Strongyloides, Dengue Fever, and Tuberculosis Conversions in New Zealand Police Deploying Overseas

Abstract

Background

Members of New Zealand Police (NZP) deploy overseas in a variety of roles. There is limited published data on travel‐related morbidity in police as a subgroup of travelers.

Methods

An audit of pre‐ and postdeployment medical files for all NZP personnel deploying overseas during 2004 to 2010 was undertaken. Of all deployments, 58.9% were within Oceania.

Results

Positive Strongyloides stercoralis serology was returned in 6.07% (95% CI: 3.80%–9.13%) at a rate of 9.00/1,000 person deployment months (pdm) (95% CI: 5.57–13.8). Dengue fever seroconversion was recorded in 4.91% (95% CI: 3.40%–6.83%) at a rate of 8.57/1,000 pdm (95% CI: 5.90–12.0). The relative risk of dengue infection was 7.47 for Timor Leste compared to all other deployment destinations. An association between seroconverting for both dengue fever and Strongyloides was found. Tuberculosis conversion was recorded in 1.76% (95% CI: 0.85%–3.21%) at a rate of 2.92/1,000 pmd (95% CI: 1.48–5.375). A single case of human immunodeficiency virus (HIV) seroconversion was recorded. There were no recorded hepatitis C seroconversions.

Conclusions

Police deploying overseas appear to have similar rates of dengue and tuberculosis conversion as other groups of travelers, and they appear to be at low risk of hepatitis C and HIV. Strongyloidiasis appears to be a significant risk; postdeployment prevalence was markedly higher than that reported in a small number of studies.

A number of countries, including New Zealand (NZ), deploy members of their police force overseas; as such, they are a special group of international travelers. Only one published study reporting health risks in police deployed overseas has been identified.¹ Considerably more data is published on military deployments,² which may share some similarities with police deployments.

New Zealand Police (NZP) personnel (both sworn officers and non‐sworn staff) deploy to a number of developing countries throughout the Pacific and Asia (Table 1). Roles include peace keeping, advising and mentoring local police, postconflict capacity building, and response to natural disasters.³ Length of deployment varies but is typically 6 months.

Table 1

Gender, nationality, and deployment location

	N (%)
Gender
Males	598 (80.4)
Females	146 (19.6)
Total	744
Nationality
New Zealand (NZ) nationals	663 (89.1)
Non‐NZ nationals	81 (10.9)
Total	744
Deployment location
Solomon Islands	380 (51.1)
Timor Leste	171 (23.0)
Thailand (Phuket)	107 (14.4)
Bougainville	44 (5.9)
Afghanistan	27 (3.6)
Tonga	10 (1.3)
Pitcairn	3 (0.4)
Fiji	1 (0.1)
Sudan	1 (0.1)
Total	744

As an employer, NZP has recognized that it has a duty of care to minimize health risks associated with overseas deployments; personnel undergo comprehensive pre‐ and postdeployment medical reviews including testing for human immunodeficiency virus (HIV), hepatitis C virus, dengue fever virus, tuberculosis, and Strongyloides stercoralis. The rationale to screen for these particular diseases varies with respect to risk of infection, future potential personal and public health impact, and feasibility of testing. Audit of these results will also help rationalize predeployment health preparation and in‐country anti‐infection strategies.

The soil‐transmitted helminth, S stercoralis, is widespread in the tropics and subtropics.⁴ The helminth can autoinfect facilitating ongoing infection many years post travel.⁵ Ongoing infection can cause considerable morbidity ⁵ and is a risk for disseminated disease (with high case fatality rates) in those who are immunosuppressed in the future.⁶ Personnel infected can be offered treatment to reduce these health impacts.

The impact of dengue fever virus on international travelers has been well documented, accounting for hospitalizations,⁷ outpatient consultations,⁸ and being a major cause of fever, in returning travelers.⁹ Previous infection with dengue fever virus is considered one risk factor for more severe disease with subsequent infections with different serotypes.¹⁰ Given this, a case can be made to establish past exposure before deploying to endemic areas. Screening for the infection caught while on deployment will allow returning personnel to make choices regarding future travel to dengue endemic areas.

International travel has been documented as a risk factor for infection with tuberculosis.¹¹ Early detection of infection with tuberculosis can reduce future disease through treatment of latent tuberculosis.¹²

Hepatitis C is an infection with a global distribution but with higher prevalence in many developing countries.¹³ Behavior putting travelers at risk of HIV has been well documented ¹⁴ and travel‐related HIV infections have been reported in returning travelers.¹⁵ Early detection of HIV and hepatitis C infection is likely to have a positive impact on health outcomes.

