Vargová et al.

BMC Psychology (2024) 12:208 BMC Psychology

https://doi.org/10.1186/s40359-024-01687-9

RESEARCH Open Access

War-related stress scale

Lenka Vargová1,2, Bibiána Jozefiaková3, Martin Lačný4 and Matúš Adamkovič2,5,6*

Abstract
Background The current war in Ukraine has affected the well-being of people worldwide. In order to understand
how difficult the situation is, specific stressors associated with war need to be measured. In response, an inventory of
war-related stressors including its short form, has been developed.
Methods A list of potential war-related stressors was created, and the content validity of each item assessed. The list,
along with other validated scales, was administered to a representative sample of the Slovak population (effective
N = 1851). Exploratory factor analysis, confirmatory factor analysis, convergent validity analysis and network analysis
were carried out to determine the optimal scale (long and short form) focused on war-related stressors.
Results The full version of the scale consists of 21 items, further divided into three factors: society-related stressors,
person-related stressors, and security-related stressors. The short version of the scale comprises nine items loaded
onto one factor. These items cover concerns for one’s safety and future, access to necessities, potential worsening
of the economic situation, and the risk of conflict escalation, including a nuclear threat. The results of the network
analysis indicate that concern about escalation and fear of an economic crisis play a central role.
Conclusions The scale attempts to encompass a wide spectrum of areas that are affected by war and its potential
consequences on individuals who reside outside the conflict zone. Given the complexity of the issue, researchers are
invited to modify the scale, tailoring it to specific cultural, geographical, and temporal contexts.
Keywords War, War-related stress, Stressors, Mental health

Background
The world had barely recovered from the COVID-19
pandemic when the Russian Federation invaded Ukraine
in February 2022. According to the DSM-5, war is con-
sidered a significant stressor that meets the criteria of
trauma. Trauma has been defined as “actual or threat-
*Correspondence: ened death, serious injury, or sexual violence” [1]. Expo-
Matúš Adamkovič sure to war can thus be related to various mental health
matho.adamkovic@gmail.com disorders such as anxiety, depressive disorders, trauma
1
Faculty of Education, University of Presov, Prešov, Slovakia
2
Institute of Social Sciences, Centre of Social and Psychological Sciences, and stressor-related disorders (acute stress disorder or
Slovak Academy of Sciences, Košice, Slovakia post-traumatic stress disorder), and addictive disorders
3
Olomouc University Social Health Institute, Palacky University, Olomouc, [1]. According to WHO estimates, the prevalence of
Czechia
4
Institute of Political Science, Faculty of Arts, University of Presov, Prešov, mental disorders in a conflict affected population stands
Slovakia at 22.1%; with a 13% prevalence of mild depression, anxi-
5
Faculty of Humanities and Social Sciences, University of Jyväskylä, ety, and PTSD, and a 4% prevalence of their moderate
Jyväskylä, Finland
6
Faculty of Education, Charles University, Prague, Czechia forms [2, see also 3].

© The Author(s) 2024. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use,
sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and
the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this
article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included
in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will
need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The
Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available
in this article, unless otherwise stated in a credit line to the data.
Vargová et al. BMC Psychology (2024) 12:208 Page 2 of 11

