Journal of Travel Medicine, 2023, 1–9

https://doi.org/10.1093/jtm/taad117
Original Article

Original Article

The Ready-To-Go Questionnaire predicts health

outcomes during travel: a smartphone
application-based analysis

Downloaded from https://academic.oup.com/jtm/article/30/8/taad117/7260577 by guest on 07 January 2025

Julian D. Maier , MD1,*, Alexia Anagnostopoulos, MD MPH1, Anna Gazzotti, MD1,
Silja Bühler, MD MSc2, Vasiliki Baroutsou, MSc1, Christoph Hatz, MD1,3,4,
Milo A. Puhan, MD PhD5, Jan Fehr, MD1,† and Andrea Farnham, PhD1,†
1 Department of Public & Global Health, Division of Infectious Diseases, Epidemiology, Biostatistics and Prevention

Institute, University of Zurich, Zurich, Switzerland, 2 Division of Hygiene and Infectious Diseases, Institute of Hygiene and
Environment, Hamburg, Germany, 3 Swiss Tropical and Public Health Institute, Basel, Switzerland, 4 University of Basel,
Basel, Switzerland and 5 Department of Epidemiology, Epidemiology, Biostatistics and Prevention Institute, University of
Zurich, Zurich, Switzerland
*To whom correspondence should be addressed. Email: julian.maier@uzh.ch
† Jan Fehr and Andrea Farnham contributed equally.

Submitted 14 July 2023; Revised 24 August 2023; Editorial Decision 29 August 2023; Accepted 31 August 2023

Abstract
Background: The Ready-To-Go (R2G) Questionnaire is a tool for rapid assessment of health risks for travel
consultation. This study aims to assess the utility of the R2G Questionnaire in identifying high-risk travellers and
predicting health events and behaviour during travel in the TOURIST2 prospective cohort.
Methods: TOURIST2 data were used to calculate the R2G medical and travel risk scores and categorize each
participant based on their risk. The TOURIST2 study enrolled 1000 participants from Switzerland’s largest travel
clinics between 2017 and 2019. Participants completed daily smartphone application surveys before, during and
after travel on health events and behaviours. We used regression models to analyse incidence of overall health
events and of similar health events grouped into health domains (e.g. respiratory, gastrointestinal, accident/injury).
Incidence rate ratios (IRR) are displayed with 95% confidence intervals (95% CI).
Results: R2G high-risk travellers experienced significantly greater incidence of health events compared to lower-risk
travellers (IRR = 1.27, 95% CI: 1.22–1.33). Both the medical and travel scores showed significant positive associations
with incidence of health events during travel (IRR = 1.11, 95% CI: 1.07–1.16; IRR = 1.07, 95% CI: 1.03–1.12, respectively),
with significant increases in all health domains except skin disorders. Medical and travel risk scores were associated
with different patterns in behaviour. Travellers with chronic health conditions accessed medical care during travel
more often (IRR = 1.16, 95% CI: 1.03–1.31), had greater difficulty in carrying out planned activities (IRR = –0.04, 95%
CI: –0.05, –0.02), and rated their travel experience lower (IRR = –0.04, 95% CI: –0.06, –0.02). Travellers with increased
travel-related risks due to planned travel itinerary had more frequent animal contact (IRR = 1.09, 95% CI: 1.01–1.18)
and accidents/injuries (IRR = 1.28, 95% CI: 1.15–1.44).
Conclusions: The R2G Questionnaire is a promising risk assessment tool that offers a timesaving and reliable means
to identify high-risk travellers. Incorporated into travel medicine websites, it could serve as a pre-consultation
triage to help travellers self-identify their risk level, direct them to the appropriate medical provider(s), and help
practitioners in giving more tailored advice.

Key words: behaviour, high-risk traveller, mHealth, risk stratification, tourist, travel medicine, triage tool

© International Society of Travel Medicine 2023. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc/4.0/),
which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact
journals.permissions@oup.com
2 Journal of Travel Medicine, 2023, Vol. 30, 8

Introduction Methods
Despite the enormous surge in worldwide mobility over the last This study uses data from the prospective cohort study
decades, much still remains unknown about how travel affects TOURIST2, which used a smartphone application to collect data
health and behaviour. The current increase in post-pandemic on health and behaviour of travellers.13,14 The baseline and trip
travel offers a key moment to reflect on how travel medicine data from the TOURIST2 study were used to calculate the R2G
consultations can be improved, as travellers seem to be more scores for each study participant. Patterns in the incidence of
aware of health- and travel-related risks and want to take health events and behaviours during travel were then examined.
appropriate precautions in order to minimize the risk.1 Travel
medicine often relies on non-specific pre-travel advice based on
destination and vaccine recommendations packed into a short TOURIST2 study design and data collection
consultation. An improved understanding of how individual The prospective cohort study TOURIST2 enrolled a total of
traveller characteristics and plans predict health behaviour and 1000 travellers from the travel medicine clinics of Zurich and
outcomes during travel would allow practitioners to provide Basel (Switzerland). Participants were eligible for inclusion if they

Downloaded from https://academic.oup.com/jtm/article/30/8/taad117/7260577 by guest on 07 January 2025

