Ineffectiveness of the Current Strategy to
Prevent Hepatitis A in Travelers
Gaston De Serres, Bernard Dtrval, Ramak Shadmani, Nicole Boulianne, Gina Pohani,
Monika Naus, Monique Dotrville Fradet, Louis Rochette, Brian]. Ward,
a d Kevin C. Kain
Background: Each year, a large number of Canadians travel t o regions of the world where hepatitis A remains endemic.
Many o f these travelers are not immune and the current preventive strategy relies wholly on self-referral t o a travel clinic.
All of the costs associated with such a visit are assumed by the traveler. We estimated the effectiveness of this strategy.
Methods: This case-control study included 108 travel-related hepatitis A cases with onset of disease between 1997 and
1999 and 620 controls who traveled during the same period.
Results: Hepatitis A was strongly associated with high-risk travel (Odds Ratio = 7.2,95% Confidence Interval 1.76-29.4).
but only 7% of cases were found i n this category. The risk of hepatitis A was 5 times lower i n travelers w h o visited a travel
clinic than in those who did not (80% efficacy). However, only 14% of the controls visited a travel clinic. As a result, the
effectiveness of the current strategy is estimated t o be 11% (80% of 14%).
Conclusions: Hepatitis A in travelers can be prevented effectively by attendance at a travel clinic. Unfortunately, most
travelers do not visit such clinics prior t o departure. Even if ell high-risk travelers were t o visit a travel clinic and receive
vaccination, this would have negligible impact on the number of travel-related hepatitis A cases (-7% reduction). The
current strategy for the prevention of hepatitis A in travelers is ineffective and should be reexamined.
Hepatitis A (HA) is the most common vaccine preventable disease of travelers’,2and expert committees recommend vaccination of all susceptible people traveling
to an endemic country."^^ Individuals who travel for
prolonged periods in poor hygienic conditions (e.g.,
backpackers) are recognized to have an elevated risk of
HA.’ O n the other hand, it is widely believed by travelers and caregivers ahke, that short-term (<2 weeks) and
higher budget travel (e.g., destination resorts, tours) presents little or no risk for the acquisition of HA. However, these latter tourists are also at risk of acquiring HA.’
As short-term tourists make up the large majority of travelers to endemic countries, they may account for the
majority of travel-associated HA cases.
In the two most populous Canadian provinces,
Quebec and Ontario, travel-related HA accounts for up
to one-third of all reported HA cases.6 As in the United
States, the current strategy in Canada for the prevention
of travel-related HA is to immunize travelers bound for
endemic countries in either self-designated “travel clinics” or in family physician offices or clinics. The effectiveness of this strategy has never been evaluated. In the
current study, we reviewed attendance at travel clinics in
order to estimate the proportion of HA cases that are prevented by this strategy.
Gaston 08 Serres, MD, Bernard Duval, MD, and Nicole
Boulienne, MSc: lnstitut National de Sante Publique du
Quebec, Quebec and Public Health Research Unit, CHUL
Research Center, Leva1 University, Quebec; Ramak
Shadmani, MD, and Louis Rochette, MSc: Public Health
Research Unit, CHUL Research Center, Lava1 University,
Quebec; Gina Pohani, MSc, and Monika Naus, MD: Ontario
Ministry of Health and Long-Term Care, Toronto; Monique
Douville Fredet, MD: lnstitut National de Sante Publique du
Quebec, Quebec and Ministere de la Sante et des Services
sociaux du Quebec, Quebec; Brian J. Ward, MD: McGill
Centre for Tropical Diseases, Montreal General Hospital,
Montreal; Kevin C. Kain, MD: Centre for Travel and Tropical
Medicine, University of Toronto, Toronto, Canada.
Part of this study was presented at the 4th Canadian
National Immunization Conference in Halifax, December
2000 and another part a t the 4th Annual Conference on
Vaccine Research, Washington DC, April 23-24,2001. This
study was funded by an unrestricted grant from Glaxo
SmithKline Beecham Pharma.
