Undiagnosed and Potentially Lethal Parasite Infections Among Immigrants and Refugees in Australia

Sonia R. Caruana MPH, Heath A. Kelly FAFPHM, Joanne Y.Y. Ngeow MB BS, Norbert J. Ryan PhD, Catherine M. Bennett PhD, Ley Chea MB BS, Sophy Nuon MB BS, Narin Bak FRACP, Susan A. Skull FAFPHM, Beverley‐Ann Biggs PhD, FRACP
DOI: http://dx.doi.org/10.1111/j.1708-8305.2006.00045.x 233-239 First published online: 1 July 2006


Background Intestinal parasite infections are a major cause of ill health in many resource‐poor countries. This study compares the types and rates of these infections and their risk factors in recently arrived and long‐term immigrants in Australia.

Method Cross‐sectional surveys of 127 East African and 234 Cambodian immigrants and refugees were undertaken in 2000 and 2002, respectively, to assess the burden of intestinal parasites and collect demographic information. Serum samples were assessed for eosinophilia and Strongyloides stercoralis and Schistosoma antibodies, and feces examined for ova, cysts, and parasites.

Results Intestinal parasites were identified in 77/117 fecal samples from East African and in 25/204 samples collected from Cambodian participants. Eleven percent (14/124) of East Africans and 42% (97/230) of Cambodians had positive or equivocal serology for S stercoralis. Schistosoma serology was positive or equivocal in 15% (19/124) of East African participants.

Conclusion Potentially serious intestinal parasite infections are common among recent and longer term immigrants despite multiple visits to health care providers. Immigrants and refugees from high‐risk countries would benefit from comprehensive health checks soon after resettlement.

Each year, a large number of people migrate permanently to Australia. Southeast Asian countries were a major source of immigrants in the 1980s,1 and from the late 1990s, Australia has sponsored a large humanitarian intake from Africa.2 Most immigrants have resettled in the Eastern states of Australia, particularly in Sydney and Melbourne.3

Many immigrants arrive in Australia with undiagnosed or poorly managed medical conditions that relate to poverty and oppression, the breakdown of health systems during economic and social unrest in their home country, and long periods spent in refugee camps.4 Studies undertaken in Australia and elsewhere have shown that intestinal parasite infections were a particular problem in immigrants from the Mekong region who settled in the 1980s.5–8 Studies conducted 2 and 6 years after resettlement in Canada indicated persistence of some intestinal parasite infections among Cambodian immigrants despite treatment on arrival. The most dangerous of these was Strongyloides stercoralis9 that may persist in humans for 40 years or more and cause serious morbidity and mortality.10

We have previously documented undiagnosed chronic strongyloidiasis in 24% of a group of 95 Laotian immigrants who had lived in Melbourne for an average of 12 years.11 Despite treatment with albendazole, one of these participants died of disseminated strongyloidiasis after the inappropriate administration of corticosteroids by a clinician who was not aware of the patient’s clinical history.12 The prevalence of strongyloidiasis in the 23,000 Cambodian immigrants living in Australia is unknown.

East African immigrants to the United States, Europe, and Israel have also been shown to harbor numerous intestinal parasites, including S stercoralis and Schistosoma mansoni.13–17 In Australia, 18% of children recently arrived from East Africa had pathogenic parasites identified in stool samples. Eleven percent and 2% had positive or equivocal Strongyloides and Schistosoma serology, respectively.18 The burden of parasitic infections in East African adults who have resettled in Australia has not been documented.

As migration to Australia and other developed countries increases, it is important to develop evidence‐based strategies for comprehensive approaches to improving the health of new arrivals. In this study, we aimed to determine whether long‐term immigrants from Cambodia living in Melbourne have a high prevalence of undiagnosed chronic strongyloidiasis and to compare the type of parasite infections in this group with those in recently arrived immigrants from East Africa.