more personalized advice targeted to the needs of their clients. travelled to Brazil, China, India, Peru, Tanzania, or Thailand,
In 2022, Gazzotti et al.2 introduced a novel medical pre-travel could use a smartphone during their trips, planned a travel
risk stratification tool, the Ready-To-Go (R2G) Questionnaire. for ≤4 weeks between 2017 and 2019, and were aged 18 or
This self-assessment tool assigns individuals a travel risk estimate older.13,14 The selected destinations were the most frequently
based on their planned itinerary (e.g. destination country, travel visited by travellers attending the mentioned travel clinics, ensur-
purpose) and their pre-travel health status (e.g. chronic diseases, ing a sufficient sample size. The study was restricted to those
medication). Previous efforts to predict high-risk travellers before travelling 4 weeks or less to ensure comparability (long-term
travelling using questionnaires have been limited. One study travellers are thought to have different risks) and feasibility (the
focused on travellers’ risk perception (TRiP) and another focused likelihood of completing a smartphone app-based survey daily
on the Domain-Specific Risk-Taking Scale (DOSPERT). Neither for months is thought to be low). Participants were asked to com-
included medical conditions, planned activities or incidence of plete a pre-travel questionnaire that collected basic demographic,
health events during travel in their analyses.3 ,4 Furthermore, medical, travel and risk-taking information. Furthermore, they
while pre-travel risk assessment forms are readily available to completed a daily electronic questionnaire about health events
guide practitioners during the pre-travel consultation, they do and behaviours 10 days before, during and for 14 days after
not give information on individual travel risks.5–7 their trip on a smartphone application. The daily electronic ques-
An estimate of a person’s travel risk allows better identifi- tionnaire captured data on 6 health event domains (accidents/in-
cation of individual needs for pre-travel preparation, especially juries, body aches, gastrointestinal symptoms, mental health,
for those at high risk of adverse outcomes. The full spectrum respiratory/flu like symptoms and skin infections or rashes) and
of health risks faced by travellers is wide, ranging from infec- nine health behaviour domains (alcohol/drugs, animal contact,
tious diseases to traffic accidents to exacerbations of existing health care utilization, avoidance of mosquito bites, food con-
chronic diseases.8 Given the limited time frame of pre-travel sumption, transportation use, physical activity, medication use
consultations, it is challenging to adequately address all possible and compliance, and sexual behaviour). Additionally, travellers
risks for every traveller. In the past, researchers have attempted rated their ability to complete planned daily activities and their
to improve the quality of travel consultations by prioritizing overall daily travel experience on a 5-point Likert scale. Every
vaccine recommendations, proposing decision aids and analysing daily questionnaire was automatically geotagged at the time of
traveller perception of vaccines.9–11 Improving health communi- completion using the Global Positioning System. In addition, geo-
cation by using evidence-based tools that prospectively identify location was collected automatically by the application every
high-risk travellers could help travel medicine practitioners give 15 minutes. Participants self-reported each health event based
more targeted advice to travellers at high risk for specific health on its subjective severity using a Likert sliding scale that ranged
issues.12 from 1 (mild) to 4 (severe). A moderate or severe health event
While the R2G Questionnaire has already been validated was defined as one that was rated 3 or 4 by the participant.13,14
in 100 travellers using criteria defined by an expert panel,2 All travellers received a standard pre-travel consultation prior
it is also important to evaluate its utility in predicting actual to their trips. More information on the TOURIST2 study design
health behaviours and outcomes in travellers on a larger scale. and recruitment is described in detail elsewhere.13,14
To do so, we used the data of an existing prospective cohort
study (TOURIST2) tracking 1000 travellers and their health and
behaviour during their travel from Switzerland to Brazil, China, R2G Questionnaire
India, Peru, Tanzania and Thailand.13,14 In this study, we aim to The R2G Questionnaire is a medical triage tool developed to
assess how well the R2G Questionnaire predicts actual traveller identify different levels of travel-related risk. It is designed to be
behaviour and health outcomes during their trips by calculating completed in <5 minutes by any traveller prior to travel using
the R2G risk scores for TOURIST2 travellers. Specifically, we basic information about the planned itinerary and health status.
aim to (i) compare the overall incidence of health events for It was developed by travel medicine experts and validated with
travellers in low-, moderate-, substantial- or high-risk, and (ii) 100 travellers.2 It consists of nine questions, with higher points
determine which health behaviours and adverse health outcomes assigned to ‘riskier’ answers. Six questions are used to calculate
are predicted by the R2G risk scores. the medical risk score and three questions are used to calculate
Journal of Travel Medicine, 2023, Vol. 30, 8 3

Table 1. Definition of R2G risk categories and score cutoffs. The R2G risk categories were assigned according to the risk categories pre-
defined by the developers of the tool.2 The R2G medical and travel risk score groups were defined according to breakpoints observed
in the TOURIST2 data. Conditions that contributed to a high-risk medical score include chronic medical conditions, medication intake,
allergies, and age. Questions that contributed to a high-risk travel score include risky destinations, long travel duration and specific travel
purposes (like volunteer work, visiting friends and relatives, and travelling to remote regions). Full explanation on how the R2G scores
were calculated are shown in Appendix A1–A32

R2G risk categories2 Definitions

Low-risk category Medical and travel risk score each ≤10 points
Moderate-risk category Medical risk score ≤ 20 and travel risk score of 15–50 points
Substantial-risk category Medical risk score ≤ 20 and travel risk score of ≥55 points
High-risk category Medical risk score ≥ 20 points
R2G medical risk score groups
No risk Medical risk score of 0
Low risk Medical risk score of >0–20

Downloaded from https://academic.oup.com/jtm/article/30/8/taad117/7260577 by guest on 07 January 2025

Moderate risk Medical risk score of >20–60
High risk Medical risk score of >60
R2G travel risk score groups
Low risk Travel risk score of ≤100
Moderate risk Travel risk score of >100–130
High risk Travel risk score of >130–160
Very high risk Travel risk score of >160