Methods
The authors had no financial or other conflicts of interest to
disclose.
Cases
All laboratory-confirmed HA cases reported in
adults (218 years old) in Quebec and Ontario with
onset between January 1997 and November 1999 who
had traveled to an HA endemic country (Appendix) in
the 6 weeks preceding the onset of their disease were
Reprint requests: Or. Gaston De Serres, lnstitut National de
Sent6 Publique du Quebec, 2400 d‘Estimauville, Beauport,
Quebec, Canada, G1E 7G9.
J Travel Med2002; 9:lO-16.
10
D e S e r r e s et a l . , H e p a t i t i s A i n T r a v e l e r s
eligible. Cases were interviewed by telephone by regional
public health department staff using a standardized questionnaire between November 1999 and May 2000.
Cases were asked to supply demographic information
(age, gender, country of birth, education level, household income), and the number of trips they had made
to endemic countries since 1990. The countries’ name(s)
as well as date and duration for each of these trips were
recorded. For the three most recent trips, additional
information was collected: purpose of travel (tourism,
business, volunteer work), organization of trip (fully
organized such as all inclusive tours, partly organized,
self-organized),self-assessed type of accommodations (first
class, middle class, low budget or camping, visiting
friends or relatives) and proportion of time spent in each
type, self-assessed type of food facilities (first class, middle class, low budget or camping, friends or relatives) and
proportion of meals taken in each setting, occurrence
of gastroenteritis, a self-evaluation of their personal attitude toward hygiene while traveling, on a scale ranging
fi-om 1 (not at all careful) to 10 (extremely careful), awareness of health risks preventable by vaccination, attendance
at a travel clinic before departure and, for those who did
not attend, the reasons for this decision. Vaccination data
was not collected.
Controls
To obtain controls, a phone survey was conducted
among adults (2 18 years old) residing in Quibec and
Ontario selected by random digit dial. The sample was
proportional to the population weight of each province
and to that of their respective metropolitan areas (Montrial and Toronto). Households in which there was an
adult who could communicate in either English or
French were eligible. Respondents were randoinly
selected fi-om within each household. Initially, 12,067 telephone numbers were generated. Of these, 3,328 were not
in use, 1,357 were commercial telephone numbers, 986
did not answer despite 6-10 attempts at different times,
and 533 were not eligible. Among the 5,863 eligible
households, 4,002 (68%) agreed to participate. The participation rates were 64% and 76% in Ontario and
Quibec, respectively. All participants who had taken at
least one trip to a HA endemic country since January
1997 (N = 620) were used as controls. Only the last three
trips of these individuals were used and the data fi-oni these
control trips were obtained with the same questionnaire as the cases.
Statistical Analyses
Prior to statistical analyses, trips were categorized into
four levels of risk as follows: “high-risk” travel was
defined as that lasting 2 4 weeks with 2 50%)of the nights
spent in low budget hotels, “low-risk’’ included travel for
11
2 weeks or less with all nights spent in first class hotels.
The “intermediate-risk” group included all other travels except for a fourth category “visiting friends and relatives” (VFR) when more than 50% of the nights had
been spent in the homes of friends or relatives. The
proportion of meals eaten in each type of food facility
was closely correlated with the type of acconimodation,
thus only this last variable was used in the risk group definition. The purpose of the trip was classified into the
following categories: “tourism” included those trips
which were entirely for leisure, “business” and “volunteer work” included trips for business or volunteer work,
respectively, even if some time had also been devoted to
leisure.
Multivariate analyses were performed by logistic
regression. The niodel included variables sipificandy associated with the risk of acquiring HA in the univariate
analyses. As risk levels were defined by two variables (duntion of stay in endemic country and type of acconiniodation), these two factors were not included in the
analyses where risk levels were used.