Mental health problems and stress can also affect peo- their (potential) consequences. These areas encompass
ple living outside the conflict area [4, 5] and is caused by the negative outcomes of the situation on people’s lives,
a number of war-related threats. Firstly, people’s sense which can be labeled as war-related stress.
of safety is disrupted with those living in surround-
ing countries and further afield fearing conflict escala- Methods
tion or nuclear war [6]. This is especially the case when Scale construction
the threat is definable and proximate [7]. The associ- The scale construction followed scale development
ated crisis can also have a global economic impact that guidelines [e.g., 24] that include (1) determining what
involves higher inflation, supply chain disruptions, stock the scale should measure and clarifying the construct’s
swings, reductions in investment and economic uncer- content, (2) generation of an item pool and item qual-
tainty [8]. People are thus affected by the worsening of ity check, (3) public and expert evaluation of the items’
the economic situation, potential lack of goods and com- content validity, (4) setting the measurement format, (5)
modities (even essential products), disrupted medical a piloting and understandability check, (6) psychometric
services, community consequences (e.g., functioning of properties and (7) scale length reduction.
public institutions, dealing with migration, etc.; [9] and (1) Firstly, the areas of life which could be most affected
the often difficult-to-predict development of the politi- by war were identified. The focus was on the areas that
cal situation [10]. The loss of purchasing power among cause the most stress for people who are not living
citizens can mean subsequent risks to political stability as directly in a warzone or a military occupied country.
well [11]. According to the McKinsey Global Survey [12] The areas were identified and clarified using a combina-
on economic conditions, geopolitical conflicts remain the tion of the following steps: (a) literature screening; (b) the
top-cited risk to global economic growth over the next 12 research team assumed that the stress caused by war is
months, while inflation continues to be in second place. similar to the stress related to the COVID-19 pandemic
Even populations that are not directly involved in con- in many aspects. The primary inspiration was thus drawn
flicts face instability, fear and various worries which can from the validated measures of COVID-19 stress [e.g., 25,
contribute to a mental health burden [13]. Despite the 26]; (c) a non-systematic screening of online discussion
scarcity of studies examining the mental health effects of content (e.g., online journals, social networks) to poten-
war on people living outside conflict areas, the available tially identify further stressful topics. (2) Based on this,
evidence suggests the impairment of mental health (e.g., an initial list of items was created, followed by a discus-
increased likelihood of depression and anxiety, PTSD, or sion to establish which areas were not covered sufficiently
a worsened quality of life) as a result of enduring stress- and which items were redundant or low on content valid-
ors related to the war, consequential impacts and global ity. A similar discussion was held with two groups of
threats [14–17]. The mental health impact of war has undergraduate students. The list was continually updated
been observed in directly-affected Ukrainian citizens as throughout this process. The resulting version consisted
well as in populations residing outside the conflict zone of 22 statements. (3) The content validity of the 22 state-
[18]. ments were independently rated by five experts (all expe-
Civilians and refugees from war-affected areas facing rienced in quantitative research and survey designs; two
complex traumas as well as those outside the conflict of them also working in clinical practice). The ratings can
areas watching the war and suffering from helplessness be found at https://osf.io/f9kbx as well as in the analytic
and hopelessness are at risk for impaired mental health code. (4) Given the nature of the statements, a progres-
[19]. The variations of impairment may arise from dif- sively increasing unidimensional response format was
ferent stress factors they have to face, emphasizing the chosen with seven response options (ranging from 1 = No
importance of applying distinct interventions and poli- concerns/difficulties; 7 = Great concerns/difficulties). The
cies [20]. Many countries can be faced with an increased utilized response format allows for easy gradation and
need for mental health care as a result, and it is important discrimination between the options (increasing vari-
to be aware of the situation and prepare the healthcare ance). Indeed, utilizing a wider scale (e.g., 1 to 100) would
system for it [21, 22]. However, studies usually trace emo- potentially increase measurement error and cognitive
tional reactions and the impact of war on people’s mental load on respondents. (5) Before the data collection, the
health retrospectively [23]. Furthermore, there is a dearth items were piloted on a heterogeneous group of volun-
of measures that specifically focus on war-related stress- teers together with other scales administered in the sur-
ors [6], and a scarcity of studies focused on this particu- vey1 (6 and 7). Besides validating the standard version of
lar issue, especially in populations living outside the war
zone [14]. The present study therefore aims to develop 1
Note that the piloting process has not been sufficiently documented as it
was not expected to publish this scale as a stand-alone paper at that time.
and validate a measure covering a wide spectrum of The initial aim was to create a comprehensive list of relevant war-related
areas that are affected by war or an armed conflict and stressors to map the experiences of the Slovak population, not to validate
Vargová et al. BMC Psychology (2024) 12:208 Page 3 of 11