Cross‐sectional surveys were undertaken in the East African and Cambodian communities in 2000 and 2002 as previously reported.19,20 Both surveys were conducted after extensive community consultation. Ethics approval was obtained from the Royal Melbourne Hospital Research Foundation Clinical Research Ethics Committee, and written informed consent was obtained from all participants. The East African survey included immigrants (who arrived in Australia between 1997 and 2000) aged 16 years or older from Ethiopia, Eritrea, Kenya, Somalia, and Sudan. The Cambodian survey included longer term settlers aged 15 years or older.

In both surveys, convenience samples comprised participants recruited from the patient lists of two inner‐city Melbourne community health centers and two private general practitioner (GP) clinics. We also used an extensive advertising campaign to encourage community members to attend these facilities for health checks.19,20 To assist with the surveys, we employed community workers or registered nurses from the relevant communities. Interpreters were available at all consultations. Questionnaires were conducted by community workers using face‐to‐face interviews in the languages of participants in the East African survey and in English or Khmer in the Cambodian survey. Information collected included demographic profile, knowledge about and potential risk factors for infection with intestinal parasites, and current utilization of health services. In addition to interviews, participants were invited to provide blood and fecal specimens.

Blood samples were collected for the estimation of hematological markers and serological testing. Blood was separated, and the serum stored prior to use at 4°C for up to 2 weeks in the East African and at −20°C in the Cambodian surveys. Serum samples were tested in batches for S stercoralis antibodies (immunoglobulin G) using an in‐house enzyme‐linked immunosorbent assay (ELISA) with soluble antigens from third stage Strongyloides ratti larvae.21 The sensitivity and specificity of this test are 93 and 95%, respectively.22 Based on a reference range determined from a comparison of the distribution of ELISA optical density (OD) values in uninfected controls with sera of those with microscopically confirmed infection, an OD value of >0.5 was considered positive. An OD of 0.3 to 0.5 was considered equivocal and <0.3 negative. In the East African survey, serum samples to detect Schistosoma antibodies were also batch tested using an indirect hemagglutination assay for antibody to adult S mansoni antigen (Cellognost Schistosomiasis H commercial kits—Dade Behring Inc., Newark, DE, USA). The sensitivity of this assay is reported as 95% for S mansoni, 90% for Schistosoma haematobium, and 50% for Schistosoma japonicum.23 Significant Schistosoma antibody titers were considered to be 1:32 or higher. The normal range for eosinophils was 0.04 × 109/L to 0.40 × 109/L, and for hemoglobin concentration 115 to 165 g/L (females) and 135 to 180 g/L (males).

One fecal sample was collected from each patient into sodium acetate–acetic acid–formalin fixative using a standard kit (Para‐Pak, Meridian Diagnostics Inc., Cincinnati, OH, USA). Fecal concentrates were made using a modified formalin–ethyl acetate procedure and commercially supplied tubes (Johns Parafilter, Selby Scientific, Clayton, Vic, Australia).24 A small amount of the fecal concentrate was resuspended in normal saline and iodine and placed onto a glass microscope slide where cysts, ova, and larvae were identified by light microscopy on the basis of morphology and size. Staining of fecal smears for Entamoeba histolytica was also performed. Processing and testing of all blood and fecal specimens were undertaken at the Victorian Infectious Diseases Reference Laboratory, the state’s largest accredited public health reference laboratory.

The results were reported to participating or nominated GPs who were responsible for appropriate patient management. Feedback on survey findings was also provided directly to the communities. Data analysis used Epi Info, Version 6 (Centers for Disease Control, Atlanta, GA, USA) and Stata Software, Version 8 (Stata Coorporation, College Station, TX, USA 2003). Univariate analysis of categorical variables derived from the questionnaires and laboratory results was performed using the chi‐square or Fisher’s exact test. The relative risk (RR) and 95% confidence interval (95% CI) were used to measure the strength of association between a predictor variable and an outcome variable and to assess whether an association between the predictor and the outcome variables was statistically significant.