We estimated the incidence of HA in travelers in the
most frequently visited countries. The population-time
of travel was estiiiiated by multiplying the sun1 of days
of travel spent in each country by the sanipling factor
(1 5,351.1 18 adult population of Quibec and Ontario
[Statistics Canada, Web site <www.statcan.ca>,
lYYY]/4,002 participants = 3.81 1). The incidence was
estimated by dividing the number of cases acquired in
each country by this denominator.
The eficacy of a travel clinic visit to protect against
HA was estimated as (1-Odds Ratio [OR]) X 100‘%
where OR is the odds ratio of H A for travelers who did
not visit a travel clinic versus those who did. The effectiveness of a travel clinic visit was determined by multiplying the calculated eficacy by the use of travel clinics
among control travels. Proportions were coiiipared using
the two-tailed Pearson x’ test.
Results
Between January 1YY7 and November 1999,208
cases of HA were reported in Quibec (n = 98) and
Ontario (n = 110) in adult travelers who had returned
from travel in an endemic area during the 6 weeks prior
to the onset of their disease. Interviews were completed
for 108 of these cases, for a participation rate of 62% and
43% in QuCbec and Ontario, respectively. IXefusals were
rare (n = 8) but it was not possible to contact 92 cases
in spite ofniultiple calls (QuCbec; n = 34[35%). Ontario;
n = 58[53%)]).
Among the 4.(w)2 adults who participated in the population survey, 620 had traveled to a HA endemic country since January 1YY7 and were used as controls. The
Journal o f Travel Medicine, Volume 9 . Number 1
12
Table 1 Characteristics of Cases and Controls
Gender
Male
Female
Born in
Canada
HA endemic country
Nonendemic country
for HA (excluding
Canada)
Province
Qukbec
Ontario
Age group (years)
5 24
25-34
35-44
45-54
554
Education level (years)
< 12
12
13-1 5
110
Household income
($/year)*
< 20.000
2O,OOO-3Y ,999
40,000-59,999
cases
N = 108
n (%)
Controls
N = 620
54 (50)
54 (50)
300 (48)
320 (52)
79 (73)
19 (1X)
430 (69)
119 (19)
10 (9)
71 (11)
61 (56)
47 (44)
202 (33)
418 (67)
29 (27)
30 (2X)
26 (24)
18 (17)
79 (13)
136 (23)
145 (24)
123 (20)
1 1x (20)
5 (5)
(%)
40 (37)
51 (48)
62 (10)
113 (18)
164 (26)
277 (45)
12 (16)
16 (21)
13 (17)
36 (47)
45 (9)
102 (20)
135 (26)
239 (46)
49 (45)
27 (25)
19 (18)
13 (12)
416 (67)
120 (19)
59 (10)
25 (4)
8 (7)
8 (7)
2 60,000
Number of trips since 19%)
1
2
3
4+
N
P-WlUF
.8
.96
.0001
,001
.02
.004
,001
‘Thirty-oiic case\ .lnd 90 control\ dld not .InswcT this question.
controls had made 962 trips during the 1997-1999 study
period and detailed data were available for 905 of them
(i.e., the last three trips).
Cases and controls were similar in terms of gender
and country of birth (Table 1). Cases resided more frequently in Quibec, were younger, had a higher education level and lower household income, and had taken
a greater number of trips to endemic countries between
1990-1 999 (see Table 1).
Countries in the Americas (Mexico, the Caribbean,
Central and South America) were the most frequently
visited by both cases (69%) and controls (73%).However,
cases traveled to Africa twice as frequently as controls (I(’%,
vs. 6%) (Table 2). The four most visited countries were
Mexico, Ilominican ILepublic, Cuba, and Jamaica. The
risk of HA however, was low in Cuba and HA was not
reported after travel to Jamaica (see Table 2). India and
the Philippines were not frequently visited but were
among the five countries that accounted for the largest
number of cases (Mexico, Dominican Republic, Cuba,
India, Philippines).