the scale, the aim was also to create and validate a short Measures
form of the scale (fewer than 10 items). A purely explor- The following measures were administered and used to
atory approach was utilized to examine the possible fac- examine the convergent/divergent validity. Depression
tor structure of the full scale. Given there was not only was measured using the Quick Inventory of Depressive
one solid a priori theoretical justification, a data-driven Symptomatology (QIDS-16-SR) [27]. The QIDS-16-SR
solution was preferred (none of the several predictions consists of 16 items (each item represents a symptom of
was found to be favorable) to see how the items would depression) covering nine depression symptoms from
cluster into factors. A combination of a data-driven the DSM-5. For each item, the respondent expresses
approach and evaluation of the content validity of the the degree of symptom severity over the last two weeks
items were used to create a short version of the scale. on a 4-point scale. Examples of areas covered include
Subsequently, both versions of the scale were cross-val- “Energy level“ or “General interest“. Anxiety was mea-
idated using a confirmatory approach. Furthermore, the sured using the Generalized Anxiety Disorder Screener-7
convergent validity of the scales was examined by corre- (GAD-7; [28]) which consists of seven items. Participants
lating them with a set of external criteria. are asked to rate how often they have been bothered by
any of the presented problems over the last two weeks
Participants and data collection procedure on a 4-point scale ranging from “not at all” to ”nearly
An online questionnaire was administered to a represen- every day”. Examples of these items include “Worrying
tative sample of Slovak inhabitants (N = 2127). The par- too much about different things” or “Becoming easily
ticipants were sampled based on quota characteristics annoyed or irritable”. The 10-item Perceived Stress Scale
for gender, age, education, and region and were recruited (PSS; [29]) was used to assess subjectively perceived
by a Slovak agency specialized in online data collec- stress. The PSS items are rated on a 5-point scale, rang-
tion. The data collection was part of a bigger project ing from ”never” to ”very often”. These include items such
and thus included several scales. Ethical permission was as “In the last month, how often have you felt nervous
granted by the Ethics Committee at the Faculty of Arts, and stressed?”. Sleep difficulties were measured using
University of Presov. Informed consent was obtained the Insomnia Severity Index (ISI; [30]). The ISI consists of
from each participant prior to the data collection. The seven items which assess the different aspects of insom-
study was conducted in compliance with the Declara- nia as well as assessing participants’ perception of noc-
tion of Helsinki guidelines. The data was screened for turnal and diurnal symptoms of insomnia. These items
missing values and careless respondents who were sub- include “How worried/distressed are you about your cur-
sequently excluded (see https://osf.io/zut2m https://osf. rent sleep problem?”. Loneliness was measured using the
io/zut2m?%20view_only=0a11ac371e3145c39c4b261e6 short form of the Loneliness Scale (USL-8; [31]) which
54ff001). The effective sample consisted of 1851 partici- consists of eight items. Examples of items were “People
pants. The demographic characteristics of the sample are are around me but not with me” or “I feel isolated from
presented in Table 1. others”. The Brief Resilience Scale (BRS; [32]) was used to
measure resilience. The BRS consists of six statements
which are rated on a scale from 1 = ”strongly disagree”
the measure or to group the items into factors. Please also note that the
scale was part of a longer survey for the purposes of a longitudinal project
to 5 = ”strongly agree”. Examples of these items include “I
focused on the mental health of Slovak inhabitants (APVV-20-0319). tend to bounce back quickly after hard times”. COVID-
related anxiety was measured by the COVID anxiety
Table 1 Demographic characteristics scale (CAS; [33]). The CAS consists of seven items that
Variable Percent- were answered on a 5-point scale, with a higher num-
age or
Mean ± SD
ber representing higher levels of anxiety. This included
Gender (female) 53.81%
items such as: “I have trouble relaxing when I think about
Age 44.36 ± 15.13 COVID-19”. Items adapted from the COVIDiSTRESS
Partner status (married/in a relationship) 70.45% survey [34] were used to measure COVID-related stress.
Education (university degree) 27.66% The items in this questionnaire are formulated as state-
Residence (urban) 60.54% ments assessing the presence of concerns and difficul-
Economic status (employed) 62.55% ties in various areas possibly affected by the COVID-19
Subjective socioeconomic status 5.39 ± 1.39 pandemic, e.g., “difficulties in daily functioning” or
Equivalized household net income per month (in euros) 725 ± 445 “worrying about getting infected”. The statements were
Note: Subjective socioeconomic status was measured using a Cantril ladder; the rated on a 7-point scale, with a higher number indicat-
scores ranged from 1 to 10 with a higher score indicating a higher subjective ing a higher level of stress. For more detailed information
socioeconomic status. At the time of the data collection (March 2022), the
average monthly gross income in Slovakia was 1212 euros (Statistical Office of about the measures see https://osf.io/bskr4 https://osf.io/
the Slovak Republic, 2024), which works out at around 850 euros net bskr4?view_only=533d3324476e42018d661813b6ecace8.
Vargová et al. BMC Psychology (2024) 12:208 Page 4 of 11

