Parasitic skin diseases: health care-seeking in a slum in north-east Brazil



Ectoparasitic diseases are endemic in many poor communities in north-east Brazil, and heavy infestation is frequent. We conducted two studies to assess disease perception and health care seeking behaviour in relation to parasitic skin diseases and to determine their public health importance. The first study comprised a representative cross-sectional survey of the population of a slum in north-east Brazil. Inhabitants were examined for the presence of scabies, tungiasis, pediculosis and cutaneous larva migrans (CLM). The second study assessed health care seeking behaviour related to these ectoparasitoses of patients attending a Primary Health Care Centre (PHCC) adjacent to the slum. Point prevalence rates in the community were: head lice 43.3% (95% CI: 40.5–46.3), tungiasis 33.6% (95% CI: 30.9–36.4), scabies 8.8% (95% CI: 7.3–10.6) and CLM 3.1% (95% CI: 2.2–4.3). Point prevalence rates of patients attending the PHCC were: head lice 38.2% (95% CI: 32.6–44.1), tungiasis 19.1% (95% CI: 14.7–24.1), scabies 18.8% (95% CI: 14.4–23.7) and CLM 2.1% (95% CI: 0.8–4.5). Only 28 of 54 patients with scabies, three of 55 patients with tungiasis, four of six patients with CLM and zero of 110 patients with head lice sought medical assistance. The physicians of the PHCC only diagnosed a parasitic skin disease when it was pointed out by the patient himself. In all cases patients were correctly informed about the ectoparasitosis they carried. The results show that tungiasis and pediculosis, and to a lesser extent scabies and CLM, are hyperendemic but neglected by both population and physicians, and that prevalence rates of tungiasis and scabies at the PHCC do not reflect the true prevalence of these diseases in the community.


The parasitic skin diseases scabies, tungiasis, pediculosis and cutaneous larva migrans (CLM) are very common in poor communities in Brazil, particularly in the many overcrowded favelas (slums) of the big cities (Lima et al. 1984; Araujo et al. 2000; Heukelbach et al. 2001; Linardi 2001; Serra-Freire 2001). However, incidental observations indicate that these ectoparasitoses are neglected by both the medical profession and the affected population (Heukelbach et al. 2001). The patients themselves are often unaware about the infection they carry and/or do not seem to be concerned about being infested. This is worrying, as in the favelas with their poor housing, crowding and lack of hygiene, untreated patients are sources for further spread of parasitic skin diseases. Heavy infestation is common and associated with severe pathology (M. Eisele, J. Heukelbach & H. Feldmeier unpublished manuscript). There were no data about disease perception and health care seeking behaviour of infected Brazilian slum dwellers, and to fill this gap and determine the public health importance of pediculosis, scabies, CLM and tungiasis, we conducted two studies.

The first study comprised a representative cross-sectional survey of the population of a defined area of a favela in which inhabitants of all age groups were examined for parasitic skin diseases. The second study assessed health care seeking behaviour of patients attending the Primary Health Care Centre (PHCC) adjacent to the slum. The results show that the four ectoparasitic diseases are very common, but rarely diagnosed by the physicians of the PHCC. Health care seeking behaviour varied considerably according to the disease.

Materials and methods

Study area

The study took place in Vicente Pinzón II, a typical favela in Fortaleza, the capital of Ceará State in north-east Brazil, which is the poorest region of the country. The favela is close to the beach and has a total population of about 15 000 inhabitants. Two-thirds of households have access to piped water. Sixty per cent of the population have a monthly family income of less than two minimum wages (1 minimum wage = US$ 60). Adult illiteracy is 30%, unemployment rates are high, and crime is common (Family Health Program 1999). Many houses are made of improvised materials and do not have cemented floors. Waste is collected by city authorities merely at the boundaries of the slum, and rubbish is scattered throughout the area. There is no public sewage system, and hygienic conditions are precarious.

The area is covered by the so-called Family Health Programme (FHP), which was implemented in the 1990s as a result of the reorganization and decentralization of Primary Health Care Services (Haines et al. 1995). The FHP's objectives are: Primary Health Care delivery; community participation in health prevention and promotion; team work between health professionals and community health workers (Ministry of Health 1998). Families are visited regularly by community health workers. Patients can attend the PHCC during office hours (8–11 am and 2–5 pm) and participate in various prevention programmes focused on high-risk groups. Patients do not pay for services and receive free medication at the PHCC pharmacy, if it is available. The favela was divided into five areas by the FHP teams. Three of these (Luxou, Morro de Sandras and Antônio Carneiro) are known to be highly infested by ectoparasites (J. Heukelbach, unpublished observation). Approximately 9000 people live in these areas. The socio-economic characteristics and the prevalence rates of common infectious and non-infectious diseases are similar in the three areas (Family Health Program 1999).