The incidence of HA in travelers per month of stay
was calculated for each continent and for the most frequently visited countries. The overall incidence rate was
5.0 X IO-’/month. Incidence rates per 100,000 months
of travel were highest in travelers to Central/South
America and Africa (5.8 and 5.0, respectively) and lowest among travelers to Eastern Europe (2.0).Among the
five most visited countries (see Table 2), Mexico had the
highest incidence rate (11) whereas the lowest rates were
observed in Cuba (4.6)and Jamaica (no cases reported).
The risk of HA was greater in high-risk travelers
and those visiting friends and relatives (VFR) (Table 3).
Despite the fact that the proportion of trips classified as
high-risk was 3 times greater among cases than controls,
only 7% of HA cases were observed in this risk category. The majority (63%)of HA cases occurred in the
low and intermediate risk categories. Even low risk
travel involved a substantial risk of HA as evidenced by
the fact that this category accounted for 25% of the cases
(see Table 3). Cases were less aware than controls of the
health risks associated with travel (41% vs. 59%). The proportion of travelers who visited a travel clinic prior to
departure was approximately 3 times lower in cases than
in controls (5%vs. 14%, p < .0001).Compared to controls, cases traveled for longer periods of time (see Table
3). Cases had taken self-organized trips more frequently
than controls (62%vs. 46%).There was no difference
between cases and controls with respect to the purpose
of travel (p = .86).The self-reported score of personal
attitude toward hygiene was lower in cases than in controls (see Table 3).
Compared to controls, cases stayed less frequently in
first class hotels (Table 4). The proportion of cases who
stayed exclusively in a first class hotel was 1.6 times
lower than in controls. The propomon of cases who spent
2 50% of their nights in low budget hotels was 2.6 times
higher than in controls (13%vs. 5%). In contrast, the pmportion of travelers who spent 250% of their nights in
the homes of fiiends or relatives was only slightly higher
among cases than in controls (29% VS. 23%). The trends
for the use of food facilities followed similar patterns to
those of housing (see Table 4). The reported occurrence of gastroenteritis was 3 times more fiequent in cases
than in controls (40% vs. 18% p = .001).
In the multivariate analyses, a significantly increased
risk of HA was found only with the following factors:
high-risk travel, lack of awareness of health risks, nonattendance at a travel clinic, province of residence, type of
travel, younger age, household income, and destination
(Table 5 ) .The adjusted risk of HA for high-risk trips was
De Serres et al., Hepatitis A in Travelers
13
Table 2 Travel According to Destination
N
Continent+
Americas
Africa
Asia
Eastern Europe
Country+
Mexico
Dominican Republic
Cuba
Jamaica
India
Philippines
Care Trips
= 108 (%)
Irrcidrnce per 100,000
Motrtks of Travel
Control Trips*
= 905 (%)
Moritks ofTravel
.jiv Cotitrols
74 (69)
1 1 (10)
19 (18)
4 (4)
631 (73)
48 (6)
123 (14)
67 (8)
333.0
57.4
134.5
52.4
5.8
5.0
3.7
2.0
35 (32)
12 (11)
5 (5)
0 (0)
4 (4)
5 (5)
230 (25)
69 (8)
89 (10)
35 (4)
21 (2)
13 (1)
83.3
34.8
28.7
13.8
37.8
10.5
11.0
N
9.0
4.6
0
2.8
12.5
'Among participants,620 traveled to an endemic country since January 1997 They had 962 trips during that period and detailed data wa\ availahle for
905 of them that were used as control trips. Of these for 36 control-trips destination WAS unknown.
+p-value < ,005.