the travel risk score. Both scores together define the overall risk accounting for varying travel time were used. To assess the
category for each traveller (Table 1). The medical risk score is relationship between R2G scores and incidence of accidents/in-
based on pre-existing medical conditions, current medications, juries, logistic regression models with dichotomized outcomes
allergies, adverse reactions to previous vaccinations, pregnancy were applied, due to the relatively small number of accidents.
and breastfeeding, and age. The travel risk score is based on travel To assess the relationship between R2G scores and the ability to
destination, travel duration and travel purpose.2 complete planned activities or the overall daily travel experience,
linear regression models were applied. The level of statistical
significance was set at P < 0.05. Interquartile ranges (IQR) or
R2G scores calculation and R2G risk category 95% CI were calculated where applicable. Sensitivity analyses
assignment were performed. One for participants where it was not possible
The R2G scores were calculated for each participant that com- to answer the travel purpose question and one by omitting some
pleted the TOURIST2 study.2 We used the baseline TOURIST2 R2G travel risk questions. All statistical analysis was performed
questionnaire to fill out most of the R2G questions; exceptions using R Statistical Software (version 4.2.3).15
are described in Appendix B. After calculating the medical and
travel risk score, each TOURIST2 participant was assigned to
Ethical considerations
the respective pre-defined R2G risk category.2 For analysing the
medical and travel risk score separately, we divided them into The prospective cohort TOURIST2 and subsequent analyses of
four groups (Table 1). the data were approved by the Ethics Commission of the Canton
of Zurich, Switzerland (KEK-ZH-Nr. 2014-0470, BASEC-Nr.
2017-00412). The TOURIST2 study is registered with clinicaltria
Statistical analysis ls.gov under the identifier NCT03262337.
Incidence rate (IR) of health events during travel were calculated
overall and by health domain by dividing the number of events Results
reported by the geotagged survey-days and multiplying by 1000,
resulting in IRs per 1000 travel-days. A survey-day is defined as Study population
a day where a questionnaire was filled out. Sunburn events were 793 participants completed the TOURIST2 study. Participants
common, and therefore only events that were rated 3 (moderate) were slightly more often women (432/793, 54.5%), young
or 4 (severe) were included as a health event. Incidence of (median age: 34.0, IQR: 28.0–50.0), planned approximately
behaviours during travel were calculated by summing up the 2-week trips (median trip days: 16.0, IQR: 14.0–23.0) and
daily number of specific health behaviours performed by each were mostly travelling for tourism (78.7%). Demographic
traveller and dividing by the number of geotagged survey-days, characteristics of study participants are summarized in Table 2.
multiplying by 1000 to obtain IR per 1000 travel-days. Incidence Further TOURIST2 population characteristics are described in
rate ratios (IRR) and associated 95% confidence intervals (95% detail elsewhere.13,14
CI) were calculated to compare IRs at home versus during travel R2G scores were calculated for all 793 participants. Accord-
as well as between R2G categories. To assess the relationship ing to the R2G categorization, 658 (83.0%) participants were
between R2G scores and incidence of health events (overall and considered to have a substantial risk for health events during
domain specific), negative binominal models with offset terms travel and 135 (17.0%) a high risk. R2G high-risk category
4 Journal of Travel Medicine, 2023, Vol. 30, 8

participants visited more than one country only the main destination (e.g. Tanzania) was considered. b Due to the fact that all the study countries included in the TOURIST2 study population were considered higher risk, no travellers fell into the low
Table 2. Baseline study population characteristics adopted from Farnham et al.14 Participants are represented in more than one country, if they travelled to more than one country. 750 out of
travellers had a higher incidence of health events during travel

Switzerland (at home)

(median IR = 1312, 95% CI: 1158–1722) per 1000 travel days

34.0 (27.0–50.0)

16.0 (14.0–23.0)

5.0 (0.0–10.0)
than those in the substantial risk category (median IR = 1000,

115 (95–130)
411 (54.8%)

593 (79.1%)

121 (17.5%)
49 (6.5%)
28 (3.7%)
59 (7.9%)
21 (2.8%)
95% CI: 909–1125) per 1000 travel days (IRR = 1.27, 95% CI:
1.22–1.33). This was driven mainly by an increase in gastroin-

9668
750 testinal, respiratory and body ache symptoms (Figure 1). Travel
purpose could not be answered for 8% of study participants
30.0 (25.0–37.5)

20.0 (15.0–29.0)
(n = 63). Based on a sensitivity analysis, categorizing these trav-

5.0 (0.0–10.0)
ellers as ‘other’ travelling reason versus not assigning additional

124 (91.9%)
84 (62.2%)

19 (15.2%)
85 (70–95)
points (i.e. treating them as beach holiday travellers) did not

2 (1.5%)
3 (2.2%)
Thailand

yield a major difference in the IR and IRR (<0.1%). Therefore,

1688
135

the reason for travel for these participants was assigned as

0
0
‘other’.

Downloaded from https://academic.oup.com/jtm/article/30/8/taad117/7260577 by guest on 07 January 2025

37.0 (28.0–51.0)

15.0 (12.0–19.0)

130 (120–140)
5.0 (0.0–10.0)
123 (54.7%)

186 (82.7%)

39 (17.6%)
High-risk medical score participants
14 (6.2%)
15 (6.7%)
7 (3.1%)
3 (1.3%)
Tanzania

Overall, a medical risk score of 5.0 (IQR: 0.0–5.0) was calculated

2353
225

for the study population. 452 (57.0%) participants had a medical

risk score above 0, indicating that they were travelling with a
31.0 (26.0–41.5)

21.0 (16.5–27.0)

chronic condition. Participants with a medical risk score above

135 (115–145)
5.0 (0.0–15.0)

zero were slightly more likely to be male (60.8% vs. 54.0%) and
57 (60.0%)

82 (86.3%)

17 (18.9%)
4 (4.2%)
4 (4.2%)
4 (4.2%)
1 (1.1%)

had more health events during travel (IR = 1167, 95% CI: 1000–
1134

1313 vs. 1000, 95% CI: 800–1125), but were approximately the
Peru

95

same age as travellers without chronic conditions (aged 35 vs.

33 years). Participants with a moderate medical risk score (>0
37.0 (28.0–53.0)

17.0 (13.0–24.0)

but ≤60, n = 438, 55.2%) had similar travel patterns to those

5.0 (0.0–10.0)
105 (95–115)
73 (50.3%)

96 (66.2%)

16 (11.0%)

22 (15.2%)

without chronic conditions. Participants with a high-risk medical

14 (9.7%)

12 (8.3%)
7 (4.8%)

risk score (>60, n = 14, 1.7%) tended to travel for longer periods
1917
India

of time (19.0 vs. 16.0 days), were more likely to be visiting friends
145

and relatives (VFR) (21.4% vs. 7.8%), did less risky travel overall
all 793 included participants also filled out daily questionnaires while at home in Switzerland

(travel risk score of 100 vs. 115), and were more likely to visit
34.0 (31.5–48.5)

20.0 (15.0–29.0)

Brazil or China (see Table 3). The median incidence of health

5.0 (0.0–15.0)
93 (75–110)
15 (42.9%)

32 (91.4%)

events during travel increased linearly with increasing medical

7 (21.9%)
2 (5.7%)

risk score (Table 3).