The descriptive statistics and reliabilities in the form of (2) Exploratory dataset 2 served to find an optimal
omega total coefficients are summarized in Table 2. short-scale version of the measure and for this, a
one-factor solution was sought. In the first step
Statistical analysis of this process, five researchers (the authors and
The dataset was randomly split into three equal parts two senior psychology researchers) independently
(each N = 617), with two of them being exploratory and rated the content validity of each item. Given the
one confirmatory. high inter-rater agreement (ICC = 0.87, p <.001),
the sum score of this evaluation was calculated for
(1) Exploratory dataset 1 was used to determine the each item and the distribution of the scores was
number of factors and find an optimal factor solution then examined. A cut-off of 40 was used (potential
for the full version of the scale. The KMO index range of scores = 10–50) to select the items with
and a Bartlett’s test were computed to assess the the highest content validity. These items were
factorability of the dataset. The KMO showed a then modeled using Confirmatory Factor Analysis
very high sampling adequacy with an overall value (CFA), assuming a one-factor solution. The CFA
of 0.97 (the lowest value = 0.94). The Bartlett’s test was estimated using the weighted least square mean
showed that the tested correlation matrix was not an and variance adjusted method, treating the items
identity matrix (χ2(231) = 9721.69, p <.001). A parallel as ordinal. The parameters of the model’s (mis)fit,
analysis [35] was run to determine the number of the chi-square and approximate fit indices (CFI,
factors. Based on the results, the optimal number of TLI, RMSEA, SRMR), as well as the factor loadings,
factors was found to be three. In the case of potential residual matrix, and modification indices were then
non-interpretability of the three-factor solution, examined. The potential sources of the model’s misfit
several additional exploratory factor analyses were were addressed to the level that was theoretically
run with a hypothesized number of factors ranging justifiable and the model was then re-estimated. The
from one to seven. The exploratory factor analyses resulting solution was then cross-validated on the
were estimated using the weighted least squares confirmatory dataset. In order to examine how these
method with a geominQ orthogonal rotation, and items are mutually connected, the items were further
the variables treated as polychoric. In addition, a modeled using the network analysis with the exact
network analysis was used (e.g [36]) to examine how same settings as previously described.
well the items were connected and which of them (3) Confirmatory dataset 1 was used to cross-validate
play a more central/peripheral role in the network. the findings. The three-factor model as well as the
The network was estimated using the EBICglasso short-version one-factor model were estimated using
method and by setting the tuning parameter to 0.50 CFA2. As before, the model-data fit was assessed
to get a sparse network. The items were screened for based on the chi-square test and the approximate
their factor loadings and cross-loadings, as well as for fit indices were examined. Two network models
their strength parameter. An item was excluded if its (corresponding to the list of items for the two factor
highest factor loading did not exceed the threshold models) were also computed to see how well the
of 0.40 and had the lowest strength index at the same results replicated.
time. Given the exploratory nature of the study,
items with high cross-loadings were not excluded For all three datasets, the sum score of the measures was
but classified to the factor that included more similar correlated with a set of external criteria3– the scores for
items content-wise. The factor solution was then depression, anxiety, general stress, insomnia, loneliness,
cross-validated on the confirmatory dataset. resilience, COVID-related anxiety, and COVID-related
stress.
Table 2 Descriptive statistics and reliabilities of other scales Finally, two network analyses were computed to gain a
M SD Potential range ωtotal better insight into how the specific indicators are mutu-
Depression 1.59 0.51 1–4 0.90 ally related. The first one included all the items involved
Anxiety 1.68 0.66 1–4 0.95 in the resulting factor model from the first exploratory
Perceived stress 2.54 0.76 1–5 0.75 dataset, while the second one included the items of the
Insomnia 2.14 0.79 1–5 0.94 short version of the scale as derived from the second
Loneliness 3.23 0.91 1–7 0.78
2
Resilience 3.16 0.61 1–5 0.85 With N = 617, alpha = 0.05, RMSEA0 = 0.04, RMSEAA = 0.08, and DF = 186
(three-factor solution) / 27 (one-factor solution), the power to detect the
COVID-related anxiety 2.14 1.02 1–5 0.93
difference between RMSEA0 and RMSEAA converges to 100%.
COVID-related stress 3.38 1.23 1–7 0.94 3
With N = 617, alpha = 0.05, and using a two-sided test, the study design had
Note: For all constructs, higher scores indicate higher levels of the construct an 80% power to detect a correlation coefficient of 0.11.
Vargová et al. BMC Psychology (2024) 12:208 Page 5 of 11