7.2 times greater than for low-risk trips. Awareness o f
health risks reduced the risk of acquiring H A by two-thirds
and visiting a travel clinic reduced it by 80%.The risk o f
acquiring H A during travel was 4 times greater for people living in Quebec. T h e risk of H A for travelers w h o
had taken a fully organized trip was half that ofother travelers. T h e risk of H A among travelers aged I24 years old
Table 3 Characteristics of Travels
N
Risk levels
High-risk
Casr Trips
= 108 (Yo)
Control Trips
= 905 (!!)
N
8 (7)
41 (38)
27 (25)
18 (2)
343 (38)
341 (38)
and relatives (VFR) 32 (30)
203 (22)
533 (59)
128 (14)
Intermediate
Low-risk
Visiting friends
Aware of risk
44 (41)
Visited a travel clinic
5 (5)
Duration of stay (days)
<8
30 (28)
p - Value
,002
.0001
.002
8-1 4
15+
Type of travel
27 (25)
51 (47)
368 (41)
263 (29)
272 (30)
,001
Fully organized
Partly organized
Self-organized
Purpose of travel
Tourism
Business
Voluntary work
26 (25)
13 (13)
64 (62)
389 (43)
101 (11)
410 (46)
,002
88 (83)
16 (15)
2 (2)
756 (84)
121 (13)
26 (3)
.86
18 (17)
30 (28)
59 (55)
149 (17)
165 (18)
583 (65)
,049
Personal attitude
toward hygiene
1-4 (less cautious)
5-7
8-10 (cautious)
was 5 times greater than those aged 2 50. T h e risk of H A
in travelers in the highest self-reported income group (2
$60,000/year) was 2.4 times greater than those w h o
earned $40,000-$59,999/year. Travelers to Asia o r the
Americas had 3 and 2.3 times greater risk, respectively,
than that observed among travelers t o Eastern Europe.
Efficacy and Effectiveness of a Travel Clinic to Prevent HA
T h e adjusted OR o f those w h o visited a travel
clinic compared with those w h o did not was 0.2 (see
Table 5 ) .From this adjusted OR, the efficacy o f a travel
clinic visit was estimated at 80%) ([l-0.201 X 1001%1
=
80%).Given that the use oftravel clinics among controls
was only 14% (see Table 3 ) ,the effectiveness in the prevention o f H A in travelers was 1 1 % (80%of 14%).Even
if every traveler attending a travel clinic receives immunization with fully protective vaccine (efficacy =100%),
the effectiveness of such a strategy would only increase
to 14%.Another 5% of travelers reported that they had
visited their family physician before leaving. If all of
these physicians were also to provide a fully protective
vaccine to all travelers, this would add only modestly t o
the effectiveness of the current strategy.
Discussion
As expected, trips of long duration and under poor
living conditions increased the risk of HA. However. only
7%)of H A cases occurred in these high-risk travelers. I n
Lct, our data clearly demonstrate that H A is a risk for
everyone traveling in an endemic country regardless o f
the perception of risk. Similar results have been found
in previous studies.',' Although specific vaccination history was not obtained in this study, a pretravel visit to a
J o u r n a l of T r a v e l M e d i c i n e , V o l u m e 9, N u m b e r 1
14
Table 4
Proportion of the Nights Spent orthe Meals Eaten per Class of Accommodation or Food Facilities
Food
Housing
Propor/iorr of /lie
N(~/its/Sprrir
Casc Trips
h k ~ n l s / E ~ / ( ~ I l N = 108 (%)
Control Trips
N = 905 (?A)
p - Value
Cme Trips
N = 108 (%)
Control Trips
N = 905 (%)
,006
71 (66)
13 (12)
14 (13)
10 (9)
320 (35)
171 (19)
167 (19)
244 (27)
.13
40 (37)
29 (27)
26 (24)
13 (12)
406 (45)
214 (24)
157 (17)
125 (14)
,006
58 (54)
34 (31)
10 (9)
722 (80)
127 (14)
43 (5)
10 (1)
p - Value
First-class
( )'%,
1-40'%1
5()-99%1
1 ( )O'%I
(65)
(4)
(4)
(27)
426
42
50
386
(47)
(5)
(6)
(43)
79 (73)
5 (5)
9 (8)
15 (14)
602
52
48
202
(67)
01 (84)
3 (3)
4 (4)
831
27
20
26
(92)
(3)
(2)
(3)
69
4
4
29
,0001
Middle-class
( 1%
1 -49'%1
51)-O",
1 ( )( Y%>
(6)
(5)
(22)
.2
Low-budget
( l'%,
1-49'%,
5(t-ow,
1 ( n V%,
1 0 (9)
6 (6)
.0001
Fricnds/rrlativrs
(1%
1-40'%1
3 )-99%l
1 1%
68 (03)
(7)
7 (0)
25 (23)
682 (75)
17 (2)
63 (7)
142 (16)
.0006
travel clinic reduced the risk of acquired HA by 80%.
