On 29 November 2002 the management of the local hospital in Barbastro (Huesca) reported to the local public health authorities the existence of a scabies outbreak, with at least four cases in staff (two nurses, one auxiliary nurse and one stretcher-bearer) on the internal medicine ward of the hospital. The first case (in a nurse) had onset of symptoms on 5 November. The index case (in an auxiliary nurse) was recognised on 25 November and had onset of symptoms on 13 November. The health authorities began an epidemiological investigation to study and control the outbreak.

Methods
Clinical histories of all patients with symptoms compatible with mite infestation and admitted to the internal medicine ward of the hospital in the four weeks prior to the beginning of symptoms in the first case were reviewed. The preliminary investigation identified a patient as probable source of the outbreak (probable primary case). This person was a man aged 92 years residing in a nursing home and admitted to the hospital on 1 November 2002 for a respiratory illness. He presented with generalised lesions of long duration (at least three months) that were treated with anti-inflammatory and antipruritic medication. He was not diagnosed with scabies at that time, and a cutaneous biopsy was not made. He died on 9 November.
An investigation carried out in the patient's nursing home from 3-5 December identified other scabies cases. The patient's roommates and caregiver were also diagnosed with scabies.
Consequently, the outbreak of scabies in the hospital at Barbastro was initially linked to exposure to an infested patient who came from a nursing home in the same village. The aims of the epidemiological investigations were to identify the causal factors and circumstances that caused the outbreak; to describe its development, estimating frequency and effect rates; to define and implement recommendations to control the outbreak.

The following definitions were initially used:
• Outbreak period was defined from the date of the probable primary case's admission to the internal medicine ward of the hospital (1 November 2002) to the date of the end of treatment of all cases, plus the maximum incubation period for scabies (19 February 2003).
• Suspected case: hospitals caregivers of patients admitted to the internal medicine ward during the outbreak period, or patients admitted to the internal medicine ward during the same period and who had presented with one of the following symptoms: pruritus of several days evolution (generalised or localised) or appearance of cutaneous lesions, irrespective of their extension and severity, suggesting scabies infestation.
Case contacts presenting with scabies signs and symptoms were also considered as suspected cases. They were all diagnosed and followed up by the hospital's dermatology department.
• Confirmed case: suspected case with confirmation of Sarcoptes scabiei by direct vision, pathological anatomy, or microbiological analysis.
• Contact: person without signs or symptoms of scabies but fulfilling at least one of the following criteria:
a) Having been part of the hospital staff and in charge of patients residing on the internal medicine ward during the outbreak period.
b) Having been admitted to this ward during the outbreak period.
c) Having had intimate contact with a case. Sexual partners and family members were considered to be contacts.
• Incubation period: delay between the date of the first physical contact with an infested person and the date of the appearance of the first symptoms in every case.
A standardised questionnaire was used for the outbreak investigation.
During the outbreak period, there were 59 HCWs (41 cleaning staff and 18 hospital stretcher-bearers) working regularly on the internal medicine ward of the hospital: 49 of them (83%) were interviewed.
In addition, an active case search for scabies was conducted among the 140 patients admitted to the internal medicine ward of the hospital during this period, in collaboration with the coordinator of the primary health centre in Barbastro. One hundred and twenty four of them (89%) were located and evaluated.
The sexual partners (8) and household contacts of the cases were interviewed by physicians of this primary health centre or their respective general practitioners.
Topical treatment with 5% permethrin cream, applied once to the entire body except the face and soles of the feet (the preparation is left on the body for at least 8 hours), was indicated for cases and contacts simultaneously (1,2). Bed linen was washed once on the day after the first treatment. Ten days later, cases and contacts were reviewed by the dermatology department of the hospital and treated again if it was considered to be necessary (1,2).
A database was made using Epi Info 6.0.

Relative risk for caregivers working on the internal medicine ward was estimated, taking into consideration that physicians had a lower level of physical contact with the scabies patients; the p-value was also estimated, using Fisher's test of exact probability with a significance level of 95% (alpha = 0.05).

Results
Between 5 November 2002 (onset of first symptoms for the first case) and 5 January 2003 (onset of symptoms for the last case), a total of 17 scabies cases were reported among HCWs who had worked on the internal medicine ward of the hospital in Barbastro (11) and patients admitted to this ward (6). Figure 1 shows the epidemic curve.

Among the 41 caregivers assigned to this ward, seven (17%) were infested with mites during the outbreak period. Three (17%) of the 18 hospital stretcher-bearers who were in contact with patients on this ward suffered from scabies, as did a member of the ambulance service, who transported the patient who was the probable source of the outbreak. Table 1 shows the attack rates for the different work groups on this ward.