Study design

For the cross-sectional study, one area (Morro de Sandras) was randomly selected using the random number calculator of the Epi Info software package (version 6.04d). Here, all 327 households (1460 individuals) were visited twice by one of the investigators (T.W.). Each member of the family present at one or the other occasion was thoroughly examined for the presence of scabies, tungiasis, head lice and CLM. As the examination had to be performed in places where privacy was not fully guaranteed, pubic lice were not looked for. Body lice, myiasis and ticks do not occur in the study area.

Inhabitants of the favela seeking medical assistance at the outpatients' clinic of the PHCC from 12 February 2001 to 9 March 2001 (16 working days) were eligible for the health care seeking behaviour study. Only patients who came spontaneously to the PHCC were included in the study, i.e. those who had been convened by a physician or were participating in prenatal care, cancer prevention or vaccination programmes were excluded. Sample size calculation was based on the assumption that point prevalence rates at the PHCC were about 20%; at a confidence level of 95% and an absolute precision of 5%, the necessary number of patients was calculated to be about 250. Therefore, in order to obtain a sufficient number of patients with ectoparasitoses during the time frame available for the study, individuals from the three areas Luxou, Morro de Sandras and Antônio Carneiro were included. A total of 288 patients could be enrolled.

First participants were interviewed about their motives for seeking medical assistance using a standardized and pre-tested questionnaire. The patients were then allocated to one of the four physicians by the PHCC staff, according to the routine of the health post. As checked after the study, this resulted in random distribution of patients to each of the four physicians. The physicians were fully informed about the study, although they did not know whether a particular patient participated in the study or not. When the patient had been examined and received the prescription, he/she was asked to enter another examination room where the whole body was thoroughly examined (excluding the genitals) by one of two experienced investigators (E.v.H. or B.R.) for scabies, tungiasis, head lice and/or CLM.

In both studies, all diagnoses were made clinically. The following findings were considered to be diagnostic. Scabies: erythematous papular, vesicular, pustular or bullous lesions associated with itching. Tungiasis: a red-brownish itching spot with a diameter of 1–3 mm (early stage), lesions presenting as a white patch with a diameter of 4–10 mm with a central black dot (mature stage), black crust surrounded by necrotic tissue (late stage with dead parasite), as well as lesions altered through manipulation by the patient (partly or totally removed fleas leaving a characteristic sore in the skin and suppurative lesions caused by the use of non-sterile perforating instruments). Head lice: presence of nits, nymphs or adult parasites on the scalp or hair. CLM: characteristic serpinginous lesions associated with severe itching.

Statistical analysis

Data were entered twice into a database using Epi Info software package (version 6.04d) and checked for validity. Calculation of 95% confidence intervals of point prevalence rates and chi-square were calculated using the respective Epi Info modules.

Ethical considerations

Permission to perform the studies at the PHCC and in the community was obtained from the Health Secretary of Fortaleza Municipality. The physicians and the medical staff of the PHCC were informed about the study and agreed to co-operate. Community associations of the favela (associações dos moradores) gave their consent for the community study. Prior to the studies, meetings with staff members, community health workers and community leaders were held in which the objectives were explained. In both studies, informed oral consent was obtained from each patient after explaining the objectives of the study. In the case of minors, the guardians were asked for consent.


There was no significant difference in the overall sex distribution between the patients of the two study groups. However, the male/female ratio varied considerably in the age groups: about half of the patients at the PHCC aged 0–14 years, but only 24% of the patients at 15 or more years of age were males. In the community, again half of the individuals up to 14 years were males, but in the age groups of 15 years or more 38% were males (both P < 0.001). Age and sex distribution of individuals examined in the community and at the PHCC are summarized in Table 1.

Table 1.  Age and sex distribution of individuals examined in the community (n = 1185) and the Primary Health Care Centre (PHCC) (n = 288)
Age group
Males nFemales nTotal n (%)*Males nFemales nTotal n (%)*
  • *

     Percentage of all individuals examined;

  •  in these age groups the proportions were significantly different between the community and the PHCC.

0–4116132248 (20.9%)6546111 (38.5)
5–98187168 (14.2%)162844 (15.3)
10–149879177 (14.9%)111324 (8.3)
15–194779126 (10.6%)41418 (6.3)
20–39119191310 (26.2%)144054 (18.8)
40–594866114 (9.6%)52227 (9.4)
≥60142842 (3.5%)3710 (3.5)
Total5236621185 (100%)118170288 (100)

In the community-based sample, we examined 1185 of the 1460 individuals (81%) living in the area Morro de Sandras. A total of 275 individuals were absent during two visits. Point prevalence rates were: head lice 43.3% (95% CI: 40.5–46.3), tungiasis 33.6% (95% CI: 30.9–36.4), scabies 8.8% (95% CI: 7.3–10.6) and CLM 3.1% (95% CI: 2.2–4.3); 62.5% (95% CI: 59.7–65.3) had any parasitic skin disease, and 23.5% (95% CI: 21.1–26.0) suffered from at least two ectoparasitoses.