Unfortunately, even if every traveler visiting a travel
clinic were to be vaccinated and fully protected against
HA, the proportion of travelers that attend such clinics
is to low (1 4?4) that only a small proportion of all cases
would be prevented by the current strategy. Even adding
the 5%who went to see their family physician would not
significantly improve this result.
In this study, we are likely to have missed HA cases
that occurred in adults traveling for prolonged periods,
as many were probably sick abroad and not reported.'
Although these cases do not contribute to direct costs
for the Canadian health care systeni, they are certainly
worth preventing. Mitclassification of risk factors due to
recall bias is not likely to have affected our results. The
observation that cases were 3 times more likely than controls to report gastroenteritis can be considered as a
validity check for their increased risk of enteric disease.
The most remote travel events included in the current
study would have taken place less than 3 years before the
interview, which is a short period of tinie considering
thc importance of wch events for most people. Moreover, the data collected were largely general information,
not likely to be forgotten. The classification of lodging
facilities and restaurants was clearly subjective and there
was probably some misclassification between the high and
middle classes. However, it is very likely that those who
recalled the use of low class hotels and/or low budget
restaurants were truly i n facilities with questionable
66
6
29
7
(61)
(6)
(27)
(6)
631 (70)
62 (7)
147 (16)
62 (7)
.06
hygienic conditions. To skirt these limitations, we concentrated our analyses on travelers who reported travel
involving low class-low budget facilities.
In this study, residents from QuCbec were at increased
risk of HA compared with those fiom Ontario. This may
be due in part to the greater participation rate of QuCbec
cases (62% vs. 43% for Ontario). However, in another
recent study, young QuCbec travelers also reported greater
risk of exposure to body fluids than Ontario travelers.'
One possible explanation for these findings could be cultural differences in the approach to explorative and highrisk behaviors between predominantly French (QuCbec)
versus predominantly English (Ontario) populations.
The decreased risk of HA associated with older age
could result from safer behavior patterns in older versus
younger travelers. However, the lesser risk in older subjects may also have been influenced by the greater proportion of individuals who would be expected to have
natural inlmunity with advancing age. Seroprevalence data
in MontrGal found that HA virus antibody was present
in 82% of individualsborn before 1930,600/0 of those born
between 1930 and 1940,49% of those between 1941 and
1950 and 29% of those born between 1951 and 1960,
and only 10%)and 1% of those born between 1961 and
1970 and between 1971 and 1980, respectively.*
The incidence of HA calculated for our travelers was
estimated to be 5 X 10-j per month of travel or 60 X 10-i
per year oftravel. Since the overall participation rate in cases
was 52??1,ourestimate should be doubled (120 X I@ year).