China

397

In a negative binomial model controlling for age, sex, planned

35

0
0
0

trip duration, destination, travel purpose and the travel risk

score, the medical risk score was significantly associated with
35.0 (29.0–50.0)

16.0 (14.0–23.0)

110 (100–125)
5.0 (0.0–10.0)

overall incidence of health events (IRR = 1.11, 95% CI: 1.07–

123 (67.2%)
89 (48.6%)

24 (13.1%)

31 (17.3%)

1.16, Table 4), meaning there was a 11% increase in incidence

1 (0.5%)
1 (0.5%)
2 (1.1%)

of health events for every 10 points increase in medical risk

Brazil

2184
183

score. In health domain-specific models, the medical risk score

was also significantly associated with incidence of gastroin-
34.0 (28.0–50.0)

16.0 (14.0–23.0)

testinal symptoms (IRR = 1.13, 95% CI: 1.07–1.18), respiratory

115 (95–130)

and flu-like symptoms (IRR = 1.14, 95% CI: 1.07–1.23), body

5.0 (0.0–5.0)
432 (54.5%)

624 (78.7%)

135 (17.0%)
Overall study

57 (7.2%)
28 (3.5%)
63 (7.9%)
21 (2.6%)

aches (IRR = 1.14, 95% CI: 1.07–1.21) and mental health events
19 341

(IRR = 1.11, 95% CI: 1.03–1.20). Not associated with the medi-
793

cal risk score were incidence of skin disorders (IRR = 1.06, 95%
CI: 0.99–1.14) and occurrence of accidents/injuries (OR = 1.00,
Participants considered to be high risk

95% CI: 0.90–1.10) (Appendix C1–C6).

As incidence of health events may normally be higher for med-
Visiting friends and relatives

ical risk travellers even at home, the IRR of health events during
(R2G high risk category)a, b

travel versus at home (before and after travel) in Switzerland was

Number of survey-days

or moderate risk category.

calculated. For travellers with a moderate or high medical risk

Number of travellers

score, the IR during travel was significantly higher for overall

Planned trip days
Reason for travel

health events (IRR = 1.38, 95% CI: 1.30–1.45), gastrointesti-

Volunteering

Medical scorea
Travel scorea

nal symptoms (IRR = 1.92, 95% CI: 1.69–2.19), respiratory/flu-

Female sex

Business
Tourism

Missing

a Here, if

like symptoms (IRR = 1.37, 95% CI: 1.23–1.52) and symptoms

concerning the skin (IRR = 2.22, 95% CI: 1.90–2.60) compared
Age

to at home. However, the IR was lower during travel than

Journal of Travel Medicine, 2023, Vol. 30, 8 5

Downloaded from https://academic.oup.com/jtm/article/30/8/taad117/7260577 by guest on 07 January 2025

Figure 1. Heat maps showing IRs per 1000 travel days for each health event domain in different medical or travel score groups. Colouring is done
per row.

at home for mental health events (IRR = 0.71, 95% CI: 0.61– High-risk travel score participants
0.83) and no significant difference was seen for accidents/injury Overall, a travel risk score of 115 (IQR: 95–130) was calculated
(IRR = 1.28, 95% CI: 0.95–1.74) and body aches (IRR = 1.11, for the study population. 153 (19.3%) of participants had a high
95% CI: 0.98–1.26). or very high travel risk score, indicating highly risky planned
Higher medical risk scores were also significantly associated travel. Participants with a high or very high travel risk score
with certain health behaviours during travel compared to at were younger (aged 31 vs. 35 years), slightly more often women
home, including higher alcohol and illicit drug consumption (58.2% vs. 53.6%), planned longer trips (19 vs. 16 days), and
(IRR = 1.09, 95% CI: 1.01–1.17), accessing medical care or need- were more likely to travel for volunteering (11.1% vs. 1.9%) and
ing medical help more often (IRR = 1.16, 95% CI: 1.03–1.31). less likely for business (3.9% vs. 8.1%) (Table 5). The median
They had a lower incidence of certain health behaviours, such as incidence of health events during travel increased linearly with
consuming risky foods (IRR = 0.97, 95% CI: 0.95–1.00). They increasing travel risk score (Table 5).
also more frequently reported difficulty in carrying out planned In a negative binomial model controlling for age, sex, planned
activities (beta coefficient = −0.04, 95% CI: –0.05, −0.02), and trip duration, destination, travel purpose and the medical
the overall travel experience was worse (beta coefficient = −0.04, risk score, the travel risk score was significantly associated
95% CI: –0.06, −0.02) (Appendix C7–C14). with overall incidence of health events (IRR = 1.07, 95%
6 Journal of Travel Medicine, 2023, Vol. 30, 8

Table 3. Characteristics of participants grouped according to medical risk score. A higher medical score indicated that they were travelling
with chronic conditions (e.g. high blood pressure, asthma). B = Brazil, C = China, I = India, P = Peru, Ta = Tanzania, Th = Thailand, T = Tourism,
B = Business, V = Volunteering, O = Other

Medical Score No risk (0) Low risk (0–20) Moderate risk (20–60) High risk (>60) All

Participants 341 (43.0%) 317 (40.0%) 121 (15.3%) 14 (1.7%) 793 (100%)
% Women 157 (46.0%) 196 (61.8%) 71 (58.7%) 8 (57.1%) 432 (54.5%)
Age (median) 33.0 (27.0–47.0) 33.0 (27.0–50.0) 42.0 (29.0–55.0) 52.5 (30.3–67.0 34.0 (28.0–50.0)
Country visited B = 77 (22.6%), C = 13 B = 71 (22.4%), C = 12 B = 26 (21.5%), C = 5 B = 5 (35.7%), B = 179 (22.6%),
(3.8%), I = 55 (3.8%), I = 68 (4.1%), I = 19 C = 2(14.3%), C = 32 (4.0%), I = 145
(16.1%), P = 43 (21.5%), P = 30 (15.7%), P = 16 I = 3(21.4%), P = 1 (18.3%), P = 90
(12.6%), (9.5%), Ta = 79 (13.2%), Ta = 38 (7.1%), Ta = 1 (7.1%), (11.3%),
Ta = 104(30.5%), (24.9%), Th = 57 (31.4%), Th = 17 Th = 2 (14.3%) Ta = 222(28.0%),
Th = 49 (14.4%) (18.0%) (14.0%) Th = 125(15.8%)

