|
|
||||||||
| ABSTRACT |
|
|
|---|
| INTRODUCTION |
|
|
|---|
Recent study has focused on empirical treatment of intestinal parasites and the use of broad-spectrum anti-helminth medications, particularly albendazole.6,7 Treatment of helminths improves growth and nutritional status among children,8 an important issue for refugees. Studies have found albendazole to be highly effective in a single 400-mg oral dose for treatment of most helminth infections while possibly requiring multiple doses for Trichuris.9,10 In regard to protozoans, while a single dose of albendazole does not appear to be sufficient for treatment of Giardia,11,12 it does appear to inhibit growth on Entamoeba histolytica; although this did not effect the organisms viability.13,14 It has also been suggested that empirical treatment of immigrants in the United States with albendazole would be a cost-effective public health strategy.15
Within one year prior to departing for the United States, refugees undergo a medical screening evaluation. In 1997, the U.S. Centers for Disease Control and Prevention (CDC) implemented an enhanced medical assessment for a sub-population of Somalian refugees in Kenya.16,17 After the enhanced assessment of 390 Barawan Somalians indicated that 38% had intestinal parasites, the CDC implemented treatment of all non-pregnant Barawan women and other Barawan refugees more than two years of age with albendazole, 600 mg in a single oral dose, prior to departure for the United States.16,17 Subsequently, in May 1999, the CDC began implementation of empirical treatment of all refugees departing from other sites throughout sub-Saharan Africa.
Once in the United States, many refugees undergo another medical screening. In mid-1995, refugee medical screening in Massachusetts was consolidated as the Refugee Health Assessment Program (RHAP), under the auspices of the Massachusetts Department of Public Health. RHAP contracted a limited network of clinical sites to deliver clinical services. While content of the health assessment varies greatly among states, the Massachusetts RHAP is among the more comprehensive and includes microscopic evaluation of a single stool specimen for ova and parasites.
| MATERIALS AND METHODS |
|
|
|---|
400.107. This analysis and public release of its findings were reviewed and approved by the Bureau of Communicable Disease Control, Massachusetts Department of Public Health. Overseas treatment protocol. In Kenya, departing refugees received their treatment from one of two sources: a single physician contracted to perform these medical assessments or the medical facilities of the International Organization for Migration (IOM). The IOM also coordinated the medical assessments of all other refugees departing from elsewhere in Sub-Saharan Africa. IOM data on the treatment program are available for October 1, 1999 through September 30, 2001. The data illustrate the extent of the program in sub-Saharan African during most of the time period studied. During these two years, a total of 28,406 refugees departed for resettlement sites anywhere in the United States, including Massachusetts. The IOM confirmed albendazole treatment of 20,865 (81%) African refugees. Data specific for those destined for Massachusetts are not available. This treatment was usually given as the refugees departed from Africa.
An additional 7,074 either had unconfirmed treatment by non-IOM physicians (mostly the single physician in Kenya) or departed from areas in which the IOM had not yet implemented the program; however, while the IOM could not confirm receipt of treatment of patients seen by this Kenyan physician, the physician has attested to the CDC that refugees evaluated in his clinic were also treated within three days of departure. From data available only for the second year of this period, 82% of this category had unconfirmed treatment by non-IOM physicians. The remainder typically departed from areas in which the IOM had not yet implemented the program. Lastly, 467 were pregnant women and children less than two years of age. (Cookson S, Division of Global Migration and Quarantine, CDC, July 26, 2001 and Kimata C, International Organization for Migration, June 20, 2003, unpublished data).
Dataset construction and group assignment. The RHAP data were used to create a dataset of all African refugees resettled in Massachusetts during the study period. For this study, the dataset was retrospectively analyzed to describe the prevalence of intestinal parasitoses among African refugee arrivals. The dataset was subdivided between those who arrived prior to and those who arrived on or after May 1, 1999, the date when the CDC implemented the universal pre-embarkation albendazole treatment program. Refugees in the former group, as well as all children less than two years of age, were designated as the no-albendazole group. All other refugees were designated as the albendazole group. While pregnant women also did not receive albendazole, the RHAP database cannot reliably identify women who were pregnant overseas; however, while not a randomly selected sample, pregnant women should have been evenly distributed before and after the implementation date. Other variables included age, gender, country of departure, original nationality, and parasites detected. Parasites were defined as helminth or protozoan. Diagnoses of strongyloidiasis or schistosomiasis made by serology were not included as these may not have indicated active, acute infections.