Methods

Seven years (2004–2010) of pre‐ and postdeployment medical files of NZP personnel were audited. Dengue fever, HIV, hepatitis C, and tuberculosis results were available for the full period. Three years (2007–2010) of testing for infection with S stercoralis was also available. [This was introduced after the description of a cluster of cases, including some NZP personnel, in the Regional Assistance Mission to Solomon Islands (RAMSI).] ¹ Potential participants were NZP personnel who had been overseas on official duties and returned to NZ. Any period of time spent continuously overseas was counted as one deployment.

Disease‐specific antibody serology tested for predeployment exposure to dengue fever, HIV, and hepatitis C. Baseline tuberculosis status was determined by two methods. Prior to 2007, tuberculin skin testing (TST) by way of a two‐step Mantoux was used; from 2007, this was replaced by a tuberculin interferon gamma assay, Quantiferon TB Gold (QFG).

Dengue fever seroconversion was defined as a change from negative to positive dengue immunoglobulin G (IgG). A tuberculosis conversion was defined as either a Mantoux increase of 10 mm or more or a change from a negative to positive QFG assay. Strongyloidiasis was considered positive on the basis of positive serology (IgG enzyme immunoassay).

Prevalence and comparative analysis was calculated using OpenEpi software. Conversion rates were calculated as per 1,000 person deployment months (pdm). CIs for these estimates were calculated as follows. For proportions, Fisher's exact CI was used; CIs for rates were calculated using the Byar approximation to the Poisson option; CIs for relative risks were calculated using Taylor series analysis.

Results

During the study period, a total of 649 NZP personnel undertook 744 deployments to nine countries. Destination and demographic data are summarized in Table 1. The Solomon Islands was the most common deployment destination, and the majority of those deployed (80.4%) were males.

The prevalence and rates of conversions are summarized in Table 2. Positive Strongyloides serology was returned in 21 personnel. Comparing the two larger deployment destinations, the Solomon Islands had a higher rate at 19.3/1,000 pdm (95% CI: 12.1–29.1) compared with 11.7/1,000 pdm (95% CI: 5.60–21.6) in Timor Leste [a relative risk of 1.64 (0.78–3.47)]. Personnel who seroconverted for dengue fever were 1.66 (1.15–2.32) times as likely to also have a positive or equivocal Strongyloides result (Table 3). Looking at this from another angle, the rate of Strongyloides on deployments where some returned dual positive results was 48.3/1,000 pdm (95% CI: 20.8–95.3), while the rate on deployments that recorded no dual positivity was 13.8/1,000 pmd (95% CI: 9.03–20.3).

Table 2

Conversions in New Zealand Police deploying overseas

Disease	Number converted	Prevalence (%) (95% CI)	Rate/1,000 person deployment months (pdm) (95% CI)
Strongyloides	21/346	6.07 (3.08–9.13)	9/1,000 (5.57–13.80)
Dengue fever	33/672	4.91 (3.40–6.83)	8.57/1,000 (5.90–12.0)
Tuberculosis	10/569	1.76 (0.85–3.21)	2.92/1,000 (1.48–5.38)
HIV	1	<1 (1)	—
Hepatitis C	0	0 (0)	—

Table 3

Strongyloides by destination

Deployment destination	n	Strongyloides rates (both positive and equivocal) per 1,000 pdm (95% CI)
Solomon Islands (SI)*	22	19.3 (12.1–29.2)
Timor Leste (TL)*	10	11.6 (5.6–21.6)
Afghanistan	2	24.6 (2.8–88.9)
Bougainville	3	16.7 (3.4–48.9)
Fiji	1
Pitcairn	1
Total	39	16.7 (11.9–22.84)

pdm = person deployment months.

*

Risk rate: SI : TL 1.6 (0.78–3.47).

Twelve personnel [1.98% (95% CI: 1.08–3.35)] tested positive for dengue fever prior to their first deployment. Dengue fever seroconversion was recorded in 33 (4.91%) personnel (Table 2). Personnel deploying to Timor Leste seroconverted at a rate of 23.7/1,000 pdm (95% CI: 15.19–35.28) compared to 3.20/1,000 pdm (95% CI: 1.40–6.00) in those deploying to all other countries combined. The relative risk of Timor Leste compared to all other destinations was 7.47 (3.47–16.1).