De Serres e t al., H e p a t i t i s A in Travelers
15
Table 5 Adjusted Odd Ratios of HA in Multivariate
Analyses
Multivariate analyses
95%
OR
Risk levels
High-risk
Intermediate-risk
Low-risk
Visiting friends and
relatives (VFR)
Aware of risk
Visited a travel clinic
Province
Ontario
Qutbec
Type of travel
Fully organized
Partly organized
Self-organized
Age group (years)
Confidence Interval p-value
7.2
1.3
ref
1.76-29.40
0.73-2.32
-
.006
1.1
0.3
0.2
0.52-2.45
0.19-0.53
0.07-0.63
.8
ref
4
2.4-6.7
.0001
0 . 8 4 3 .96
1.50-4.93
.13
.001
0.69
0.61
0.44
0.20
0.36-1.32
0.31-1.21
0.21-0.92
0.07-0.58
.3
.2
.03
.003
0.98
0.51
0.42
ref
0.40-2.39
0.24-1.08
0.20-0.88
.08
1.9
2.3
3.0
0.63-5.66
0.9 1-5.97
1.11-8.1 1
.3
.08
.03
.4
-
,0001
.005
ref
1.8
2.7
ref
I24
25-34
35-44
45-54
55+
Household income
($/year)
< 20,000
20,000-39,999
40,000-59,999
60,000+
Continent
Africa
Americas
Asia
Eastern Europe
.97
.02
ref
ref = reference group, OR = odds ratio.
Given that the reported rate of HA in the Canadian popper year,”’”the risk of HA in travelulation is 6 X
ers is 20 times greater than the general population. Our rate
was 30-60 times lower than the incidence estimated by Steffen for Swiss travelers (3-6 X 10-’/month or 360&7200
X 10-’/year).’ However, the incidence estimated by Steffen was restricted to nonimmune travelers, an approach that
increases the incidence of hepatitis A.
In the current study, the vaccination status of the
travelers was not verified. A travel clinic visit was only
80%effective to protect against HA. In other words, some
of those who went to a travel clinic sustained the dicease. This observation can be explained in two ways. I t
is possible that these cases represent vaccine failures.
Although the hepatitis A vaccines available at the time
the study trips occurred were highly effective, rare vac-
-
cine failures have been reported.” Alternately, and more
likely, some of the travelers in this study may not have
been offered (or may have refused) active immunization
against HA despite clearly articulated national guidel i n e ~ . ~Such
, ‘ omissions (or refusals) are most likely to
have occurred in those perceived by physicians and
travelers to be at lower risk (i.e., destined for first class
resorts).
In Canada, the cost of travel related disease prevention is currently left entirely to the traveler based
on the assumption that individuals wealthy enough to
travel should have sufficient resources to pay for preventive services. Although travel is certainly more common among individuals with higher incomes, it is not
limited to these people: 29% of the controls in our study
had a household income < $39,999 Can (< $27,000 US)
and 9% earned <$20,000 Can (< $13,500 US). Moreover, even those who have the resources to pay for
pretravel care must actually make it to the clinics to
receive this service. Attendance at travel clinics might
reasonably be expected to increase with public education o r free access. However, even then, we expect that
a substantial proportion of travelers would not go to
these clinics because they do not feel themselves to be
at risk. Half of the travelers gave this reason for not
attending a travel clinic.’’ As travel to HA endemic
countries becomes more and more frequent in our
population and the proportion of these travelers with
natural immunity decreases, other strategies have to be
considered to increase the rate of immunization in
travelers. Both risk awareness and vaccine accessibility
have to be increased. Universal immunization should
be seriously considered. Meanwhile, public health
authorities should increase their effort either directly
or through other channels like pharmacists or travel businesses to inform travelers that no trip is completely risk
free. Immunization would prevent HA not only in the
travelers themselves but also in those secondarily infected
after their return.
From a public heath perspective, any program with
an effectiveness of 1 1%) is clearly inadequate. A prompt
reevaluation of the current strateby is long overdue.