Downloaded from https://academic.oup.com/jtm/article/30/8/taad117/7260577 by guest on 07 January 2025

Planned trip days 16.0 (14.0–23.0) 16.0 (13.0–23.0) 16.0 (14.0–23.0) 19.0 (11.5–27.3) 16.0 (14.0–23.0)
(median)
Reason for travel T = 279 (81.8%), T = 240 (75.7%), T = 96 (79.3%), B = 3 T = 9 (64.3%), B = 1 T = 624 (78.7%),
B = 23 (6.7%), B = 31 (9.5%), (2.5%), VFR = 13 (7.1%), VFR = 3 B = 58 (7.3%),
VFR = 23 (6.7%), VFR = 25 (7.9%), (10.7%), V = 6 (21.4%), V (0.0%), VFR = 64 (8.1%),
V = 12 (3.5%), O = 4 V = 11 (3.5%), O = 10 (5.0%), O = 3 (2.5%) O = 1 (7.1%) V = 29 (3.7%), O = 18
(1.2%) (3.2%) (2.3%)
Median travel risk 115 (100–130) 115 (95–130) 115 (100–130) 100 (95–127.5) 115 (95–130)
score (IQR)
Median IR of health 1000 (800–1125) 1000 (909–1250) 1312 (1158-1727) 1479 (500–2500) 1036 (1000-1200)
events (95%CI)

Table 4. Negative binominal regression models showing the association between the R2G scores (predictor) and incidence of health events
(outcome). Reference destination is Tanzania (the destination with the lowest incidence of health events) and reference purpose is tourism
(the most common travel purpose).

Predictors IRR (95% CI) P-value

R2G Medical Risk Score 1.11 (1.07–1.16) <0.001∗

R2G Travel Risk Score 1.07 (1.03–1.12) 0.002∗
Age 0.98 (0.97–0.98) <0.001∗
Sex (female = 1) 1.13 (1.00–1.28) 0.056
Planned trip duration 1.00 (0.99–1.00) 0.480
Destination Brazil 1.20 (0.98–1.47) 0.069
Destination China 1.73 (1.21–2.51) 0.003∗
Destination India 1.58 (1.27–1.96) <0.001∗
Destination Peru 1.39 (1.12–1.72) 0.003∗
Destination Thailand 1.47 (1.09–1.97) 0.009∗
Travel Purpose: Business 0.91 (0.71–1.19) 0.490
Travel Purpose: Other 0.94 (0.54–1.79) 0.836
Travel Purpose: Study 1.47 (0.88–2.62) 0.161
Travel Purpose: VFR 0.91 (0.72–1.15) 0.419
Travel Purpose: Volunteer Work 0.76 (0.55–1.07) 0.106

Statistical significance as defined in the methods is shown with ∗ .

CI: 1.03–1.12, Table 4). In health domain-specific negative for overall health events (IRR = 1.57, 95% CI: 1.49–1.66),
binomial models, the travel risk score was also significantly asso- accidents/injury (IRR = 1.94, 95% CI: 1.46–2.59), gastrointesti-
ciated with incidence of gastrointestinal symptoms (IRR = 1.07, nal symptoms (IRR = 2.05, 95% CI: 1.83–2.31), respiratory/flu-
95% CI: 1.01–1.13), accidents and injuries (OR = 1.28, 95% like symptoms (IRR = 1.53, 95% CI: 1.37–1.70), body aches
CI: 1.15–1.44) and body aches (IRR = 1.13, 95% CI: 1.06– (IRR = 1.22, 95% CI: 1.09–1.37) and symptoms concerning
1.21). Not associated with the travel risk score were incidence the skin (IRR = 3.07, 95% CI: 2.57–3.67). However, the IR
of mental health disorders (IRR = 1.03, 95% CI: 0.95–1.12), was lower during travel than at home for mental health events
respiratory and flu-like symptoms (IRR = 1.07, 95% CI: 0.99– (IRR = 0.84, 95% CI: 0.72–0.98).
1.16) and skin disorders (IRR = 1.05, 95% CI: 0.98–1.14) Except for increased reporting of animal contacts (IRR = 1.09,
(Appendix C1–C6). 95% CI: 1.01–1.18), higher travel risk scores were not signif-
For travellers with a high or very high travel score, the IR icantly associated with certain health behaviours during travel
during travel compared to at home was significantly higher (Appendix C7–C14).
Journal of Travel Medicine, 2023, Vol. 30, 8 7

Table 5. Characteristics of participants grouped according to travel risk score. A higher score indicates that they planned more risks
during travel (e.g. travelling to high mountain regions, backpacking). B = Brazil, C = China, I = India, P = Peru, Ta = Tanzania, Th = Thailand,
T = Tourism, B = Business, V = Volunteering, O = Other

Travel Score Low risk (0–100) Moderate risk High risk (130–160) Very high risk (>160) all
(100–130)

Participants 264 (33.3%) 376 (47.4%) 138 (17.4%) 15 (1.9%) 793 (100%)
% Women 139 (52.7%) 204 (54.3%) 78 (56.5%) 11 (73.3%) 432 (54.5%)
Age (median) 34.0 (28.0–48.0) 37.0 (28.0–51.0) 31.0 (25.3–42.8) 30.0 (24.0–37.5) 34.0 (28.0–50.0)
Country visited B = 53 (20.1%), C = 22 B = 109 (29.0%), B = 17 (12.3%), C = 0 B = 0 (0.0%), C = 0 B = 179 (22.6%),
(8.3%), I = 69 C = 10 (2.7%), I = 67 (0.0%), I = 8 (5.8%), (0.0%), I = 1 (6.7%), C = 32 (4.0%), I = 145
(26.1%), P = 3 (1.1%), (17.8%), P = 37 P = 44 (31.9%), P = 6 (40.0%), Ta = 8 (18.3%), P = 90
Ta = 0 (0.0%), (9.8%), Ta = 69 (50.0%), (53.3%), Th = 0 (11.3%),
Th = 117 (44.3%) Ta = 145(38.6%), Th = 0 (0.0%) (0.0%) Ta = 222(28.0%),