Because of the large numbers of refugees departing from Kenya and the relative homogeneity of that refugee population (mostly Somalian during the time studied), refugees departing from Kenya were analyzed separately. Only a very limited number of the Barawan Somalians resettled in Massachusetts, so their receipt of albendazole prior to May 1999 did not necessitate alteration of the classification of other refugees departing from Kenya. Somalians who did not have a prior country specified in the database were presumed to have departed from Kenya. Refugees who did not have a country of departure listed but were not Somalian, were presumed to have departed from other African countries. Lastly, 41 African refugees departed from countries outside of Africa. This number included 30 Somalians who departed from Pakistan. Because of their limited numbers and uncertain duration of stay outside of Africa, these refugees were not included in the analyses.
Analytic methods. Analyses included descriptive statistics of the whole sample as well as the subsets of Kenyan departures and those from other African countries. Bivariate analyses include t tests and chi-square analyses comparing Kenyan departures to other African departures as well as comparing the albendazole group to the no-albendazole group. Dependent variables included detection of any parasites, helminths, protozoans, and specific parasite species in the stool. Lastly, multivariate logistic regression analyses were used to control for Kenyan departure, age, and sex in assessing the impact of albendazole treatment on the detection of parasites in the stool specimen. Because of the direct relationship between Kenyan departure and uncertain timing of or likelihood of receipt of albendazole, as well as the fact that infants and toddlers were not treated, regression models initially included interactions terms for age with albendazole and Kenyan departure with albendazole; the former term was not statistically significant and was dropped from subsequent analyses. Subsequently, because of the possible interaction between departure origin and albendazole treatment, separate regression models were run also on data subsets of Kenyan departures and other African departures.
| RESULTS |
|
|
|---|
Of the 1,254 with stool results, 789 (63%) refugees departed from Kenya, and 465 from elsewhere in Africa. The sample was 45% female. In addition, 47% were children or adolescents less than 18 years of age. The mean age was 22 years. Forty-nine percent of the sample should have received albendazole (Table 1
and Figure 1
). No differences in demographics were noted between those who departed from Kenya compared with other African countries and those who received albendazole compared with those who did not. These demographics also did not differ from those of the overall group of 1,738 African arrivals.
|
|
|
After assigning children less than two years of age to the no-albendazole group, women and children were still equally likely to be in both the albendazole and no-albendazole groups; however, Kenyan departures were far less likely to be in the albendazole group (odds ratio [OR] = 0.17, 95% confidence interval [CI] = 0.130.22). Of the 43 children less than two years of age, 15 (35%) had a positive stool. In addition, Kenyan departures were more likely than other African departures to have a positive stool (OR = 1.82, 95% CI = 1.442.31). Kenyan departures were also 4.1 times as likely to have helminths (95% CI = 2.556.51) and 1.6 times more likely to have protozoans (95% CI = 1.272.05) than other African departures. This pattern was seen in both adults and children (Table 2
).
Among adults, 62% of the no-albendazole group compared with only 39% of the albendazole group had a positive stool (OR = 0.39, 95% CI = 0.290.55). Those with helminths decreased from 21% to 3% (OR = 0.10, 95% CI = 0.05-0.22) and those with protozoans from 55% to 38% (OR = 0.50, 95% CI = 0.360.69). Among children, 66% of the no-albendazole group compared with 58% of the albendazole group had a positive stool (OR = 0.69, 95% CI = 0.490.98). Those with helminths decreased from 27% to 4% (OR = 0.12, 95% CI = 0.060.24); however the prevalence of protozoans was virtually unchanged at 58% and 57%, respectively, in the no-albendazole and albendazole groups.
