| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The aim of this study was i) to precisely estimate the incidence rate of morbidity in family members traveling to tropical and subtropical countries, ii) to investigate the types of disease they experienced, iii) to identify potential risk factor(s), and iv) to compare the findings between children and corresponding adults. We therefore conducted a prospective controlled study among 157 children (cases) and their adult relatives (controls) who attended our travel clinic for pre-travel consultation.
|
MATERIALS AND METHODS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Study procedures. During the pretravel consultation, families received standardized pretravel information. We updated routine vaccinations. Specific vaccinations, such as hepatitis A and B, typhoid, meningitis, yellow fever, and rabies, were done if deemed necessary, following the recommendations of the Swiss expert group for travel medicine (http://www.safetravel.ch). Malaria chemoprophylaxis was prescribed when appropriate. A questionnaire was filled out for each child and adult with information on personal demographics, general health, vaccination status, and type of travel planned.
Self-administered questionnaires were then given to be completed weekly during travel and 4 weeks after return. This time period was chosen to cover the period of malaria prophylaxis after return (so that full adherence could be assessed), as well as the incubation period of most illnesses acquired during travel. These questionnaires included questions on preventive measures (use of repellent and mosquito net; avoidance of tap water, salad, and ice cubes) and health issues with both open-ended and directed questions. When an illness occurred, parents were requested to answer 21 questions about symptoms (general wellness, irritability, fatigue, fever, shivers, sweating, joint pain, muscular pain, rhinitis, otalgia, cough, headache, dizziness, abdominal pain, diarrhea, constipation, vomiting, rash, seizures, sleep disorder, motion sickness), to mention whether they had attended a medical practitioner, and whether they had taken any medication or not.
In addition, all parents were contacted by phone 2 months after return to check for new events and clarify the answers to the questionnaires if necessary.
Data analysis. Incidence rates of morbid episodes during the total observation period were calculated using a Poisson distribution. Days of illness were censored from the total observation period for all of the incidence calculations. A conditional maximum likelihood estimate of the rate ratio between both groups was calculated for each type of morbid episode using the polynomial multiplication method.10 This also provided mid-P estimates. For diseases whose incidence rate exceeded 2 episodes per 100 person-weeks, we compared incidence rates during travel and during the 4 weeks post-travel using the same method.
Life tables of the occurrence of morbid episodes were plotted for both groups. The occurrence of morbid episodes per week of observation both during travel and after return was plotted for both groups.
For diseases whose incidence exceeded 3 episodes per 100 person-weeks, we first investigated potential predictors in a univariate analysis using the conditional maximum likelihood estimate of rate ratios. We then entered variables whose P value was < 0.2011 into a multivariate analysis using a Poisson regression model. The fit of the data to a Poisson distribution in the regression was tested by the deviance goodness-of-fit test. The adequacy of the fitted model was tested using the G2 deviance (likelihood ratio) test statistic.12 Data were analyzed using SPSS 12.0 (SPSS, Inc., Chicago, IL) and StatsDirect (StatsDirect Ltd, Cheshire, U.K.).
|
RESULTS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Study population.
There were 157 child–adult pairs. Demographic and personal characteristics are summarized in Table 1
. The following comorbidities were identified in children from their pretravel questionnaire: 23 (14.6%) had recurrent otitis media, 8 (5.1%) asthma, 17 (10.8%) eczema, and 17 (10.8%) hay fever. For adults, 85 (54.1%) presented at least 1 cardiovascular risk factor, and 17 (10.8%) were taking regular medication. Comorbidities were as follows: 10 (6.3%) asthma, 15 (9.6%) varicose veins, 4 (2.5%) hypertension, 10 (6.3%) eczema, 26 (16.6%) hay fever, 7 (4.5%) migraine, 7 (4.5%) depression, 2 (1.3%) cancer, and 1 (0.6%) was diabetic. Adults occupied the following professional categories13: 14% were stay-at-home parents, 37% were in classes 1 (legislators, senior officials, managers) or 2 (professionals), 30.6% in classes 3 (technicians and associate professionals) or 4 (clerks), and 14.6% in classes 5 (service, shop and market sales), or 7–9 (7, craft and related trades; 8, plant and machine operators and assemblers; 9, elementary occupations). Six participants did not indicate their profession.
|
Preventive measures. One hundred eight (68.8%) children and 117 adults (74.5%) received all vaccinations recommended.14 Malaria chemoprophylaxis was prescribed to 74 (47.1%) families; 61 (82.4%) children and 65 (87.8%) adults were fully adherent (no missed dose). A further 52 (33.1%) children and 50 (31.8%) adults traveling to low-level endemic areas were prescribed standby treatment. None of them made use of it. One hundred forty-seven children and 147 adults (93.6%) used a mosquito net, a repellent lotion, or had an air-conditioned room. Fifty (31.8%) children and 26 (16.6%) adults were compliant to all 3 of the following measures of food-borne disease prevention: no tap water, no ice cubes, and no salads.
Morbid episodes.
Ninety-six (61.1%) children and 88 (56.1%) adults presented at least 1 morbid episode during follow-up. In total, there were 157 episodes of illness in children and 140 in adults. The median time until the first episode of disease was 7.2 days in children and 8.0 days in adults, according to the survival analysis (Figure 1). Incidence rates for overall morbidity as well as specific illnesses are shown in Table 2. There were no significant differences in the incidence rates of morbid episodes between children and adults, except for fever, which occurred significantly more frequently in children (RR = 2.5, 95%; CI 1.3–5.0; P = 0.003). Incidence rates were significantly higher during travel than after return for all types of disease (Table 3), except for rhinitis in adults.
|
|
|
Eighty-one (51.6%) children and 73 (46.5%) adults took medication. In total, 136 courses of medication were given to children and 135 to adults. The main classes of drugs used in children and adults were probiotics (respectively 38 and 25 courses), antipyretics/painkillers (respectively 26 and 25 courses), and anti-diarrheals (respectively 17 and 22 courses). Other drugs used included ENT and cough medication, antibiotics, antiemetics, and antihistamines. There was no significant difference in medication used between both populations. Antibiotics were prescribed to children for the following diagnoses: pharyngitis (4 cases), otitis media (3 cases), sinusitis (1 case), urinary tract infection (1 case), and gastroenteritis (1 case). They were prescribed to adults for otitis media (4 cases), gastroenteritis (4 cases), pharyngitis (3 cases), sinusitis (2 cases), giardiasis (1 case), and gynecologic infection (1 case). All courses of antibiotics were medically prescribed. Two (1.3%) adults and none of the children were hospitalized. One admission was for diverticulitis and the second for abortion, both after return.
Predictors of disease.
Results of analyses for predictors of morbid episodes are shown in Table 4. Multivariate analyses indicated that incidence of overall morbidity increased with rising level (classes 1–2) of parental occupation (P = 0.004) and with use of ice cubes (P = 0.034) but was decreased by history of previous travel to the tropics (P = 0.014). Incidence of fever was significantly higher in the 0- to 5-year age group (P = 0.004) and with use of ice cubes (P = 0.018). Abdominal pain was more frequent with parental occupation class 1–2 (P = 0.039). The incidence rate of headache was significantly lower in children of age 0–5 years (P = 0.018), in subjects who had previously traveled to the tropics (P = 0.049), and in those who used mosquito nets (P = 0.023) but was higher in persons who were poorly compliant with their malaria chemoprophylaxes (P = 0.007).
|
|
DISCUSSION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Around 60% of both children and adults presented at least 1 episode of illness during their follow-up. Previous studies based on post-travel interviews have shown variable figures: 15–38% in Swiss travelers,8,16 36%15 in Scottish package tourists, and 64%9 in American subjects. It is likely that the high proportion in our study is primarily due to the absence of recall bias with families that knew they would have to fill in the questionnaire, probably inclining them to closely observe their own health status. It is also likely that other factors, such as destination,17 length of travel, and duration of follow-up, and also variable definitions of what constituted an episode of illness influenced the results of each study.
We showed that there is a steep rise in the number of episodes of illness during the first 15 days after departure; it then levels off to return to a baseline by 
30 days of follow-up. Half of the children and adults who developed illnesses did so during the first 8 days after departure. We were not able to obtain baseline rates of illness before departure or incidence rates of illness in the general population (these data are not available in Switzerland); this would have enabled us to estimate the fraction of illness attributable to travel. However, given the length of our follow-up, we believe that the baseline attained at the end of follow-up approaches these rates.
Apart from fever, there were no differences between children and adults in the incidence of morbid episodes. This demonstrates that for this cohort of travelers, age was not a major determinant of morbidity during travel. The environment, which was exactly the same for children and their adult controls (with same levels of protective measures adherence), seems to be a much stronger determinant. The higher incidence of fever in children was principally due to an excess of febrile episodes in the youngest age group (0–5 years). Steffen and others7 previously described a tendency toward more fever in children traveling to the tropics; however, none of them was < 7 years old. Exposure to micro-organisms that are foreign to the home environment is inherent to travel to the tropics, and in young children who have not yet achieved mature immunity, this may manifest by an excess of febrile episodes. It is also possible that parents may be more likely to record children’s temperature in this age group if they appear unwell.
Diarrhea and abdominal pain were the most frequent complaints encountered both in children and adults. This is in line with all of previous studies of morbidity in travelers to the tropics.7–9,15,16,18 Management of traveler’s diarrhea, especially in children, relies essentially on the use of oral rehydration. This was systematically discussed in the pretravel consultations; 13% of the children took oral antidiarrheals (loperamide), despite concerns that this treatment can induce paralytic ileus, especially in children < 2 years of age.19,20 The youngest child to take loperamide in our group was 3 years old. The same type of behavior has been described by Pitzinger and others.21 Antidiarrheals are not recommended or prescribed for children in our travel clinic, nor are antibiotics for all travelers. Antidiarrheals were purchased through other channels, either in Switzerland or at travel destination, but antibiotics were always medically prescribed. Such behavior highlights the necessity to approach pretravel consultation, not only through prescription of adequate vaccines or drugs but also through specific advice on health-seeking behavior and drug use while sick abroad. Nineteen percent of all children and 15% of all adults sought medical attention, with a comparable figure of 12% reported by Hill,9 but 50% of both adults and children used medication. This demonstrates that self-medication was very frequent. The nature of the drugs corresponded to the main symptoms that were reported, with a high consumption of antipyretics, antidiarrheals, and probiotics. An appropriate strategy would be to recommend a travel-medicine kit with a summary sheet that highlights the lower age limits for the different drugs included and their precise indication in terms of symptoms or signs.
Analysis of predictors of morbid episodes showed that a higher class of parental occupation was associated with both an increase in overall illness, particularly frequent abdominal pain. This could be explained by an increased awareness and propensity to report unusual symptoms in these families, associated with a higher level of education. Use of ice cubes was associated with a higher incidence of fever episodes (not diarrhea), maybe as a marker of more risky behavior. As usual, standard advice to avoid ice cubes was poorly followed because 40% of children and nearly 50% of adults did not comply with this recommendation. How to improve this is a recurrent question with no satisfying answer.
The most important limitation of our study is the selection bias because our population was recruited among travel clinic attendants, who were therefore well prepared for their journey compared with a group of people who would have been taken randomly from a travel agency, for example, or at the airport. However, the design of our study, which included a formal informed consent, thorough initial pretravel assessment, and a detailed questionnaire to fill in every week during and after travel, required full understanding and some training that would have been impossible to perform in an airport or travel agency. We believe that precise estimates of incidence rates can only be realistically obtained through such thorough questioning and close follow-up of travelers. Such an approach often implies a relatively small sample size, as in this study, that may have limited our ability to identify weak potential risk factors for certain diseases as well as less-frequent disease episodes, such as travelers’ malaria. Only 1 child had a presumptive diagnosis of malaria during travel and received artesunate treatment. Our incidence of 3/1000 travelers is within the range of larger studies based mostly on post-travel questionnaires, i.e., 0.7–18 cases/1000.9,22–27 The latter observation suggests, as expected from the study subjects’ selection, that the incidence rates estimated from our study sample are probably at the lower range of the ones that could be observed in the general population of travelers who often do not seek pretravel advice.28,29 Likewise, no cases of typhoid fever, meningitis, or hepatitis were reported. This may have been an effect of adequate immunization during the pretravel clinic, but again the attack rates for these diseases are low8,9 and our sample size may have limited our ability to detect such episodes.
In conclusion, the similar incidence of morbidity in children and adults, as well as the mildness of the morbid episodes, challenges the common view that it is unwise to travel with small children. In view of the high proportion of self-medication for mild symptoms, pretravel advice should also tackle appropriate response to disease.
Received February 17, 2007. Accepted for publication June 13, 2007.
Acknowledgments: We warmly thank Marie-José Puelma, who was instrumental in the recruitment of families, as well as the other nurses and physicians of the travel clinic. The authors thank the study subjects, especially the parents who agreed to regularly fill in the questionnaire.
* Address correspondence to Blaise Genton, Swiss Tropical Institute, Socinstrasse 57, 4002 Basel, Switzerland. E-mail: Blaise.genton{at}unibas.ch 
Authors’ addresses: Catherine Newman-Klee, Department of Pediatrics, Hôpital de Morges, Ch. du Crêt 2, 1110 Morges, Switzerland, Telephone: +41 21 802 06 77, Fax: +41 21 545 01 10. Valérie D’Acremont and Blaise Genton, Ifakara Health Research & Development Centre, P.O. Box 78373, Dar es Salaam, Tanzania, Telephone: +255 787 957883 (V.D’A.), +255 786 190069 (B.G.). Christopher Newman, Department of Pediatrics, Lausanne University Hospital, 1010 Lausanne, Switzerland, Telephone: +41 21 545 06 07, Fax: +41 21 545 01 10. Mario Gehri, Children’s Hospital, Ch. de Montétan 16, 1000 Lausanne, Switzerland, Telephone: +41 21 627 28 29.
Reprint requests: Blaise Genton, Swiss Tropical Institute, Socinstrasse 57, 4002 Basel, Switzerland, E-mail: Blaise.genton{at}unibas.ch.
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
year period. Arch Dis Child 88: 432–434.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