East african survey

The East African survey included 127 participants aged between 16 and 75 years (median 32 years), of whom 81 (64%) were female. Participants were born in Somalia (n= 68), Eritrea (n= 29), Ethiopia (n= 21), Sudan (n= 6), Egypt (n= 2), and Djibouti (n= 1). More than one third (n= 48) had very limited or no understanding of English. Participants had been in Australia for 0.7 to 42.6 months (median 19.1 months). The majority (88%) arrived in Australia from an African country in which they were not born, and more than half (n= 57) had spent time in a refugee camp prior to arrival (median 68 months, range 1–240 months). Sixty (47%) participants lived in a household with five or more people (range 1–11 persons). While almost two thirds (n= 82) could nominate a regular GP, half (n= 63) had experienced problems with accessing and utilizing health services, with language the main problem. Symptoms reported by participants since arriving in Australia included fever (34%), stomach pain (30%), weight loss (25%), and diarrhea (13%).

Fecal specimens

Of the 117 participants who provided a single stool sample, ova, cysts, or parasites were detected in 77 (66%; 95% CI 58%–74%) (Table 1). Pathogenic parasites were detected in 17 (22%; 95% CI 12%–32%). Only one pathogen was found in each sample. Gender, age, and duration of residence in Australia did not vary significantly between those with pathogenic parasites and those without. Intestinal parasites were detected in half the participants who reported having experienced fever, stomach pain, weight loss, and/or diarrhea. There was no correlation between the presence of symptoms and parasitic infection (defined as a fecal sample in which pathogenic parasites were identified and/or a positive serology test; RR = 1.00; 95% CI 0.64–1.58).

View this table:
Table 1

Pathogenic and nonpathogenic intestinal parasites isolated in stool samples collected in the East African (n= 117) and Cambodian (n= 204) surveys by gender

OrganismEast AfricanCambodian
Females (%)Males (%)Females (%)Males (%)
Strongyloides stercoralis002 (1)8 (4)
 Hookworm0004 (2)
Giardia lamblia3 (3)3 (3)00
Trichuris trichiura5 (4)000
Entamoeba histolytica2 (2)2 (2)00
Hymenolepis nana01 (1)00
Dientamoeba fragilis1 (1)000
Blastocystis hominis33 (28)22 (19)4 (2)4 (2)
Iodamoeba butschlii1 (1)3 (3)01 (0.5)
Entamoeba coli15 (13)6 (5)1 (0.5)3 (1.5)
Endolimax nana12 (10)14 (12)00
Entamoeba hartmanni1 (1)000
  • More than one intestinal parasite found in some samples.

Serological markers

A total of 124 participants were assessed for the presence of S stercoralis and Schistosoma antibodies, and the findings are reported in Table 2. Gender, age, residence in a refugee camp, household size, or time spent in Australia were not associated with seropositivity for strongyloidiasis or schistosomiasis.

View this table:
Table 2

Strongyloides and Schistosoma serology and eosinophil results for the Cambodian and East African surveys

Positive (%)Equivocal (%)Negative (%)Positive (%)Equivocal (%)Negative (%)
East African survey
 Serology results (n= 124)2 (2)12 (10)110 (89)14 (11)5 (4)105 (85)
Strongyloides larvae in feces000000
Cambodian survey
 Serology results (n= 230)82 (36)15 (7)133 (58)NDNDND
Strongyloides larvae in feces1000NDNDND
  • ND = not done.

Hematological markers

Of the 124 participants who provided a sample, 11 participants (9%; 95% CI 3%–15%) had an elevated eosinophil count (Table 2). Gender was not associated with eosinophilia (RR = 1.4; 95% CI 0.5–4.4). The median age of those with eosinophilia was 40 years (range 17–55 years) compared to 31 years in those with a normal eosinophil count (range 16–75 years). Eight participants (7%), all of whom were female, were anemic (median hemoglobin 105 g/L, range 83–113 g/L).