exploratory dataset. Since network analysis is usually an theoretically well-justified. The estimated network of the
exploratory technique, and has been reported as such, items revealed that fear of the war spreading worldwide
these networks were estimated on the whole dataset to or to the country where the participant resided in addi-
achieve better stability and accuracy of the estimates. tion to fear of an economic crisis, were the most central
The analyses were performed in R with psych [37], indicators of the war-related stress construct.
lavaan [38] and bootnet [39] serving as the main pack-
ages. The data and analytic workflow are documented at Cross-validation of the results
https://osf.io/jv2np/. The three-factor solution retained from the first
exploratory dataset showed an unsatisfactory fit:
Results χ2(186) = 2087.76, p <.001; CFI = 0.93, TLI = 0.92,
Factor structure - full version RMSEA = 0.13 95% CI [0.12, 0.14], and SRMR = 0.06.
The parallel analysis suggested that the optimal number The fit improved once the covariance terms (1 and
of factors is three. In comparison to other factor solu- 2, 7 and 18, 13 and 14, 13 and 15, and 14 and 15) were
tions, the three-factor model had sufficient psychometric added: χ2(181) = 1582.91, p <.001; CFI = 0.95, TLI = 0.94,
properties (TLI = 0.95; RMSEA = 0.06, 95%CI [0.06, 0.07]; RMSEA = 0.11 95% CI [0.11, 0.12], and SRMR = 0.06
SRMR = 0.02) and was well-interpretable in terms of clar- although it still did not reach satisfactory values (espe-
ity of the factors. There was only one item (difficulty in cially in the chi-square statistics). There were no further
distinguishing between true and fake information in the major, theoretically justifiable modifications which could
media) that had a factor loading below 0.40 as well as have been done. As in the second exploratory dataset, the
being low on centrality indices, especially strength. This short form one-factor solution showed a very good fit in
item was thus excluded from the model and the three- most of the AFIs (CFI = 0.99, TLI = 0.98, RMSEA = 0.11
factor model was re-estimated. The re-estimated fac- 95% CI [0.09, 0.13], and SRMR = 0.03) but still had a sig-
tor model had about the same psychometric properties nificant chi-square test (χ2(22) = 177.69, p <.001). All fac-
(both AFIs and factor loadings) as the previous version tor loadings and descriptive characteristics of the items
although no factor loading was smaller than 0.40. The are summarized in Table 3.
items with high cross-loadings (> 0.40) were categorized
to a factor that was more similar content-wise. In sum- Convergent validity
mary, the three factors are: (1) society-related stressors The correlations of the factors (three-factor model) and
(items no. 7, 8, 9, 10, 16, 17, 18, 20), (2) person-related the total score (short-form model) from the respective
stressors excluding security stressors (items no. 3, 4, 5, 6, exploratory and confirmatory datasets are available in
11, 19, 21), (3) safety-related stressors (items no. 1, 2, 12, Tables 4 and 5. The strongest positive relationship was
13, 14, 15). detected between the war-related stressors (all three
domains of the full version and the summary score of the
Factor structure - short scale shortened version) and (1) COVID-related stressors, (2)
After the raters’ evaluation of the content validity of the pandemic anxiety, (3) and general anxiety and depres-
items, nine items (with a total score equal to 40 or more) sion. There were negative relationships observed in the
were selected for the short form of the scale. Most of the case of resilience, which was also the smallest correlation
selected items had had very high centrality indices when (from − 0.11 to − 0.19 in the confirmatory dataset).
a network involving all the items was estimated in both
the exploratory datasets. The selected items (item no. 1, Network analysis
2, 4, 7, 12, 13, 14, 15, and 17) were then loaded onto a Figures 1 and 2 show the visualizations of the networks,
single factor. This model was estimated using CFA. With centrality indices, their stability, and the difference in the
the exception of RMSEA, the model showed decent AFIs: strength of the items. In the network that involved all 21
CFI = 0.95, TLI = 0.93, RMSEA = 0.21 95% CI [0.19, 0.22], items, items no. 7 and 14 were the highest on the strength
SRMR = 0.07, although it was still disconfirmed based indicator, while items no. 20, 15, and 3 were significantly
on the chi-square test (χ2(27) = 737.20, p <.001). After lower in strength compared to the rest of the items. In
examining the residual matrix and modification indices the network involving the short-scale items, items no.
(MIs > 10), a covariance term was added between the fol- 14 and 13 had the highest strength indicator. Additional
lowing items: 1 and 2, 7 and 18, 13 and 14, 13 and 15, and outputs of all the analyses can be found at https://osf.io/
14 and 15. This improved the fit of the model (CFI = 0.99, jv2np/.
TLI = 0.98, RMSEA = 0.11 95% CI [0.10, 0.12], and
SRMR = 0.03) although the chi-square still indicated a sig-
nificant model-data deviation (χ2(22) = 177.50, p <.001).
There were no further modifications which could be
 Vargová et al. BMC Psychology