A total of 288 visitors of the PHCC were eligible; none refused to participate. Prevalence rates in this sample were: head lice 38.2% (95% CI: 32.6–44.1), tungiasis 19.1% (95% CI: 14.7–24.1), scabies 18.8% (95% CI: 14.4–23.7) and CLM 2.1% (95% CI: 0.8–4.5). About 159 individuals (55.2%) had any parasitic skin disease, 52 (32.7%) presented with two or more.

The overall prevalence rates of tungiasis and scabies in the community were significantly different from the prevalence rates detected in the patients of the PHCC (both P < 0.0001). The rates of head lice and CLM did not differ between the study groups. Figure 1 depicts the age-specific prevalence rates of the four ectoparasitic diseases in the community and the PHCC group. For pediculosis, age-specific point prevalence rates were similar in both groups, whereas the age-specific prevalence rates of tungiasis were consistently lower in patients visiting the PHCC. In contrast, the age-specific prevalence rates of scabies were consistently higher. Age-specific prevalence rates of CLM were too low for a meaningful comparison between the two groups.

Figure 1.

Age-specific prevalence rates of four ectoparasitic diseases in the community (n = 1185) and at the PHCC (n = 288). (a) pediculosis; (b) tungiasis; (c) Scabies; (d) cutaneous larva migrans.

The reasons why patients visited the PHCC, diagnoses by PHCC physicians and the true number of patients with one or more ectoparasitoses are summarized in Table 2. Only about half of the patients with scabies (28 of 54) sought medical assistance because of this disease; none of the 28 patients made a wrong diagnosis. The physicians of the PHCC diagnosed scabies in 26 of 54 cases.

Table 2.  Number of patients with an ectoparasitosis as the reason for visiting the Primary Health Care Centre (PHCC), diagnoses made by physicians of the PHCC, true number of patients affected with one or more ectoparasitosis (all = 288) and prevalence in the community (n = 1185)
for visiting
physician [B]
Patients who
diagnosed their
True number
of patients
affected [C]
Prevalence of
at PHCC (%)
Prevalence of
in community
Significance of
difference [A]
vs. [C]
of difference
[B]vs. [C]
Scabies2826285418.88.8P < 0.002P < 0.001
Tungiasis3335519.133.6P < 0.0001P < 0.0001
Head lice00011038.243.3P < 0.0001P < 0.0001
CLM44462.13.1P = 0.5P = 0.5
Any ecto-  parasitosis35333515955.262.5P < 0.0001P < 0.0001

Only three patients (1%) attended the PHCC because of tungiasis, all of them showing signs of severe super-infection, and only in these cases the physicians diagnosed the presence of T. penetrans. However, the true prevalence was 19%.

Thirty-eight per cent of the patients attending the PHCC had pediculosis. Surprisingly, this caused no one to seek medical attention. Neither was a single case of head lice diagnosed by the physicians of the PHCC. Conversely, two-thirds of patients with CLM attended the PHCC for this reason. The two cases which escaped the attention of patients and physicians were children aged 1 and 2 years, with multiple infections: one suffered from CLM, tungiasis, scabies and pediculosis, the other from CLM, tungiasis and scabies. Only in the first case, the reason for attending and the physicians' diagnosis was an ectoparasitosis, namely scabies.

The percentage of patients attending the PHCC because of their ectoparasitic disease differed between those knowing they had scabies/tungiasis, scabies/pediculosis, tungiasis/CLM or pediculosis/CLM, respectively (all P < 0.0001), as well as between tungiasis/pediculosis (P = 0.01). It did not differ significantly between scabies/CLM (P = 0.5). The four physicians at the PHCC had clinical experience ranging from 1 year to more than 20 years in Primary Health Care. However, there was no difference in the physicians' performance in establishing a correct ectoparasitic diagnosis (P > 0.05 for each of the four infections).


Health care seeking behaviour and disease perception have been studied for several bacterial and parasitic infections, such as sexually transmitted diseases, tuberculosis, malaria and childhood diarrhoea (Schultz et al. 1994; Liefooghe et al. 1997; Granich et al. 1999; Fonck et al. 2001). However, these aspects have never been systematically investigated in parasitic skin diseases. In one single study in Tanzania, the community perception of any kind of skin disease has been looked for (Satimia et al. 1998). However, only one ectoparasitosis, namely scabies, was prevalent in the community.

In our population the four major parasitic skin diseases were very common. When data from the community survey and the PHCC were compared, two distinct patterns emerged: whereas the presence of CLM – and to a lesser extent also scabies – caused patients to visit the PHCC, which neither tungiasis nor pediculosis did. This is exemplified by the finding that none of the 110 patients with head lice attended the Health Centre for this reason. The three patients who consulted a physician of the PHCC because of tungiasis showed severe super-infection of the lesions and in fact needed systemic antibiotic treatment.