Acknowledgments
We gratefully acknowledge Colette Couture for
the coordination of this study and Sophie Auger,
Claude Boulianne, Lise Fillion, Stephanie Michaud, Julie
Picard, Danielle Vachon, Monali Varia, Prabo Dwight,
Samuel Thomas, Deborah Owens, Cecilia Alterman,
Jessica Chan, Gemnia Vena, Lucille Repetski, Julie
Chircop, Julie Marianayagam for collecting and entering data.
16
J o u r n a l o f Travel M e d i c i n e . V o l u m e 9, N u m b e r 1
References
travail. Qukbec: MinistPre de la santP et des services sociaux
du Qukbec, 1997.
Correid JII, Shafer RT, Patel V, et al. Blood and body tluid
exposure as a health risk for international trave1ers.J Travel
Med 2001 ;8:263-266.
I’aynient I? Antibody levels to sekcted enteric viruses In a
French-Canadian population 111 the province of QuPbec
(Canada). lniniunol Infect Ilic 1991; 1:317-322.
Health Canada. Nonfiable diseases annual tununary 1997.Can
Coin Ihs R e p 1999;2S(S6):1-159.
Health Canada. Nonfiable disease annual sunmiary 199X.Can
Coin IXs R e p 20(H);26(SS)(Suppl): 1-1 5 1.
Feinstone SM, Gust ID.Hepatias A m c m e . In Plotkin SA, Orenstein WA. eds. Vaccine. 3rd Ed. Phdadelphia: WB Saunderc,
1999:65047 I .
I>uval U, Ile Serres G. Shadmani I t , et al. A population based
comparison between m ~ ~ l who
e n consulted mvel clinics and
those who did not. [submitted]
1. Steffen I<. I h k of hepatitis A in travellers. Vaccinr 1992;
1f J(S1ippl 1):6Y-72.
7
-.
Wolfr. MS. tiepatiti\ A dnd the American trave1er.J Infect Dis
1‘)‘)s; 171 (SUPpl l):s?‘)-s32.
3. (:enters for I)isease C h t r o l and Prevention. I’revention of
hepatitis A through xtive or piscive inimunization - recoiiinienddtions of the Advisory <:onimitter on Irnniunization Practices (ACII’). MMWIL IW9; 4X(RI<-l2):1-37.
4. National Advitory <:oniniittce on Iiiiinun~zation.Canadian
iiniiiuinzatiwi guide. 5th Ed. Ottawa. I )ep.rrtnlent of N.itioiial
He.ilth .ind Welfare, I998:2lS.
~
5. StcKcn I<, K.wc MA, Shapiro C N , et .]I., E p i d e n i i o l o ~and
prrventioii o f hcp.ititlc A i n travclcrc. JAMA 1094; 272:
885-1 ( I f )f J.
0. Ihv;il 11. Uoucher E I h n I\. et .II. Contrble dc I’hcpatite A
par I ’ i r n i i i t i ~ ~ i c ~.IU~ t iQo ~u ~i . 1 ~rapport
~:
fin.11 du groupe de
7.
X.
9.
10.
1 1.
12.
Appendix HA Endemic Countries
The countries where hepatitis A is endemic
Africa:
Central and South America:
Asia and Oceania:
Eastern Europe:
Forliter Soviet Republics:
Middle East:
All the countries.
All countries including Mexico and the Caribbean.
All countries excluding Japan, Singapore, H o n g Kong, Australia,
and N e w Zealand.
Albania, Bosnia, Bulgaria, Croatia, Cyprus, Czech Republic, Greece. Hungary,
Macedonia, Malta, Poland, Roniania,Slovakia. Slovenia, Yugoslavia.
Arnienia, Azerbagan, Belarus, Estonia. Georgia. Kazakhstan, Kyrgyzstan, Latvia,
Lithuania, Moldova. Russia. Tajikistan, Turkmenistan, Ukraine, Uzbekistan.
l r m , Iraq, Israel, Jordan, Kuwait, Lebanon, Oman, Saudi Arabia, Syria, Turkey,
United Arab Einirntes. Yemen.