During the audit period, 63 personnel had positive baseline tuberculosis giving a predeployment prevalence of presumed latent tuberculosis of 10.38% (95% CI: 8.07–13.08). Those who gave their nationality as being a New Zealander (and therefore more likely to be NZ born) had a relative risk of 0.62 (0.33–1.17) for latent tuberculosis.

During deployment, a tuberculosis conversion was documented in 10 personnel (Table 2). Rates of conversions were higher in those deploying to the Solomon Islands compared with Timor Leste; however, this was not statistically significant (Table 4).

Table 4

Tuberculosis conversions by country of deployment

Country	Tuberculosis conversions	Total screened	Total deployment months	Rates/1,000 person deployment months	95% CI
Solomon Islands	8	307	1,998	4.01	1.72–7.89
Timor Leste	2	137	863.3	2.32	0.26–8.30
Other (Afghanistan, Bougainville)	0	125	—	—	—
Total	10	569	—	2.92	1.48–5.38

There was one HIV seroconversion and no recorded seroconversions for hepatitis C. Both had 0% predeployment prevalence.

Discussion

This is the first identified published audit of conversions for Strongyloides, dengue fever virus, tuberculosis, HIV, and hepatitis C in police deploying overseas.

While published work on travelers and strongyloidiasis has focused on two groups (immigrants from endemic countries to developed countries ¹⁶ and military veterans ⁵ ), it has been described in returning travelers in two prospective studies.^17,18 In one, 0.25% (at a rate of 3.2/1,000 person months) were found to seroconvert for S stercoralis during short‐term travel,¹⁷ and in another, 0.8% of returning travelers had a positive S stercoralis polymerase chain reaction.¹⁸ These studies suggest that strongyloidiasis is a rare disease of returning travelers. The prevalence of S stercoralis infection (6.07%) found in this audit is therefore surprisingly high. A clear explanation for this is not obvious. It is possible that NZP are deploying to areas with high prevalence (as the cluster of cases diagnosed in the Solomon Islands might indicate). Indeed, Solomon Islands‐based personnel were shown to have higher rates of Strongyloides‐positive serology, but positive results were not restricted to this destination. While the literature suggests that Strongyloides is rare in travelers, what is not clear is whether more infection would be uncovered in if it was actively sought. The results of this audit suggest that it might be a greater risk than previously thought.

Dengue infection has been recorded in up to 19.5% of a cohort of returning travelers,¹⁹ 4.3% of aid workers,²⁰ 6.6% of military deploying to East Timor,²¹ and in 7.7% of one US army unit in Somalia.²² The 4.9% (95% CI: 3.40%–6.83%) prevalence observed in our audit was of the same magnitude as that observed in these studies. The rate per 1,000 months exposed observed (8.57) is not dissimilar to that seen in Israeli travelers ²³ but is less than that described in Dutch short‐term travelers.²⁴ The baseline 1.98% positive dengue serology in our audit was similar to that found in a German study.¹⁹ Because NZ is not endemic for any human flavivirus, positive baseline dengue was assumed to represent past infection associated with previous travel to, or residency in, endemic countries or a cross‐reaction to vaccination ²⁵ against other flaviviruses. In this audit, it was observed that those who had seroconverted for dengue fever were more likely to also test positive for infection with S stercoralis. Why it is not clear, it could be explained by personal attributes (are those who are less fastidious with their insect personal protection methods also less likely to take care to avoid helminthic infections?) or environmental conditions (do conditions which favor one also favor the other?).

Higher rates of dengue conversion were noted in those deploying to Timor Leste, and while this is likely to reflect local disease patterns, it could be inflated by cross‐reactivity to vaccination against Japanese encephalitis,²⁵ which is required for deployments to Timor Leste and Thailand but not others.

The observed 1.76% of NZP personnel converting with tuberculosis compares favorably with that published in a recent systematic review.¹¹ The observed rate of 2.9/1,000 pdm is more than that observed in Peace Corps Volunteers ²⁶ but very similar to long‐term travelers from Holland.²⁷

Of interest was the amount of latent tuberculosis uncovered by baseline testing. Comprehensive data and an accurate incidence of latent tuberculosis in the NZ population are lacking ²⁸ ; therefore, it is not clear if the 10.4% measured in this group is typical of the wider NZ population.