Downloaded from https://academic.oup.com/jtm/article/30/8/taad117/7260577 by guest on 07 January 2025

Th = 8 (2.1%) Th = 125(15.8%)
Planned trip days 16.0 (13.0–22.0) 16.0 (13.0–23.0) 19.0 (15.0–24.8) 23.0 (18.5–33.5) 16.0 (14.0–23.0)
(median)
Reason for travel T = 214 (81.1%), T = 294 (78.2%), T = 108 (78.3%), B = 5 T = 8 (53.3%), B = 1 T = 624 (78.7%),
B = 33 (12.5%), B = 19 (5.1%), (3.6%), (6.7%), B = 58 (7.3%),
VFR = 7(2.7%), V = 1 VFR = 43(11.4%), VFR = 14(10.1%), VFR = 0(0.0%), V = 6 VFR = 64(8.1%),
(0.4%), O = 9 (3.4%) V = 11 (2.9%), O = 9 V = 11 (8.0%), O = 0 (40.0%), O = 0 (0.0%) V = 29 (3.7%), O = 18
(2.4%) (0.0%) (2.3%)
Median medical risk 5.0 (0.0–10.0) 5.0 (0.0–15.0) 5.0 (0.0–10.0) 5.0 (0.0–5.0) 5.0 (0.0–5.0)
score (IQR)
Median IR of health 1000 (867–1222) 1000 (826–1111) 1372 (1125-1647) 1889 (950–2400) 1036 (1000-1200)
events (95%CI)

Discussion overall lower enjoyment of travel. While they did have more
This study used data from the TOURIST2 cohort to calculate the adverse mental health outcomes than other travellers, the inci-
R2G scores for each study participant and to identify patterns in dence during travel was lower than at home for all groups,
the incidence of health events and behaviours during travel. In indicating that travelling can improve mental health. However,
our analysis, the R2G scores correlated with overall and health it is important to note that TOURIST2 was a cohort of rela-
domain specific incidence of health events during travel, except tively young travellers. Future studies should implement the R2G
for skin disorders. This indicates that the R2G questionnaire Questionnaire in older populations.
could serve as a tool for travel medicine practitioners to prospec- In our study, high-risk medical travellers were more likely to
tively identify clients at high risk of adverse events during travel. be VFR, who are known to have increased risk of health events
Participants with high-risk medical scores (i.e. travellers with during and after travelling.25–27 Our findings suggest an associa-
chronic diseases) also showed different patterns in behaviour tion between chronic medical conditions and VFR, which could
during travel compared with both low-risk medical scores and offer a partial explanation for this observed pattern. Moreover,
those with high-risk travel scores. The descriptive characteriza- it appears that individuals with chronic health conditions do not
tion of high-risk travellers using the R2G Questionnaire is similar allow these conditions to dictate their choice of destination or
in most respects to known aspects of high-risk travellers, and activities, as reported by other studies.28,29
matches that of previous studies, further confirming the validity
of the R2G Questionnaire.16,17 High-risk travel scores
Those with a high R2G travel risk score based on their planned
High-risk medical scores itinerary showed distinct patterns in traveller characteristics
Those with chronic medical conditions and therefore a higher and health outcomes and had a higher risk of adverse health
medical risk score had a significantly higher incidence of health outcomes during travel. These travellers were slightly more often
events during travel than at home. Similar results were also women, had longer trips, and were volunteering more often. High
reported by other studies,18,19 but to our knowledge this is risk travellers were also characterized by La Rocque et al to
the first analysis that was able to show which types of health have longer trip durations.17 Participants with a higher travel
outcomes are specifically higher in travellers with medical condi- risk score also had more frequent animal contact during travel.
tions, and show that incidence of health events is higher than at An association between animal contact and health events during
home.20–24 Increased incidence of health events was seen only for travel was also reported by Muehlenbein et al.30 Further, the fact
medical risk scores over 20 points, indicating that travellers with that accidents, injuries and body aches happened more often
mild chronic illnesses are not necessarily at higher risk.2 Med- while doing ‘riskier’ trips suggests that these travellers were
ical risk travellers also accessed medical care more frequently, taking more physical risks. Despite these additional risks, no
had increased difficulties in carrying out planned activities, and significant interruption to planned trip activities was seen.
8 Journal of Travel Medicine, 2023, Vol. 30, 8

While it was expected that the R2G travel risk score would changed, the TOURIST2 destinations and risk factors continue
be associated with increasing risk of adverse health outcomes to be highly relevant during the current post-pandemic surge in
during travel, it is surprising that the R2G travel risk score is travel.
not associated with the risky health behaviours during travel that
are part of standard travel medicine advice (e.g. consuming raw Conclusion
or unwashed vegetables, mosquito protection). Further studies
In conclusion, the R2G risk assessment tool offers an effective,
should follow-up with R2G high travel risk travellers to identify
flexible means to quantify medical and travel risk using risk
relevant additional risk behaviours, informing a larger discussion
parameters like chronic medical conditions, travel destinations
within the travel medicine community about what it means to
and purpose of travel, creating a personalized traveller risk
have risky health behaviour during travel.
profile. It also demonstrates the power of mHealth innovation
in testing travel medicine tools detecting high-risk groups by
Using the R2G Questionnaire to inform the travel tracking the actual incidence of health events during travel. In
medicine consultation future studies, the R2G Questionnaire can be incorporated into

