immunologic protection from throat GAS carriage and/or infection.
This program is ongoing and one important challenge will be to ensure its sustainability over the long term. Similar programs have waned in the past either because of financial constraints or because of reduced enthusiasm by the community in the face of reduced disease prevalence, and scabies infestation rates have quickly returned to hyperendemic levels.4, 7 Further challenges include instituting measures to target the residual pool of lower limb pyoderma secondary to insect bites, cuts and scratches.4, 18-20 This will require education about the need for skin hygiene, application of antiseptic solutions to traumatized skin, the use of insecticides and the judicious use of penicillin for established skin sepsis. It also remains to be seen if scabies programs can succeed in larger, less isolated communities. A major obstacle to meeting this challenge in Australian Aboriginal communities is the poor resourcing of health services. Rural Aboriginal communities are characterized by inadequate housing, education, sanitation and health infrastructure,21 and the inequalities in resourcing of Aboriginal communities compared with mainstream Australia have been well-described.22 Ultimately reduced rates of scabies and streptococcal diseases will result from improvements to these fundamental services, including clinical services. In the meantime the program described here offers a practical alternative strategy to reduce the burden of these diseases.
We thank the residents and council members of Minjilang community for their support and the teachers and staff of Minjilang school for their assistance. The following Aboriginal Health Workers participated in this study: Jessie Yarmirr, Rachel Nimilga, Nellie Gaburutju, Isobel Lami Lami and William Wurlurli. We also thank the medical students who helped in the collection of data, Tania Shelby-James for performing bacterial cultures and typing and Jenny Powers and Wen Dai for statistical advice. Jonathan Carapetis is supported by a scholarship from the Public Health Research and Development Committee of the National Health and Medical Research Council of Australia. The permethrin cream for the initial mass treatment in this project was supplied free of charge by Wellcome Australia.
1. Carapetis JR, Wolff DR, Currie BJ. Acute rheumatic fever and rheumatic heart disease in the Top End of Australia
's Northern Territory. Med J Aust 1996;164:146-9.
2. Gogna NK, Nossar V, Walker AC. Epidemic of acute poststreptococcal glomerulonephritis in Aboriginal communities. Med J Aust 1983;1:64-6.
3. Kristensen JK. Scabies and pyoderma in Lilongwe, Malawi: prevalence and seasonal fluctuation. Int J Dermatol 1991;30:699-702.
4. Taplin D, Porcelain SL, Meinking TL, et al. Community control of scabies: a model based on use of permethrin cream. Lancet 1991;337:1016-8.
5. Carapetis JR, Currie BJ. Group A streptococcus, pyoderma, and rheumatic fever [Letter]. Lancet 1996;347:1271-2.
6. Fraser J. Permethrin: a Top End viewpoint and experience [Letter]. Med J Aust 1994;160:806.
7. Taplin D, Rivera A, Walker JG, Roth WI, Reno D, Meinking T. A comparative trial of three treatment schedules for the eradication of scabies. J Am Acad Dermatol 1983;9:550-4.
8. Taplin D, Meinking TL. Pyrethrins and pyrethroids in dermatology. Arch Dermatol 1990;126:213-21.
9. Mackey SL, Wagner KF. Dermatologic manifestations of parasitic diseases. Infect Dis Clin North Am 1994;8:713-43.
10. Healy MJR. Glim: an introduction. Oxford, UK: Clarendon Press, 1988:94-9.
11. Burkhart CG. Scabies: an epidemiologic reassessment. Ann Intern Med 1983;98:498-503.
12. Arlian LG. Biology, host relations and epidemiology of Sarcoptes scabiei.
Annu Rev Entomol 1989;34:139-61.
13. Henderson C. Community control of scabies [Letter]. Lancet 1991;337:1548.
14. Meinking TL, Taplin D, Hermida JL, Pardo R, Kerdel FA. The treatment of scabies with ivermectin. N Engl J Med 1995;333:26-30.
15. Lawrence GW, Sheridan JW, Speare R. We can get rid of scabies: new treatment available soon [Letter]. Med J Aust 1994;161:232.
16. Currie BJ, Maguire GP, Wood YK. Ivermectin and crusted (Norwegian) scabies [Letter]. Med J Aust 1995;163:559-60.
17. Van Buynder PG, Gaggin JA, Martin D, Pugsley D, Mathews JD. Streptococcal infection and renal disease markers in Australian Aboriginal children. Med J Aust 1992;156:537-40.
18. Allen AM, Taplin D. Skin infections in Eastern Panama. Am J Trop Med Hyg 1974;23:950-6.
19. Svartman M, Potter EV, Finklea JF, et al. Epidemic scabies and acute glomerulonephritis in Trinidad. Lancet 1972;1:249-51.
20. Suite M. Cutaneous infections in Trinidad. Int J Dermatol 1990;29:31-4.
21. Munoz E, Powers JR, Nienhuys TG, Mathews JD. Social and environmental factors in 10 Aboriginal communities: relationship to hospital admissions of children. Med J Aust 1992;156:529-33.
22. National Aboriginal Health Strategy Working Party. A national Aboriginal health strategy. Canberra, Australia
: Australian Government Publishing Service, 1989:1-234.