Data were not always complete. Despite a policy of having NZP personnel likely to deploy overseas in a constant state of readiness, it has not always been possible to predict exactly who will need to deploy at short notice. The test most commonly missed predeployment was the two‐step Mantoux as this takes a minimum of 9 days to complete. Postdeployment data were not always complete; 47 (6.3%) of all personnel failed to complete all or some postdeployment testing. The test most commonly lost to follow‐up was the postdeployment Mantoux; this improved with the introduction of QFG in 2007.

If personnel had incomplete testing, their data were excluded from analysis. It is not known if those who did not have time for full predeparture testing or failed to complete postdeparture testing differed from those who did.

An overestimation of strongyloidiasis is possible as no baseline testing was done. The rationale is that NZ is considered non‐endemic for S stercoralis²⁹ with the only published case reports of strongyloidiasis in New Zealanders being in persons born and traveling outside NZ.^30,31 It is possible, however, that NZP personnel might have been exposed due to prior travel to, or residence in, endemic countries. Also, in this audit, screening was based on serology alone. For many years, isolation of the larva from fecal samples was considered the “gold standard” of diagnosis, but techniques are difficult ¹⁸ and some studies have shown low sensitivity.³² While serological tests have been quoted to have high levels of both specificity and sensitivity,¹⁷ low sensitivity has been described in travelers.³³ It would appear that the diagnosis of S stercoralis infection, especially in returning travelers where worm burden might be low, is not perfect. After discussion with local laboratories, the consensus was that, given the limitations of larval isolation, diagnosis would be made on serology alone and this might, in part, explain the high prevalence found.

Screening tools for tuberculosis infection are limited. Both tuberculin skin tests and the newer tuberculin gamma interferon assays have their limitations. TST can give false positives due to previous Bacillus Calmette‐Guerin (BCG) vaccination, previous exposure to non‐human mycobacteria, the boosting effect of serial tests, and readings are subjective.³⁴ While there is support for the substitution of tuberculin gamma interferon assays where TST has been traditionally used,³⁵ some uncertainty remains around their sensitivity, specificity, and positive predictive value.³⁶ Because many NZP personnel have received BCG vaccination as children and because pre‐ and postdeployment Mantoux testing was the cause of most incomplete testing, it was decided that, despite limitations, QFG should be the preferred test once it became available in NZ. It is recognized that both forms of testing may result in false positives causing overestimation of the prevalence of both latent tuberculosis predeployment and infections during deployment.

Conclusions

NZP personnel deploying overseas are at risk of travel‐related infectious diseases. This audit revealed positive Strongyloides serology, dengue seroconversions, and tuberculosis conversions during deployments, all of which have future health implications. Some destinations appear to carry a greater risk for specific diseases than others; in particular, deployment to Timor Leste carried a significantly higher risk for dengue infection. A positive association between Strongyloides and dengue fever was observed.

While not all risk can be fully mitigated, predeployment training and in‐country strategies should continue to focus on avoidance of insect‐ and soilborne diseases. This should include personal protection measures (including insect proofing of work and living quarters and use of repellents and permethrin‐impregnated clothing) and avoidance of skin contact with potentially fecally contaminated soil. Future study should also focus on measuring the effectiveness of these interventions.

It would also seem reasonable to continue to screen for these infections postdeployment so that future health risks can be reduced, for example, by offering treatment for latent tuberculosis.

While the prevalence of dengue and tuberculosis was of the same magnitude described in other travelers, the higher than expected prevalence of S stercoralis infection (and a positive association with dengue conversion) was surprising. Further study, including optimal testing for strongyloidiasis in returning travelers, is warranted.

This audit was made possible due to sponsorship by the Wellington Medical Research Foundation (Inc) of a University of Otago summer studentship. Ethics approval was granted internally by the University of Otago.

Declaration of Interests

The authors state that they have no conflicts of interest to declare.

Some of the data presented here was also presented as posters at the 12th Conference of the International Society of Travel Medicine Boston 2011. [Poster PO06.01 Tuberculosis conversion in New Zealand Police personnel deploying overseas: A retrospective review. And PO06.02 Strongyloidiasis Incidence (New Zealand Police International Services Group)—A Three Year Retrospective Review.] Abstracts available at http://www.istm.org/Documents/Members/MemberActivities/Meetings/Congresses/cistm12/CISTM12-Poster-Abstracts.pdf.