Downloaded from https://academic.oup.com/jtm/article/30/8/taad117/7260577 by guest on 07 January 2025

an evidence-based pre-consultation triage system for identifying
Our results show that travellers with chronic medical conditions
travellers at high-risk of specific outcomes. Including such a
and behaviorally high-risk travellers have distinct risk profiles.
tool on the website of the clinic could also help travellers self-
Travellers with chronic conditions should be advised on how
identify whether they need a special travel medicine consultation
to access medical care during travel. High-risk travellers based
for a planned trip. Such a system would allow practitioners to
on planned activities should receive more advice about rabies
focus their limited consultation time on advising travellers about
exposure and risks such as accidents and injuries related to
the health risks most relevant to their personalized risk profile
physical activity. Our data also suggest that standard pre-travel
and planned itinerary, quickly triage low-risk travellers to other
advice on avoiding diarrhoea and gastrointestinal problems is
sources of travel information, and ensure that clients are better
inadequate, as it continues to be a persistent problem.31–33 For
prepared for the health challenges they might face.
high-risk travels, it is important to be properly insured (e.g. annu-
lation, medical, repatriation). The R2G Questionnaire might also
be used to inform evidence-based risk stratification for travel Supplementary data
insurance purposes. Supplementary data are available at JTM online.

Funding
Limitations
This work did not receive any funding, however, the TOURIST2
There was a small number of TOURIST2 travellers for whom
study was supported by the Swiss National Science Founda-
the travel purpose was unknown. However, a sensitivity analysis
tion, Switzerland [grant number 320030_166524/1]. The fund-
testing the degree to which associations would change by shifting
ing body had no role in the analysis of the data or the writing of
the R2G points assigned for certain categories did not show
the manuscript, or the decision to submit the work.
significant differences in the patterns observed (Appendix D).
We did not have data on the R2G questions on pregnancy,
breastfeeding and adverse reaction to previous vaccines. How- Author contributions
ever, those cases are thought to be rare in this setting.28,34,35 Julian Maier (Conceptualization-Equal, Data curation-Equal,
Furthermore, the study only included participants with pre-travel Formal analysis-Equal, Methodology-Equal, Software-Equal,
consultations by travel clinics in Switzerland and only included Visualization-Equal, Writing—original draft-Equal, Writing—
travelling to specific countries. This may not be generalizable review & editing-Equal), Alexia Anagnostopoulos (Writ-
to all travellers or travel destinations. Even though the travel ing—review & editing-Equal), Anna Gazzotti (Writing—
risk score assigns points to specific destinations depending on review & editing-Equal), Silja Bühler (Writing—review &
local risks (including malaria), it does not account for some fresh editing-Equal), Vasiliki Baroutsou (Writing—review & editing-
water related travel risks (e.g. schistosomiasis, leptospirosis). The Equal), Christoph Hatz (Writing—review & editing-Equal),
inclusion of fresh water related health risks should be considered Milo Puhan (Writing—review & editing-Equal), Jan Fehr
in future R2G studies. Since the TOURIST2 study only included (Supervision-Equal, Writing—review & editing-Equal), Andrea
travellers to relatively high-risk countries, we are not able to Farnham (Conceptualization-Equal, Data curation-Equal,
make statements about low- and moderate-risk travellers. In Formal analysis-Equal, Methodology-Equal, Software-Equal,
future iterations of the study, low-risk countries should also be Supervision-Equal, Visualization-Equal, Writing—original draft-
studied to see if the risk points assigned for different destinations Equal, Writing—review & editing-Equal).
are appropriate. Finally, the group of travellers with very high-
risk medical (>60) and travel (>160) scores was very small, Conflict of interest
and results should be interpreted with caution. Future studies
The authors have declared no conflicts of interest.
should be conducted targeting these individuals in particular, as
they may represent particularly important groups for person-
alized advice. It is also important to note that the TOURIST2 Data availability
cohort travelled in 2019, immediately prior to the coronavirus The data that support the findings of this study are available
pandemic. While some attitudes towards risk-taking may have on reasonable request from the corresponding author, JM. The
Journal of Travel Medicine, 2023, Vol. 30, 8 9

data are not publicly available due to potentially identifiable analysis of demographic characteristics, travel destinations, and pre-
information that could compromise the privacy of research travel healthcare of high-risk US international travelers, 2009–2011.
participants. Clin Infect Dis 2012; 54:455–62.
18. Wieten RW, Leenstra T, Goorhuis A, van Vugt M, Grobusch MP.
Health risks of travelers with medical conditions—a retrospective
References
analysis. J Travel Med 2012; 19:104–10.
1. Rahman MK, Gazi MAI, Bhuiyan MA, Rahaman MA. Effect of 19. Baaten GG, Geskus RB, Kint JA, Roukens AHE, Sonder GJ, van den
Covid-19 pandemic on tourist travel risk and management percep- Hoek A. Symptoms of infectious diseases in immunocompromised
tions. PloS One 2021; 16:e0256486. https://doi.org/10.1371/journa travelers: a prospective study with matched controls. J Travel Med
l.pone.0256486. 2011; 18:318–26.
2. Gazzotti A, De Crom-Beer S, Haller S et al. Ready-to-go question- 20. Ellsbury G, Campling J, Madhava H, Slack M. Identifying UK
naire - development and validation of a novel medical pre-travel risk travellers at increased risk of developing pneumococcal infection:
stratification tool. Travel Med Infect Dis 2022; 47:102304. a novel algorithm. J Travel Med 2021; 28:taab063. https://doi.o
3. Tardivo S, Zenere A, Moretti F et al. The traveller’s risk perception rg/10.1093/jtm/taab063.