Mixed infections of both helminths and protozoans were seen in 120 refugees (10%). Of these, 73 (61%) were Kenyan departures. In analyses stratified by departure origin, albendazole treatment did not affect the likelihood of having a mixed infection. Of those with mixed infections, only 13% had multiple helminths, while 50% had multiple protozoans. The most common helminths in mixed infections included Trichuris (79, 66%) and Hymenolepis nana (28, 23%). The most common protozoans included B. hominis (82, 68%), Endolimax nana (43, 36%), and Giardia (27, 23%). Overall, refugees with the three most common non-pathogenic protozoans were more likely than other refugees to have helminths (B. hominis OR = 2.0, 95% CI = 1.462.79, E. nana OR = 1.9, 95% CI = 1.262.71, Entamoeba coli OR = 2.0, 95% CI = 1.263.28). Those with Giardia, the most common pathogenic protozoan, were no more likely than other refugees to also have helminthes. Overall, of refugees who had any of the three most common non-pathogenic protozoans, 20% also had helminth infections.
In multivariable logistic regression analyses, the strong affect of albendazole treatment on reduced likelihood of having parasites was maintained. In models controlling for age and Kenyan departure, those refugees treated with albendazole were less likely than others to have a positive stool (OR = 0.61, 95% CI = 0.470.78). Albendazole was also associated with reduced helminths (OR = 0.15, 95% CI = 0.090.24) and Trichuris (OR = 0.05, 95% CI = 0.020.13). It was not significantly associated with reduced risk of protozoans (OR = 0.79, 95% CI = 0.621.01) or Giardia (OR = 1.26, 95% CI = 0.871.82) (Table 3
).
|
| DISCUSSION |
|
|
|---|
Recent cost-effectiveness analysis projected significant cost savings of empirical treatment of immigrants in the United States with albendazole.15 This work did not focus on immigrants from countries with endemic helminths, and it also used a longer dosing regimen of 400 mg orally for five days. It is reasonable to expect that empirical treatment of refugees from endemic countries, such as those in Africa, with a single oral dose should be even more cost effective, although given the reduced prevalences of protozoan infections, use among refugees from other countries with greater proportions of protozoan infections, such as the countries of eastern Europe, might also be considered only if supported by further research.
A significant limitation of this study is the inability to track individual refugees directly. As such, we cannot account for the impact of changing demographics within refugee populations. For example, earlier cohorts of Somalians tended to depart from coastal refugee camps in Kenya, while later cohorts tended to depart from urban areas such as Nairobi. This also relates to our inability to identify those pregnant refugees who were not treated. In addition, the timing of and adherence to treatment of Kenyan departures is unclear. While those screened by the IOM had clear documentation of treatment at the time of departure, others may have been treated up to one year prior to departure. This would seem likely to reduce the effectiveness of the intervention if refugees continued to live in an area with endemic parasites. Given the more pronounced reduction of parasitoses among other African departures, it would seem that treatment upon embarkation as practiced by the IOM is the most effective strategy.
Since those who departed from other African countries were almost exclusively screened by the IOM and treated at departure, by analyzing other African departures separately and controlling for Kenyan departure in multivariate analyses, the study more accurately was able to define the effectiveness of the albendazole treatment. Lastly, while shifting demographics (e.g., refugee camps versus urban environments in Kenya) may play some role in changing prevalences, particularly of the protozoans, the study assumes that the level of endemicity of parasites in the countries of departure is relatively static over time. However, it is doubtful that any reductions of endemicity in the last two years would have matched the dramatic reductions in the prevalences of parasitoses observed after initiation of the pre-embarkation albendazole treatment program. These reductions even exceed those that would be predicted based on published research from west Africa.9
Although not unexpected, one notable finding is the higher prevalence of parasitoses among children. Throughout multivariate analyses, each annual increment of age was associated with a 1.58% reduction in risk of the various parasite outcomes. The exclusion of treatment of children less than two years of age does not seem to explain this finding. Adherence to treatment could partly explain this. However, the albendazole formulations used by the IOM include a suspension of 4% (Cookson S, January 18, 2002, unpublished data). Thus, adherence to the treatment regimen should have been reasonably good among young children. It is also possible that because of poorer personal hygiene and other exposure issues, children may have had higher baseline parasite loads and reinfection rates than adults. In regards to the assignment of children less than two years of age to the no-albendazole group, we do not believe that this inflated the prevalence of parasitoses in that group since the prevalence of parasitoses in these young children was actually lower than that of both groups overall.