Cambodian survey

A total of 234 people aged 15 to 92 years (median 45.5 years) volunteered for this study, of whom 126 (54%) were female. Most responded to every question on the questionnaire and provided biological samples for testing. Participants were born in Cambodia (n= 224), Thai refugee camps (n= 8), or Australia (n= 2). The year of arrival in Australia ranged from 1974 to 2002, with a median time since resettlement of 14 years (range <1–28 years). About three in four participants (172/232) had returned to Cambodia at least once, staying an average of 3 or 4 weeks. While 97% spoke Khmer at home, 38% (88/232) could not read their spoken language. Only 83 (36%) considered themselves proficient in English. There was a low level of education, with 71% (165/232) completing only primary school. More than two thirds (159/232) were unemployed. Most (n= 206; 89%) lived in a household with three or more persons. Ninety‐five percent of participants had made contact at least once with a primary health care provider in the 12 months prior to the study. One third (n= 78) were unaware of possible modes of transmission of intestinal parasites.

Fecal specimens

A total of 204 participants provided a single stool sample, and the results are shown in Table 1. Four participants with hookworm had traveled back to Cambodia since resettling in Australia and two with S stercoralis larvae had not. While age, time since resettlement in Australia, and travel back to Cambodia were not associated with pathogenic parasites, they were more common in male than in female participants (RR = 7.0; 95% CI 1.6–30.6).

Strongyloides serology

A total of 230 people were tested for S stercoralis antibody, and 82 (36%; 95% CI 30%–42%) were positive (Table 2). Males were more likely to have positive or equivocal serology than females (RR = 1.5; 95% CI 1.1–2.1). Age, travel back to Cambodia, or time since resettlement in Australia were not associated with S stercoralis infection (defined as S stercoralis identified in feces and/or seropositivity). All those with S stercoralis larvae and two with hookworm ova detected in stool samples had positive serology.

Eosinophilia and anemia

Thirty‐nine of the 229 participants tested (17%; 95% CI 11%–23%) had eosinophilia. Males (n= 26; 67%) were almost three times more likely to have eosinophilia compared to females (RR = 2.4; 95% CI 1.3–4.3). Thirty‐six of the 82 participants (44%; 95% CI 34%–54%) who were seropositive and three with negative serology for S stercoralis had an elevated eosinophil count (Table 2). Half of those with S stercoralis larvae or hookworm ova in stools had eosinophilia. Twenty‐eight of those (28/178, 16%) with no parasites in feces had eosinophilia. No participants reported allergies, another explanation for eosinophilia, at or around the time of the survey.

Participants with eosinophilia were four times more likely to be seropositive for S stercoralis than those with a normal eosinophil count (RR = 3.8; 95% CI 2.9–5.0). Investigation of the validity and feasibility of eosinophilia as a screening tool for the detection of S stercoralis infection (detection of larvae or seropositivity) showed that in this survey, sensitivity was 44%, specificity 98%, with positive and negative predictive values of 92 and 76%, respectively (data not shown). Of the 204 people tested, five males and six females (5%) were anemic.


This study reports the burden of intestinal parasite infections among a cross section of recent immigrants from East Africa and longer term settlers from Cambodia. The surveys were conducted using a convenience sample, which is a common approach in studies of immigrant communities as they can be difficult to access. It is therefore not possible to extrapolate the results to determine the exact prevalence of intestinal parasite infections in either of the communities. However, the findings suggest a high likelihood of unrecognized intestinal parasite infections in the wider East African and Cambodian communities in Melbourne, the combined population of which is estimated to be 15,500.25 The presence of undiagnosed and potentially serious intestinal parasitic diseases such as strongyloidiasis and schistosomiasis is of particular concern.