Downloaded from https://academic.oup.com/jtm/article/30/8/taad117/7260577 by guest on 07 January 2025

(TRiP) questionnaire: pre-travel assessment and post-travel changes. 21. Wyler BA, Young HM, Hargarten SW, Cahill JD. Risk of deaths
Int. Int Health 2019; 12:116–24. due to injuries in travellers: a systematic review. J Travel Med 2022;
4. Farnham A, Ziegler S, Blanke U, Stone E, Hatz C, Puhan MA. Does 29:taac074. https://doi.org/10.1093/jtm/taac074.
the DOSPERT scale predict risk-taking behaviour during travel? A 22. Warner JC, Hatziioanou D, Osborne JC, Bailey DJ, Brooks TJG, Sem-
study using smartphones. J Travel Med 2018; 25:25. per AE. Infections in travellers returning to the UK: a retrospective
5. Chiodini J. Travel risk assessment form. 2018; https://www.janechio analysis (2015–2020). J Travel Med 2023; 30:taad003. https://doi.o
dini.co.uk/wp-content/uploads/2018/08/1.-Travel-risk-assessment- rg/10.1093/jtm/taad003.
form-2018.pdf (8 July 2023, date last accessed). 23. Reinsberg F, Moehlmann MW, Krumkamp R et al. Symptoms of
6. Sanofi Pasteur MSD. Pre travel risk assessment form. 2019; https:// illness during travel and risk factors for non-adherence to malaria
www.gpwebsolutions-host.co.uk/734b/files/2019/08/PRE-TRAVE prophylaxis—a cross-sectional study in travellers from Germany. J
L-RISK-ASSESSMENT-FORM.pdf (8 July 2023, date last accessed). Travel Med 2023; 30:taad055. https://doi.org/10.1093/jtm/taad055.
7. Campbell. Pre-travel-questionnaire. 2014; https://tarporleygps.gpsu 24. Pisutsan P, Soonthornworasiri N, Matsee W et al. Incidence of health
rgery.net/wp-content/uploads/sites/296/2014/10/Pre-Travel-questio problems in travelers to Southeast Asia: a prospective cohort study.
nnaire.pdf (8 July 2023, date last accessed). J Travel Med 2019; 26:taz045. https://doi.org/10.1093/jtm/taz045.
8. Piyaphanee W, Stoney RJ, Asgeirsson H et al. Healthcare seeking 25. Askling HH, Rombo L, Andersson Y, Martin S, Ekdahl K. Hepatitis
during travel: an analysis by the GeoSentinel surveillance network a risk in travelers. J Travel Med 2009; 16:233–8.
of travel medicine providers. J Travel Med 2023; 30:taad002. https:// 26. Gier B, Suryapranata FST, Croughs M et al. Increase in imported
doi.org/10.1093/jtm/taad002. malaria in the Netherlands in asylum seekers and VFR travellers.
9. Steffen R, Chen LH, Leggat PA. Travel vaccines—priorities deter- Malar J 2017; 16:60.
mined by incidence and impact. J Travel Med 2023; 21:taad085. 27. Ericsson CD, Hatz C, Leder K et al. Illness in travelers visiting friends
https://doi.org/10.1093/jtm/taad085. and relatives: a review of the GeoSentinel surveillance network. Clin
10. McGuinness SL, Eades O, Seale H, Cheng AC, Leder K. Pre- Infect Dis 2006; 43:1185–93.
travel vaccine information needs, attitudes, drivers of uptake and 28. Hochberg NS, Barnett ED, Chen LH et al. International travel
the role for decision aids in travel medicine. J Travel Med 2023; by persons with medical comorbidities: understanding risks and
30(4):taad056. https://doi.org/10.1093/jtm/taad056. providing advice. Mayo Clin Proc 2013; 88:1231–40.
11. Bravo C, Castells VB, Zietek-Gutsch S, Bodin PA, Molony C, Früh- 29. Stienlauf S, Streltsin B, Meltzer E et al. Chronic illnesses in trav-
wein M. Using social media listening and data mining to understand elers to developing countries. Travel Med Infect Dis 2014; 12:
travellers’ perspectives on travel disease risks and vaccine-related 757–63.
attitudes and behaviours. J Travel Med 2022; 29:taac009. https:// 30. Muehlenbein MP, Martinez LA, Lemke AA et al. Unhealthy travelers
doi.org/10.1093/jtm/taac009. present challenges to sustainable primate ecotourism. Travel Med
12. Renshaw A, Lai I. Addressing health communication in the era of Infect Dis 2010; 8:169–75.
alternative truths: the view from medical assistance. J Travel Med 31. Chen LH, Hochberg N. The Pretravel Consultation. CDC Yellow
2022; 29:taab179. https://doi.org/10.1093/jtm/taab179. Book 2024. Atlanta, Georgia, USA: Centers for Disease Control and
13. Baroutsou V, Hatz C, Blanke U et al. TOURIST2 – tracking of urgent Prevention, 2023. https://wwwnc.cdc.gov/travel/yellowbook/2024/
risks in Swiss travellers to the 6 main travel destinations – feasibility preparing/pretravel-consultation (8 July 2023, date last accessed).
and ethical considerations of a smartphone application-based study. 32. Hatz C, Chen LH. 4 - pretravel consultation. In: Keystone JS,
Travel Med Infect Dis 2021; 39:101912. Kozarsky PE, Connor BA et al. (eds). Travel Medicine, 4th edn.
14. Farnham A, Baroutsou V, Hatz C et al. Travel behaviours and London: Elsevier, 2019, pp. 25–30.
health outcomes during travel: profiling destination-specific risks in 33. Brunette GW, Kozarsky PE, Magill AJ et al. (eds). Chapter 2 - The
a prospective mHealth cohort of Swiss travellers. Travel Med Infect pre-travel consultation. In: CDC Health Information for Interna-
Dis 2022; 47:102294. tional Travel 2010. Edinburgh: Mosby, 2009, pp. 19–241.
15. R Core Team. R: A language and environment for statistical com- 34. LaRocque R, Rao S, Yanni E et al. Demographics, medical con-
puting. 4.2.3 (2023-03-15 ucrt), 2023. ditions, and use of immunizations and chemoprophylaxis among
16. Valerio L, Martínez O, Sabrià M et al. High-risk travel abroad international travelers within the global TravEpiNet U.S. National
overtook low-risk travel from 1999 to 2004: characterization and Clinic Network. Int J Infect Dis 2010; 14:e131–2.
trends in 2,622 Spanish Travelers. J Travel Med 12:327–31. 35. Nakayama T, Onoda K. Vaccine adverse events reported in post-
17. LaRocque RC, Rao SR, Lee J et al. Global TravEpiNet: a National marketing study of the Kitasato institute from 1994 to 2004. Vaccine
Consortium of clinics providing care to international travelers— 25:570–6.