There was no statistically significant difference between the risks of becoming infested of the different groups, probably due to the low sample size: nurses 3.2 (CI 95% 0.4-25.0), auxiliary nurses 1.7 (0.2-16,7), stretcher-bearers 2.0 (0.2-17.0) and physicians 1.
Six of the 11 cases in HCWs were in women and five were in men. Two of them were confirmed by direct vision of mites, and the others fulfilled the suspected case criteria. The average age was 39.4 years (24-61 years) and the standard deviation was 12.0. Their main symptoms and signs were: pruritus (11) (5 generalised and 6 localised, 9 with nocturnal predominance), papules (9), vesicles (6), and erythema (2). The lesions were mainly located on the upper arms and forearms (73%, including axillary fold), hands (63%), thorax and abdomen (55%), and inguinal region (27%).
The average incubation period was 16.7 days (5-47), the median value was 12, and the standard deviation was 14.2.
The majority of the hospital staff working on the internal medicine ward reported that they paid attention to personal protection while caring for patients. Six percent of them (3/49), however, reported that they did not use always disposable latex gloves during caregiving. Furthermore, only three HCWs also reported systematically protecting upper arms and forearms with long-sleeved coats. There was no statistically significant relation between the attention paid to such measures and the presence of scabies, probably due to the small sample size.
No relation was found between scabies infestation and contact with cases' clothes, animals, or intimate contact with people other than their usual partner (nobody related this personal detail).
Eighteen percent of the uninfested caregivers (7/38) treated with permethrin reported adverse effects, mainly pruritus (7/7) and erythema (3/7). Among cases, two persons reported increased pruritus the next week after treatment.

Besides, 29% (11/38) of the uninfested HCWs linked with the internal medicine ward presented slight cutaneous lesions during the outbreak period, not related with this treatment, being diagnosed as psychogenic cause: pruritus (11), papules (10), vesicles (3) or erythema (2).
Among the 140 patients admitted to the internal medicine ward of the hospital during the outbreak period, six patients (4%) presented with scabies. They were four men and two women, with an average age of 78 years (range 73-84). All were dependent and bedridden patients, with a high need for maintained care by their caregivers. The mean incubation period was 31 days (19-46). The main symptoms and signs were: pruritus (five localised and one generalised, and in three cases, of nocturnal predominance), erythema (5), papules (4), scabies burrows (2) and vesicles (2). The preferred locations were upper arms and forearms (5), legs (4) and trunk (3).
One tertiary case (the wife of an HCW) was reported among household contacts of all the cases.
The cases were mainly diagnosed, treated and reviewed by the dermatology department of the hospital. There were no cases among staff of this department.

Discussion
Delay in diagnoses of scabies among elderly people due to the inspecificity of their lesions has been previously described (3,4). Primary cases in scabies outbreaks are frequently in elderly with pruritus diagnosed as senile, psychogenic or degenerative cause more than of infectious origin (3). In bedridden patients the main sign of scabies infestation is the appearance of vesicles or red papules on the back, instead of scabies burrows on the hands and fingers. That was probably the cause of the diagnosis delay for the patient who was the vehicle for this outbreak.
The delay in diagnosis of scabies amplifies the exposure period for other people such as HCWs (4). A single contact with an infested person can be enough to transmit the mites and to cause a nosocomial outbreak (3).
The nosocomial outbreaks of scabies are usually the result of delay in diagnosing in an infested patient (5,6,7) and of lack of attention to individual protection measures by staff while caring for patients (8).
In this outbreak, the hypothesis of infestation risk related to the physical contact rate with the patients could not be demonstrated, possibly due to the low number of cases. The differences in estimation of the relative risk for each group of HCWs (the physicians were considered to be the group at lowest risk) were not statistically significant. In other, similar outbreaks, extensive physical contact with infested patients was a risk factor for scabies infestation (9).
The transmission of the mites was probably person-to-person among the members of staff on the internal medicine ward, as a consequence of sharing the workspace. A similar transmission could explain the infestation of the six patients admitted to the ward during the outbreak period, which was detected by an active search. In this sense, all of them were old, bedridden and dependent patients, and needed continuous care from HCWs.

Concerning the lack of attention paid to individual protection measures, it is possible that the short-sleeved coats worn by the HCWs did not protect their upper arms and forearms sufficiently from direct contact with infested patients. The use of short-sleeved coats has been identified in other nosocomial outbreaks as a high risk factor for scabies transmission (8).
The patient presumed to be the source of the outbreak was bedridden. His need for maintained care probably played a role in the transmission of mites to his caregivers. The outbreak has not been linked to contact with fomites, patients' clothes, animals or intimate contact with people other than usual partners. The appearance of cutaneous lesions in uninfested caregivers can be due to the alarm triggered by the occurrence of the outbreak. Pruritus attributed to psychogenic causes has already been described in similar outbreaks (10).
Topical treatment with 5% permethrin cream was demonstrated to be very effective for infested cases. This treatment cured all of the scabies cases, and there were no severe adverse effects. Only one tertiary case was identified among household contacts, probably due to the care given to the case.
The implementation of guidelines to treat and strict follow up all cases and contacts (with close supervision of treatment), as well as the activation of the epidemiological surveillance system, was fundamental to the control of the nosocomial outbreak.