The prevalence of scabies in the PHCC was almost double that of the community, probably because a considerable number of individuals with scabies sought medical attention. Interestingly, as to tungiasis, exactly the opposite occurred. Unsurprisingly, pediculosis showed the same prevalence rates in both groups, as this disease did not influence health care seeking behaviour at all.

Medical treatment for scabies – topical benzyl benzoate – is available at the PHCC, but the Health Centre does not provide any drug or surgical treatment for pediculosis, tungiasis or CLM. Presumably, this impeded patients with one of these three diseases from seeking medical assistance at the Health Care Centre. Besides, scabies does not suffer any stigmatization or mystification in the population of the favela, as it is the case for tungiasis and CLM (B. Winter, unpublished observation).

In the community, tungiasis is not only common but also frequently associated with severe pathology (Feldmeier et al. 2002; M. Eisele, J. Heukelbach & H. Feldmeier unpublished manuscript). It seems therefore paradoxical that affected individuals do not seek medical care. The explanation may be that tungiasis is so intricately linked to the daily life in a favela in this part of Brazil that it has become part of normal life (Heukelbach et al. 2001). Besides, tungiasis is considered a condition that has to be resolved by the person himself or, in case of children, the respective caretakers, and not a health professional. It is conceivable that stigmatization associated with tungiasis also impedes patients seeking medical attention. Attack rates are so high that daily inspection of the feet and a subsequent visit of the health post are not feasible for many inhabitants, in particular for parents or guardians of many children. In fact, in the 2 years before starting research on ectoparasitic diseases in the area, not a single patient presented with tungiasis at the PHCC. Finally, the disease is more common in poor settings (Heukelbach et al. 2001); people with a lower socio-economic status are generally under-represented at the PHCC.

Pediculosis is even more neglected by the population than tungiasis. This is not surprising, as it is usually considered a nuisance in the study area (B. Winter, unpublished observation). In fact, for most inhabitants living in precarious conditions, acute diseases such as diarrhoea and respiratory tract infections require more attention, and day-to-day life is full of troubles and existential worries. The high prevalence, though, is a matter of concern as pediculosis is considered to be one of the most common causes of impetigo in developing countries (Burgess 1995).

Cutaneous larva migrans, although being an auto-limited disease, is feared by the local population. People believe that under certain circumstances, patients infested with larva migrans will die. This could explain why people seek medical assistance when they experience CLM although no treatment is available.

The physicians of the PHCC only diagnosed ectoparasitic diseases when mentioned by the patient. There was not a single case in which the physician diagnosed a parasitic skin disease which was not pointed out by the patient himself. Ectoparasitic diseases are widely neglected by the health professionals of the PHCC.

Point prevalence rates can be deduced from the prevalence rates of patients at the PHCC only for pediculosis and CLM. The prevalence of tungiasis is underestimated if derived from the PHCC patients, the prevalence of scabies, overestimated. Additional studies have to be carried out to further elucidate the different aspects of ectoparasitic diseases in poor communities.

In summary, ectoparasitic diseases, in particular tungiasis and pediculosis and to a lesser extent scabies and CLM, are hyperendemic in the study area. They are neglected by both population and physicians. We suggest that for effective control of ectoparasites, continued education of medical staff should not only focus on endemic diseases and public health issues, but also on increasing awareness of poverty-associated ectoparasitoses.


This study was supported in part by the Ärztekomittee für die Dritte Welt, Frankfurt (Germany), the World Health Organization, Geneva (Switzerland), Kongregation der Franziskanerinnen, Salzkotten (Germany), Laboratórios Sintofarma S.A., São Paulo (Brazil), Bayer Health Care, Leverkusen (Germany) and Merck do Brazil, Rio de Janeiro (Brazil). We thank Dr Anastácio de Queiroz Sousa and Dr Luiza de Marilac Meireles Barbosa from the Ceará State Ministry of Health (Brazil) for continuous support. The skilled assistance of Vânia Santos de Andrade Souza and Walter Antônio da Silva is gratefully acknowledged. We thank the staff of the Primary Health Care Centre Aída Santos e Silva, Fortaleza (Brazil) for their cooperation.

Jörg Heukelbach (corresponding author) and Rômulo César Sabóia Moura, Rua José Vilar de Andrade 257, Fortaleza, CE 60833-830, Brazil. Fax: +55 85 2733031. E-mail:;
Hermann Feldmeier, Evelien van Haeff, Babette Rump and Thomas Wilcke, Institute for International Health, Centre for Humanities and Health Sciences, Free University of Berlin, Fabeckstr. 60-62, 12203 Berlin, Germany.