Finally, this study was not a prospective scientific experiment. Rather, it was a retrospective epidemiologic review of public health screening data. The reliance on a single stool specimen may have also contributed to under-diagnosing of infections. Consequently, scientific limitations (such as lack of standardization of laboratory techniques, ability to control for technical variations in analysis, or changing laboratory personnel and skill levels) are inherent. In addition, surveillance data are devoid of clinical content, such as to what extent the refugees were symptomatic at the time of testing and to what extent health morbidity might be reduced. We believe that despite these limitations, the markedly decreased prevalences of helminth infections after implementation of the program speaks to the adequacy of the study methods.
In conclusion, the pre-departure empirical treatment program appears to be a highly successful strategy for the reduction of intestinal parasites, and potentially their associated morbidity, among African refugees bound for the United States. It is likely that other immigrants from similar countries in Africa might also benefit from such empirical treatment. Further applications and adaptations of this strategy, particularly ones that focus on the reduction of protozoan infections, appear warranted but should be implemented with prospective evaluation of their effectiveness. Until more universally effective strategies are developed, however, domestic testing for parasites among refugees should continue.
Received April 7, 2003. Accepted for publication July 1, 2003.
Acknowledgments: We thank Dr. Susan Cookson (Division of Global Migration and Quarantine, Centers for Disease Control and Prevention, Atlanta, GA) for her provision of information about the overseas treatment program and assistance in interpretation of the study findings and the staff of the Migration Health Services, International Organization for Migration in Africa.
Authors addresses: Paul L. Geltman and Jennifer Cochran, Refugee and Immigrant Health Program, Massachusetts Department of Public Health, 305 South Street, Jamaica Plain, MA 02130, Telephone (Geltman): 617-983-6593, (Cochran): 617-983-6596, Fax: 617-983-6597, E-mail: paul.geltman{at}state.ma.us. Cressida Hedgecock, Boston Public Health Commission AIDS Program, 774 Albany Street, Boston, MA 02118, Telephone: 617-534-4559, Fax: 617-534-2480.
| REFERENCES |
|
|
|---|
This article has been cited by other articles:
![]() |
E. Meltzer, R. Percik, J. Shatzkes, Y. Sidi, and E. Schwartz Eosinophilia Among Returning Travelers: A Practical Approach Am J Trop Med Hyg, May 1, 2008; 78(5): 702 - 709. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. J. Shah, S. A. Maloney, Y. Liu, E. W. Flagg, S. P. Johnston, S. A. Young, R. Weston, S. Merritt, P. P. Wilkins, V. Keane, et al. Evaluation of the Impact of Overseas Pre-departure Treatment for Infection with Intestinal Parasites among Montagnard Refugees Migrating from Cambodia to North Carolina Am J Trop Med Hyg, May 1, 2008; 78(5): 754 - 759. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Franco-Paredes, R. Dismukes, D. Nicolls, A. Hidron, K. Workowski, A. Rodriguez-Morales, M. Wilson, D. Jones, P. Manyang, and P. Kozarsky Persistent and Untreated Tropical Infectious Diseases Among Sudanese Refugees in the United States Am J Trop Med Hyg, October 1, 2007; 77(4): 633 - 635. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. K. GARG, S. PERRY, M. DORN, L. HARDCASTLE, and J. PARSONNET RISK OF INTESTINAL HELMINTH AND PROTOZOAN INFECTION IN A REFUGEE POPULATION Am J Trop Med Hyg, August 1, 2005; 73(2): 386 - 391. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. EKDAHL and Y. ANDERSSON IMPORTED GIARDIASIS: IMPACT OF INTERNATIONAL TRAVEL, IMMIGRATION, AND ADOPTION Am J Trop Med Hyg, June 1, 2005; 72(6): 825 - 830. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. L. Geltman and J. Cochran A Private-Sector Preferred Provider Network Model for Public Health Screening of Newly Resettled Refugees Am J Public Health, February 1, 2005; 95(2): 196 - 199. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |