Risk factors for symptoms of infection and microbial carriage among French medical students abroad.

OBJECTIVES: To investigate symptoms of infections and their risk factors among French medical students undertaking an internship abroad.
METHODS: Clinical follow up, and qPCR-based respiratory, gastrointestinal, and vaginal pathogen carriages were prospectively assessed pre-travel and post-travel, in a cohort of medical students departing from Marseille, France.
RESULTS: 293 students were included. 63.5%, 35.8%, and 3.6% of students reported gastrointestinal, respiratory, and vaginal symptoms, respectively. The acquisition rate of Enteroaggregative Escherichia coli and Enteropathogenic E. coli was 40.9% and 18.6%, respectively. A significant increase was observed for rhinovirus and Streptococcus pneumoniae by comparing the prevalence of pathogens in pre-travel and post-travel samples. Gardnerella vaginalis and Atopobium vaginae acquisition rates were 12.9% and 13.9%, respectively. Being female, primarily traveling to Vietnam, and living in basic accommodation conditions were independent risk factors for reporting respiratory symptoms. Students reporting respiratory symptoms were three times more likely to acquire S. pneumoniae. Traveling primarily to north India and Senegal were independent risk factors for diarrhea.
CONCLUSION: This study makes it possible to identify the leading infectious diseases linked to travel in a group of French medical students undertaking an internship abroad and the risk factors on which to base targeting students for reinforced pre-travel advice.

Introduction

International travelers are exposed to potential pathogens, including viruses, bacteria, or parasites, with the risk of community or hospital spread upon return, whether or not they present health problems during their trip. Therefore, there is a risk of pathogens being imported into France from endemic areas abroad, either from foreign travelers visiting France, French travelers visiting foreign countries, or migrants and expatriates treated in France with a risk of indigenous spread. This has been extensively described, for example, among French Hajj pilgrims traveling to Mecca, Saudi Arabia (Hoang and Gautret, 2018).

An Australian organization, Work the World, has enabled 15,000 medical students to participate in internships abroad since 2005 (Work the World, 2019). Developing countries are becoming popular destinations for internships. Medical internships abroad are generally hospital immersion experiences, but young medical students also participate in humanitarian missions unrelated to clinical activities such as school renovation, for example (Dao et al., 2020, Watson et al., 2019). In one Australian survey, 64% of students experienced some sort of health problems while taking part in electives abroad; travelers’ diarrhea was the most common problem (40%) (Goldsmid et al., 2003). Our preliminary report on a cohort of 134 French medical students participating in international electives showed that 73.9%, 38.8%, and 5% of them reported gastrointestinal, respiratory, and vaginal symptoms, respectively (Dao et al., 2020). We showed that the acquisition rate of Enteropathogenic Escherichia coli (EPEC) and Enteroaggregative E. coli (EAEC) was 41% and 53%, respectively. By contrast, the acquisition of respiratory viruses was low but was associated with persistent respiratory symptoms at return. Respiratory bacterial acquisition ranged from 3.3% for Streptococcus pyogenes to 15.0% for Haemophilus influenzae. Atopobium vaginae and Gardnerella vaginalis percentage of acquisition were 14.3% and 7.7%, respectively. So far, to our knowledge, the risk factors for acquisition of pathogens have not been clearly identified among medical students abroad. The relationship between symptoms and the carriage of pathogens also remains poorly understood, making it difficult to distinguish between infection and colonization. We aimed to conduct this study to investigate the risk factors for symptoms of infections among French medical students undertaking an internship abroad.

Materials and methods

Study design

A monocentric prospective cohort survey was conducted over two years (2018–2019) among medical students from the Faculty of Medicine in Marseille, France, who were planning to participate in an internship abroad during the summer. Recruitment was performed voluntarily during their vaccination and pre-travel consultation at the Institut Méditerranée Infection on the Marseille University medical campus. Participants were asked to complete an inclusion questionnaire, including demographic data, history of chronic illness, intended travel dates, and destination. All participants received advice regarding preventing diarrhea during travel (hand hygiene, safe food, and water habits), but prophylaxis for traveler’s diarrhea was not prescribed. Because dates of departure and return are different for each student, the samples were not taken at once. Participants were given two sets of “pre-travel” and “post-travel” kits, which contained questionnaire and sampling equipment (commercial rigid cotton-tipped swab applicators and viral transport media). They were also instructed how to self-collect samples, as follows: three cm in the nostril, five turns on the post wall of the pharynx, five streaks for respiratory samples; rectal samples were collected using two methods: rectal self-sampling when having a bowel movement, three centimeters through the anus, gently rubbing the inner walls of the rectum several times or stool collection after emission; vaginal samples were collected by placing the swab about three centimeters into the vagina and gently rubbing the inner walls several times, avoid touching the skin and vulva with the swab. Samples were self-collected using commercial rigid cotton-tipped swab applicators (Medical Wire & Equipment, Wiltshire, UK) and placed in viral transport media (Sigma Virocul®) for further processing at our laboratory. A document of instructions for self-sampling was also provided in the kits. During the week before travel, each student was invited to deposit their self-collected samples and a pre-travel questionnaire that collected information about their health problems and antibiotic use, if applicable. After their travel, they were invited to self-collect samples during the week following their return to France. Students were also provided with a post-travel questionnaire addressing the internship’s exact place, the type of activities during their stay, including tourism and travel to other countries over the internship period. Accommodation conditions, contact with animals or children, symptoms, onset of symptoms, and treatment during their stay were also documented. Influenza-like illness (ILI) was defined as a sore throat, cough, plus subjective fever (Rashid et al., 2008). Diarrhea was defined by at least three loose or liquid stools per 24 h.

Microbiological methods

The methods for identifying respiratory, gastrointestinal, and vaginal pathogens by PCR assay are detailed elsewhere (Dao et al., 2020). The following respiratory pathogens were screened for: Influenza A and B viruses, Human coronaviruses, Human parainfluenza viruses, Human rhinovirus, Human metapneumovirus, Adenovirus, Respiratory syncytial virus, Staphylococcus aureus, H. influenzae, Streptococcus pneumoniae, Klebsiella pneumoniae, and S. pyogenes. The following gastrointestinal pathogens were screened for: Norovirus, Rotavirus, Adenovirus, Astrovirus, Entamoeba histolytica, Giardia lamblia, Cryptosporidium spp., Salmonella spp., Shigella spp. (EIEC), Enterohemorrhagic E. coli (EHEC), Enteropathogenic E. coli (EPEC), Enteroaggregative E. coli (EAEC), Campylobacter jejuni and Tropheryma whipplei. The following vaginal pathogens were screened for: Chlamydia trachomatis, Mycoplasma genitalium, Neisseria gonorrhoeae, Trichomonas vaginalis, Mycoplasma hominis, A. vaginae, and G. vaginalis.

PCR was considered positive for virus or bacteria detection when the cycle threshold (CT) value was ≤35. Bacterial vaginosis was defined by a G. vaginalis DNA load ≥109 copies/mL (CT ≤ 18) and/or an A. vaginae DNA load ≥108 copies/mL (CT ≤ 21), as previously reported (Menard et al., 2008).

The acquisition of a pathogen was defined as negative before travel and positive when returning to Marseille, France.

Statistical analysis

STATA software version 14.2 was used to conduct statistical analysis. Differences in the proportions were tested by using Fisher’s exact or Pearson’s chi-square tests, as appropriate. McNemar’s test was used to evaluate the potential acquisition of pathogens (prevalence after versus before travel). Clinical symptoms during travel were reported only if the onset of symptoms took place during travel. Univariate analysis was used to evaluate unadjusted associations between the prevalence of symptoms during travel and multiple factors. A p-value <0.05 was considered to be statistically significant. Only variables with a prevalence equal to or more than 5.0% were considered for statistical analysis. Variables with p-values <0.2 in the univariate analysis were included in the multivariate analysis. Log-binomial regression was used to estimate factors’ adjusted risk ratios for symptoms of infections.

Ethics

The protocol was approved by our Institutional Review Board (2019-006). It was performed following the proper clinical practices recommended by the Declaration of Helsinki and its amendments. All participants gave their written informed consent.

Results

Characteristics of study participants

A total of 293 students agreed to participate and answered the post-travel questionnaire. The M/F gender ratio was 0.31, with a median age of 21 years (ranging from 18 to 25 years). Most participants (80.9%) were students in their second year of medical studies and were taking part in a non-medical humanitarian mission. The remaining participants were in their 4th year of study and were assigned to different departments of medicine or surgery for clinical training (Supplementary Table S1).

The primary travel destinations for internships were in Africa (29.0%), South East Asia (27.7%), South America (21.8%), and South Asia (18.8%). The top five primary destination countries were Vietnam (24.2%), India (18.8%), Peru (16.4%), Tanzania (10.2%), and Madagascar (9.2%). The mean travel duration was 41 days ± 11.1 days (ranging from 16 to 78 days). Accommodation conditions were judged as being “very clean” by 14.7% (43/293) of students, “clean” by 44% (129/293), and “very basic” by 41.3% (121/293). 42.7% and 79.9% of the participants reported contact with animals and local children, respectively (Supplementary Table S2). During their stays, 94.2% also traveled as tourists in the country of primary destination, and 50 (17.1%) visited other countries. The top five additional destination countries were Laos (4.8%), Cambodia (4.4%), Bolivia (3.4%), Thailand (1.7%), and Argentina (1.7%).

A total of 55/293 (18.8%) were vaccinated against influenza and only 2/293 (0.7%) against invasive pneumococcal infections at inclusion. In addition, 260/293 (88.7%) and 273/293 (93.2%) students were vaccinated against hepatitis A and hepatitis B before travel, respectively.

Overall, 5.1% took antibiotics in the week before departure, and 17.8% took doxycycline as a chemoprophylaxis against malaria during their stay.

Respiratory infections

A total of 35.8% (105/293) of students reported respiratory symptoms during travel. The median time between arrival at the travel destination and the onset of symptoms was 17 days [ranging from 1 to 58 days]. The most frequent respiratory symptoms were rhinitis (27.1%), sore throat (21.8%), and cough (20.1%), followed by fever (9.6%) and dyspnoea (6.5%) (Figure 1a). 7.9% of students declared persistent symptoms on their return to France, and 5.5% took antibiotics (ATB) for respiratory symptoms during travel.

Figure 1

Prevalence of respiratory (a) and gastrointestinal (b) symptoms during travel.

A total of 275 (93.9%) students provided paired nasopharyngeal swabs. 52.4% of students acquired at least one respiratory pathogen. 17.8% (49/275) of students acquired at least one respiratory virus with human rhinovirus (14.6%) being the most frequent. Twenty-three students were still symptomatic after returning to France. Of whom, 16 (69.6%) were positive for at least one pathogen. Bacterial acquisition rates were higher (40.7%), with S. aureus (18.9%) being the most frequent, followed by H. influenza (17.1%) (Table 1 ). A total of 6.2% of students acquired a virus-bacteria combination and 11.6%, a bacteria combination. When comparing the post- versus pre-travel prevalence of pathogens, a significant increase was observed for rhinovirus and S. pneumoniae.

Table 1

Prevalence of respiratory pathogens.

Variables	Pre-travel		Post-travel		Acquisition		p*
	n = 2751	%	n = 2792	%	n = 2753	%
Viruses
Adenovirus	1	0.4	0	0	0	0	NA
Coronaviruses	1	0.4	6	2.2	6	2.2	0.06
Coronavirus HKU1	0	0	3	1.1	3	1.1	0.08
Coronavirus 229E	0	0	2	0.7	2	0.7	0.16
Coronavirus NL63	1	0.4	1	0.4	1	0.4	1.0
Coronavirus OC43	0	0	0	0	0	0	NA
Influenza A	0	0	1	0.4	1	0.4	0.32
Influenza B	0	0	2	0.7	2	0.7	0.16
Metapneumovirus	0	0	0	0	0	0	NA
Parainfluenza virus	0	0	0	0	0	0	NA
Respiratory syncytial virus	1	0.4	2	0.7	2	0.7	0.56
Rhinovirus	25	9.1	44	15.8	40	14.6	0.015
At least one virus	26	9.5	54	19.6	49	17.8	0.001
Bacteria
Haemophilus influenzae	122	44.4	119	42.7	47	17.1	0.76
Klebsiella pneumoniae	23	8.4	28	10.0	19	6.9	0.38
Staphylococcus aureus	63	22.9	83	29.8	52	18.9	0.052
Streptococcus pneumoniae	5	1.8	15	5.4	15	5.5	0.025
Streptococcus pyogenes	3	1.1	8	2.9	7	2.6	0.096
At least one bacteria	161	58.6	180	64.5	112	40.7	0.11
Virus and bacteria combination	13	4.7	29	10.4	17	6.2	0.01
H. influenzae–rhinovirus	9	3.3	13	4.7	4	1.5	0.37
K. pneumoniae–rhinovirus	1	0.4	1	0.4	0	0	NA
S. pneumoniae–rhinovirus	0	0	1	0.4	1	0.4	0.32
S. aureus–rhinovirus	6	2.2	12	4.3	8	2.9	0.16
Bacteria combination of two bacteria	61	22.2	87	31.2	32	11.6	<0.000
H. influenzae–K. pneumoniae	12	4.4	15	5.4	3	1.1	0.49
H. influenzae–S. pneumoniae	4	1.5	11	3.9	5	1.8	0.07
H. influenzae–S. aureus	36	13.1	40	14.3	12	4.4	0.55
H. influenzae–S. pyogenes	3	1.1	5	1.8	1	0.4	0.41
K. pneumoniae–S. pneumoniae	0	0	3	1.1	3	1.1	0.08
K. pneumoniae–S. aureus	4	1.5	7	2.5	2	0.7	0.26
K. pneumoniae–S. pyogenes	0	0	0	0	0	0	NA
S. pneumoniae–S. aureus	1	0.4	6	2.1	6	2.2	0.06
S. pneumoniae–S. pyogenes	0	0	0	0	0	0	NA
S. aureus–S. pyogenes	1	0.4	0	0	0	0	NA
At least one pathogen	174	63.3	205	73.5	144	52.4	0.008

1, 2275 and 279 students provided pre- and post-travel nasopharyngeal swabs, respectively.

3Acquisition of respiratory pathogens was calculated in 275 students who provided both pre- and post-travel nasopharyngeal samples.

p-Value: pre-versus post-travel, McNemar’s test.

Gastrointestinal infections

A proportion of 63.5% of students reported at least one gastrointestinal symptom, all during travel and only one following return. The median time between arrival at the travel destination and the onset of symptoms was 13 days [ranging from 1 to 65 days]. The most frequent symptoms were diarrhea (48.1%) and abdominal pain (46.4%), followed by nausea (26.3%) and constipation (19.8%) (Figure 1b). 8.9% reported persistent symptoms on return to France, and 3.4% took an antibiotic for diarrhea during their stay.

A total of 274 (93.5%) students provided paired rectal swabs. Three students reported gastrointestinal symptoms on day one after departure. But all rectal samples pre-travel were negative for gastrointestinal pathogens; 51.5% of students acquired at least one gastrointestinal pathogen. Nine students (3.3%) acquired at least one virus (adenovirus, astrovirus, and norovirus). Bacterial acquisition rates were higher (49.3%), notably for EAEC (40.9%) and EPEC (18.6%). Additionally, 2.6% of individuals acquired Shigella spp./EIEC, 1.5% Salmonella spp., and 1.1% G. lamblia (Table 2 ). When comparing the post- versus pre-travel prevalence of pathogens, a significant increase was observed for viruses overall, EAEC, EPEC, Salmonella spp., and Shigella spp./EIEC.

Table 2

Prevalence of gastrointestinal pathogens.

Pathogens	Pre-travel		Post-Travel		Acquisition		p*
	N = 2821	%	N = 2832	%	N = 2743	%
Viruses
Adenovirus	0	0	3	1.1	3	1.1	0.08
Astrovirus	0	0	3	1.1	3	1.1	0.08
Norovirus	0	0	3	1.1	3	1.1	0.08
Rotavirus	0	0	0	0	0	0	NA
At least one virus	0	0	9	3.2	9	3.3	0.003
Bacteria
Campylobacter jejuni	2	0.7	5	1.8	5	1.8	0.26
Enteroaggregative Escherichia coli	10	3.6	119	42.1	112	40.9	<0.000
Enterohemorrhagic E. coli	3	1.1	4	1.4	2	0.7	0.65
Enteropathogenic E. coli	25	8.9	66	23.3	51	18.6	<0.000
Salmonella spp.	0	0	4	1.4	4	1.5	0.05
Shigella spp./EIEC	0	0	7	2.5	7	2.6	0.008
Tropheryma whipplei	1	0.4	1	0.4	0	0	NA
At least one bacteria	38	13.5	147	51.9	135	49.3	<0.000
Parasites
Cryptosporidium parvum/hominis	0	0	0	0	0	0	NA
Entamoeba histolytica	0	0	0	0	0	0	NA
Giardia lamblia	1	0.4	3	1.1	3	1.1	0.32
At least one gastrointestinal pathogen	39	13.8	153	54.1	143	52.2	<0.000

1, 2282 and 283 students provided pre- and post-travel rectal swabs, respectively.

3Acquisition of gastrointestinal pathogens was calculated in 274 students who provided both pre- and post-travel rectal samples.

p-Value: pre-versus post-travel, McNemar’s test.

Vaginal infections

Of the 224 female students, eight (3.6%) reported vaginal symptoms such as leucorrhoea, during their stay, and two took antibiotics for this purpose. The median time between arrival at the travel destination and the onset of symptoms was 26 days [ranging from 6 to 59 days]. Three students (1.3%) were still symptomatic on returning to France.

209 (93.3%) of the participants provided paired vaginal samples. 15.8% of the participants acquired at least one vaginal microorganism with the highest acquisition rate for A. vaginae (13.9%) and G. vaginalis (12.9%) (Table 3 ). Nine students (4.0%) had molecular criteria for bacterial vaginosis on return. When comparing the post- versus pre-travel prevalence of microorganisms, a significant increase was observed for G. vaginalis (CT < 18), indicative of vaginosis.

Table 3

Prevalence of vaginal microorganisms.

Microorganisms	Pre-travel	Post-travel	Acquisition	p*
	N = 2121(%)	N = 2212 (%)	N = 2093 (%)
Bacteria
Atopobium vaginae	42 (19.8)	42 (19.0)	29 (13.9)	0.896
Atopobium vaginae (CT <21)	1 (0.5)	1 (0.5)	1 (0.5)	1.0
Chlamydia trachomatis	1 (0.5)	1 (0.5)	1 (0.5)	1.0
Gardnerella vaginalis	59 (27.8)	55 (24.9)	27 (12.9)	0.439
Gardnerella vaginalis (CT <18)	2 (0.9)	9 (4.1)	9 (4.3)	0.035
Mycoplasma genitalium	0	2 (0.9)	2 (0.96)	0.16
Mycoplasma hominis	4 (1.9)	3 (1.4)	2 (0.96)	0.65
Neisseria gonorrhoeae	0	0	0	NA
Parasites
Trichomonas vaginalis	0	0	0	NA
At least one vaginal pathogen	82 (38.7)	78 (35.3)	33 (15.8)	0.42

1, 2212 and 221 students provided pre- and post-travel vaginal swabs, respectively.

3Acquisition of vaginal pathogens was calculated in 209 students who provided both pre- and post-travel vaginal samples.

p-Value: pre-versus post-travel, McNemar’s test.

Risk factors for symptoms of infections

Being female, primarily traveling to Vietnam, and living in basic accommodation conditions were independent risk factors for reporting respiratory symptoms. Students suffering respiratory symptoms were three times more likely to acquire S. pneumoniae during travel. Traveling primarily to north India and Senegal were independent risk factors for reporting diarrhea (Table 4 ).

Table 4

Risk factors for respiratory symptoms and diarrhea during travel.

	Risk factors for respiratory symptoms during the travel					Risk factor for diarrhea during the travel
	Univariate analyses				Multivariate analyses	Univariate analyses				Multivariate analyses
	No	Yes	RR	p	aRR	No	Yes	RR	p	aRR
	n (%)	n (%)	[95%CI]		[95%CI], p	n (%)	n (%)	[95%CI]		[95%CI], p
Gender
Female	136 (72.3)	88 (83.8)	1.50 [1.06–2.11]	0.02	1.55 [1.08–2.22], p = 0.02	121 (54.0)	103 (46.0)	0.77 [0.51–1.15]	0.19	–
Male	52 (27.7)	17 (16.2)				31 (44.9)	38 (55.1)
Age (mean ± SD)	20.7 ± 1.1	20.7 ± 1.1	t = 0.38	0.70		20.7 ± 1.1	20.6 ± 1.1	t = 0.70	0.48
			0.70
Medical history
Vaccination against influenza	83 (34.9)	22 (40.0)	1.15 [0.78–1.70]	0.48		NA	NA	NA	NA
Primary travel destination
South East Asia	71 (33.5)	34 (42.0)	1.26 [0.89–1.77]	0.18	–	104 (49.1)	37 (45.7)	0.93 [0.71–1.22]	0.61
Vietnam	73 (32.9)	32 (45.1)	1.40 [0.97–2.00]	0.07	1.57 [1.07–2.31], p = 0.02	111 (50.0)	30 (42.3)	0.85 [0.62–1.14]	0.26
South Asia	86 (36.1)	19 (34.6)	0.96 [0.65–1.40]	0.82		107 (45.0)	34 (61.8)	1.38 [1.07–1.77]	0.02	–
South India	98 (37.4)	7 (22.6)	0.66 [0.41–1.05]	0.08	–	125 (47.7)	16 (51.6)	1.08 [0.75–1.56]	0.68
North India	93 (34.6)	12 (50.0)	1.53 [0.84–2.79]	0.16	–	123 (45.7)	18 (75.0)	1.64 [1.26–2.14]	<0.01	2.57 [1.13–5.85] p = 0.025
Africa	77 (37.0)	28 (32.9)	0.89 [0.65–1.24]	0.51		105 (50.5)	36 (42.4)	0.84 [0.63–1.11]	0.21
Tanzania	99 (37.6)	6 (20.0)	0.61 [0.38–0.97]	0.04	–	137 (52.1)	4 (13.3)	0.26 [0.10–0.64]	<0.000	0.37 [0.22–0.60] p < 0.000
Madagascar	91 (34.2)	14 (51.9)	1.63 [0.92–2.90]	0.09	–	129 (48.5)	12 (44.4)	0.92 [0.59–1.42]	0.69
Senegal	98 (35.3)	7 (46.7)	1.37 [0.82–2.83]	0.40		129 (46.4)	12 (80.0)	1.72 [1.30–2.29]	0.01	3.31 [1.05–10.45] p = 0.04
South America	84 (36.7)	21 (32.8)	0.90 [0.63–1.29]	0.569		108 (47.2)	33 (51.6)	1.09 [0.83–1.44]	0.53
Peru	87 (35.5)	18 (37.5)	1.06 [0.70–1.58]	0.79		114 (46.5)	27 (56.3)	1.21 [0.91–1.60]	0.22
Condition during travel
Contact with children	18 (30.5)	87 (37.2)	1.20 [0.83–1.73]	0.33		27 (45.8)	114 (48.7)	1.06 [0.78–1.45]	0.69
Contact with animals	52 (31.0)	53 (42.4)	1.37 [1.01–1.86]	0.046	–	77 (45.8)	64 (51.2)	1.12 [0.88–1.42]	0.36
Type of accommodation
Very clean	34 (79.1)	9 (20.9)	rfr	rfr	rfr	22 (51.2)	21 (48.8)	rfr	rfr
Clean	84 (65.1)	45 (34.9)	1.49 [0.96–2.29]	0.07	–	73 (56.6)	56 (43.4)	0.86 [0.53–1.40]	0.54
Basic	70 (57.9)	51 (42.2)	1.81 [1.16–2.82]	<0.01	1.43 [1.02–1.99], p = 0.04	57 (47.1)	64 (52.9)	1.13 [0.68–1.86]	0.65
Mission during travel
Humanitarian mission in an orphanage	94 (39.2)	11 (20.8)	0.60 [0.41–0.87]	<0.01	–	115 (47.9)	26 (49.1)	1.02 [0.76–1.39]	0.88
School renovation	49 (31.4)	56 (40.9)	1.29 [0.96–1.75]	0.09	–	79 (50.6)	62 (45.3)	0.89 [0.70–1.14]	0.36
Supply of medical equipment and health advice	86 (35.0)	19 (40.4)	1.16 [0.77–1.76]	0.48		117 (47.6)	24 (51.1)	1.07 [0.79–1.46]	0.66
Internships in hospital	86 (36.3)	19 (33.9)	0.94 [0.64–1.37]	0.74		112 (47.3)	29 (51.8)	1.10 [0.82–1.46]	0.54
Internships in surgery department	97 (35.3)	8 (44.4)	1.29 [0.82–1.12]	0.45		130 (47.3)	11 (61.1)	1.29 [0.88–1.91]	0.26
Duration of travel (mean ± SD)	40.5 ± 10.7	41.8 ± 11.9	t = −0.95	0.34		39.6 ± 10.7	42.6 ± 11.4	t = −2.35	0.02	–
Acquisition of respiratory pathogens275
Rhinovirus	86 (36.6)	11 (27.5)	0.78 [0.51–1.19]	0.25		NA	NA	NA	NA
At least one respiratory virus	84 (37.2)	13 (26.5)	0.75 [0.50–1.10]	0.14	–	NA	NA	NA	NA
Haemophilus influenzae	79 (34.7)	18 (38.3)	1.10 [0.73–1.67]	0.64		NA	NA	NA	NA
Klebsiella pneumoniae	89 (34.8)	8 (42.1)	1.22 [0.65–2.29]	0.53		NA	NA	NA	NA
Staphylococcus aureus	78 (35.0)	19 (36.5)	1.04 [0.70–1.55]	0.83		NA	NA	NA	NA
Streptococcus pneumoniae	87 (33.5)	10 (66.7)	2.70 [1.10–6.62]	0.03	2.79 [1.13–6.88], p = 0.03	NA	NA	NA	NA
At least one respiratory bacteria	56 (34.4)	41 (36.6)	1.06 [0.78–1.45]	0.70		NA	NA	NA	NA
Acquisition of digestive pathogens274
Enteroaggregative Escherichia coli	NA	NA	NA	NA		78 (48.2)	58 (51.8)	1.08 [0.85–1.37]	0.55
Enteropathogenic E. coli	NA	NA	NA	NA		110 (49.3)	26 (51.0)	1.03 [0.77–1.40]	0.83
At least one gastrointestinal bacteria	NA	NA	NA	NA		68 (48.9)	68 (50.4)	1.04 [0.74–1.47]	0.81
At least one gastrointestinal pathogen	NA	NA	NA	NA		64 (48.9)	72 (50.4)	1.04 [0.74–1.47]	0.81

NA: not applicable, RR: relative risk, aRR: adjusted relative risk, CI: confidence interval, p: p-value, rfr: reference.

Discussion

Most students in our study traveled to low-income tropical countries for about two months and participated in humanitarian missions in relatively basic housing conditions, in close contact with local children and animals. Two-thirds of students reported gastrointestinal symptoms (notably diarrhea, abdominal pain, and nausea-vomiting, which are suggestive of gastroenteritis). One-third of students reported respiratory symptoms (notably rhinitis, a sore throat, and a cough, suggesting an upper respiratory tract infection). These symptoms appeared within two weeks of their arrival at the destination of their internship. Moreover, we observed a low proportion of travel-associated vaginal symptoms. Overall, symptoms were relatively mild, with fewer than 5% of students requiring antibiotics, and most symptoms resolved before students came back to France. This result is consistent with other studies realized on medical students abroad, confirming that despite reinforced pre-travel counseling, travel-associated respiratory infections and travelers’ diarrhea were very frequent among medical students, who were fully aware of how to prevent these illnesses (Goldsmid et al., 2003, Meltzer et al., 2019, Sharafeldin et al., 2010, Angelin et al., 2015).

We found a significant acquisition of human rhinovirus and S. pneumoniae, as reported previously among international travel among Hajj pilgrims (Hoang and Gautret, 2018, Jennings et al., 2015, Gautret et al., 2016). We also observed a high acquisition rate of EAEC (40.9%) and EPEC (18.6%) among health students, as documented in other studies realized in different populations of domestic and international travelers (Brehm et al., 2020; van Hattem et al., 2019; Paredes-Paredes, 2011; Paschke, 2011; Schaumburg, 2010; Sow, 2018).

Our results showed that respiratory symptoms were significantly more frequent in females. We have no explanation for this observation. Interestingly, the travel destination was distinctly associated with symptoms. Travel to India and Senegal was a risk factor for diarrhea, while travel to Vietnam was a risk factor for respiratory symptoms. Our results are discordant with those of a previous study on 649 international travelers showing that respiratory infections (sore throat or cough) were significantly increased in travelers returning from the non-tropical regions (7.6%) than those from tropical regions, including Vietnam (2.0%) (Pai and Lai, 2008). Differences by travel destinations are also known to be relative to the incidence of travelers’ diarrhea. This result was in line with most other studies that have also found that traveling to the Indian subcontinent was the highest relative risk for diarrhea, followed by African regions (Paschke et al., 2011, Zamarrón Fuertes et al., 2010, Belderok et al., 2011, Hill and Beeching, 2010, Rack et al., 2005).

In addition, the observed correlation between respiratory symptoms and very basic accommodation conditions suggests that precarious housing conditions may encourage respiratory infections. A significant association between the acquisition of S. pneumoniae and respiratory symptoms was also observed in cohorts of Hajj pilgrims (Hoang et al., 2019).

We observed no significant association between E. coli acquisition and diarrhea in our study, in contrast to other studies where EAEC or EPEC has been reported to be more frequent in travelers with diarrhea returning from several geographical areas (Schaumburg et al., 2010, Paschke et al., 2011, Lääveri et al., 2016). This may be explained by the onset of gastrointestinal symptoms occurring early during the trip, while sampling was performed on return several weeks later. Furthermore, asymptomatic carriage of potential pathogens was also observed in participants. In a study by Adachi et al., EAEC was detected in the stools of 26% of patients with traveler’s diarrhea returning from Mexico, Jamaica, or India (Adachi et al., 2001). On the other hand, a recent case-controlled study conducted on German and Dutch travelers showed that EAEC detection was not significantly different in diarrheal persons and asymptomatic controls. However, the prevalence of this bacterium among participants suffering from diarrhea during international travel was high (40.0%) (Schaumburg et al., 2010). Such results are in line with ours.

Our study has a few limitations. First, this study was monocentric and conducted on a very specific population of travelers, which impairs the generalization of our findings. Also, qPCR does not differentiate between dead and viable microorganisms. Finally, sampling was realized during the week preceding departure, and during the week following return, samples at onset of symptoms were not available. Among students who reported clinical symptoms early after their arrival abroad, we are unsure whether these students were infected before or after departure since incubation times of diseases are very different and may vary (Supplementary Table S3). Likewise, post-travel samples were collected at a significant time after the onset of symptoms during travel, and responsible pathogens may have been partly cleared.

This study makes it possible to identify the main infectious diseases linked to travel in a group of French medical students undertaking an internship abroad and the risk factors upon which to base targeting students for reinforced pre-travel advice. A sampling at the time of the onset of symptoms should be carried out in future studies to better understand the relationship between the carriage of pathogens and symptoms.

Transparency declaration

The authors declare that there are no conflicts of interest.

Funding

This study was supported by the Institut Hospitalo-Universitaire (IHU) Méditerranée Infection, the National Research Agency under the “Investissements d’avenir” programme, reference ANR-10-IAHU-03, the Région Provence Alpes Côte d’Azur and European funding FEDER PRIMI.