While considered the “gold standard” for diagnosis of strongyloidiasis, the demonstrated poor sensitivity of fecal microscopy26 has led to the wide acceptance of the ELISA as a satisfactory diagnostic and screening test for strongyloidiasis, especially among asymptomatic individuals and those at high risk of being infected.5,11,27 The very high proportion of Cambodian immigrants who were seropositive for S stercoralis in this survey is consistent with surveys undertaken in people who had recently arrived from Mekong countries and resettled in Australia during the 1980s6 and with studies showing that S stercoralis has persisted for years in some immigrant groups.9,11 It is likely that there were some false‐positive results in this group as the specificity of the Strongyloides ELISA is not 100%. Some individuals may have had occult filariasis as there is documented serological cross‐reactivity28 . Filariasis is endemic in Cambodia but as filarial worms usually only persist for 4 to 6 years,29 it is unlikely to be a significant factor in these long‐term immigrants. The few positive cases observed among the East African participants is in contrast to the high number seen in the Cambodians and other Southeast Asian immigrants.5,9,11 To the best of our knowledge, this is the first Strongyloides seroepidemiology survey undertaken among adult East African immigrants resettled in Australia. The majority of immigrants are likely to have acquired strongyloidiasis before resettlement or during home visits as S stercoralis is not endemic in southern Australia.

The finding that 15% of East African immigrants had positive or equivocal Schistosoma serology is similar to reports from other countries.15,17,30 Although the presence of antibodies does not distinguish between past and recent infection, clinicians usually elect to treat immigrants because it is impossible to ascertain whether adequate treatment has been received in the past, and chronic infection can be associated with major morbidity30 . It is also possible that there were some false‐positive results. No Schistosoma ova were detected in the fecal samples of seropositive individuals. This may have been due to the low sensitivity associated with the microscopic examination of only one fecal specimen from each participant or due to low egg production/excretion in lightly infected individuals.31

Not surprisingly, the pattern of intestinal parasites detected varied according to duration of residence in Australia and the region of origin. Helminths with a relatively short life span and protozoa were more common among the recently arrived East Africans. Transmission of Giardia lamblia, Trichuris trichiura, and E histolytica to family members, especially children, could occur, and so presumptive treatment is recommended even if the infected person is asymptomatic.32,33 One limitation of this study was that only one fecal specimen was obtained due to difficulties with transport of participants and specimens. Analysis of multiple fresh specimens may have increased the yield of pathogenic parasites. This may also have explained the absence of detection of Ascaris lumbricoides and hookworms in the East Africans, although there have been anecdotal reports that some people receive treatment with mebendazole before migrating to Australia.

Presence of eosinophilia has often been used to predict intestinal helminth infections, and this is supported among the Cambodians surveyed with participants four times more likely to be seropositive for S stercoralis if they have eosinophilia. However, only half of those with S stercoralis larvae detected in feces and less than half seropositive for S stercoralis had an elevated eosinophil count. This is possibly explained by fluctuating eosinophil counts that are reported to occur at different stages of illness.34 These findings confirm earlier results, indicating that while eosinophilia is predictive of strongyloidiasis, its absence does not exclude the diagnosis.11

GPs are now largely responsible for undertaking health checks in recently arrived immigrants/refugees in Australia. Raising awareness among GPs of the most common and serious medical conditions that may occur in immigrant groups from different geographic regions is important for early diagnosis and treatment of diseases that may initially be asymptomatic. It is also important for GPs to consider occult infections in long‐term immigrants, such as strongyloidiasis and schistosomiasis that may persist for many years, as well as acute infections that may have been recently acquired during visits to their home country.

Declaration of interests

The authors state that they have no conflicts of interest.


The project was kindly supported by the Department of Medicine (RMH/WH), The University of Melbourne, The Victorian Infectious Diseases Service, The Royal Melbourne Hospital, the Victorian Infectious Diseases Reference Laboratory, and by staff at the Western Region Health Centre, Footscray and Banyule Community Health Service, West Heidelberg, and the GP clinics. We would like to thank both the Cambodian and the East African communities for their enthusiastic participation in the surveys. We are indebted to Naysang Ouk and the East African community workers Sagal Mohammed, Munira Mahmoud, Malyun Ahmed, Kinfe Abera Kahhsay, and Abdulrahman Mohammed for their assistance in mobilizing community support, interpretation, and collection of biological specimens, Jennie Leydon for serology testing, Jim Black for his assistance with data sets, and Mirella Ozols for assistance with formatting the manuscript.


View Abstract