Table 3 Factor loadings and descriptive characteristics of the items

Item Three-factor solution Short form Descriptives
Society Person Safety
E1 C E1 C E1 C Raters’ score E2 C M SD Skew Kurt
1 Your own safety - - - - 0.47 0.84 45 0.81 0.81 4.13 1.95 -0.12 -1.11
2 Safety of your loved ones - - - - 0.50 0.86 43 0.85 0.85 4.66 1.95 -0.45 -0.94
(2024) 12:208

3 Work/school future - - 0.43 0.76 - - 31 - - 3.94 1.96 -0.05 -1.12

4 Access to essential items (e.g., food, medicine, toiletries) - - 0.43 0.85 - - 43 0.75 0.78 4.14 1.91 -0.12 -1.07
5 Social relationships with your loved ones - - 0.74 0.76 - - 24 - - 3.39 1.93 0.31 -1.05
6 Social relationships in general - - 0.57 0.79 - - 22 - - 4.03 1.88 -0.09 -1.03
7 Economic crisis in [country of residence] 0.90 0.83 - - - - 40 0.77 0.77 5.42 1.62 -0.94 0.14
8 Functioning of public institutions (e.g., governmental bodies, authorities, police) 0.48 0.77 - - - - 29 - - 3.94 1.85 -0.03 -1.00
9 Functioning of the healthcare system 0.69 0.83 - - - - 28 - - 4.66 1.84 -0.42 -0.85
10 Worsening of public health 0.49 0.83 - - - - 24 - - 4.38 1.83 -0.26 -0.92
11 Spending your free time - - 0.74 0.78 - - 14 - - 3.23 1.84 0.36 -0.89
12 Uncertainty about the future development of the situation - - - - 0.44 0.85 45 0.82 0.85 5.05 1.74 -0.71 -0.37
13 Fear of the war moving to [country of residence] - - - - 0.91 0.80 48 0.80 0.83 4.84 1.92 -0.53 -0.85
14 Concern about the war moving to the EU or the world - - - - 0.94 0.80 47 0.80 0.83 4.94 1.87 -0.60 -0.73
15 Nuclear threat - - - - 0.80 0.78 48 0.76 0.78 4.93 1.85 -0.59 -0.69
16 [Country of residence] coping with a migration wave 0.69 0.75 - - - - 35 - - 4.67 1.79 -0.39 -0.77
17 Rising prices (inflation) 0.89 0.67 - - - - 42 0.68 0.61 5.95 1.37 -1.47 1.81
18 Lack of energy resources (e.g., gas, oil, electricity) 0.72 0.82 - - - - 35 - - 5.21 1.75 -0.84 -0.22
19 Worsening of your mental health (e.g., stress, anxiety, depression) - - 0.65 0.75 - - 34 - - 3.47 1.94 0.29 -1.07
20 Development of the political situation in [country of residence] 0.82 0.71 - - - - 28 - - 5.02 1.75 -0.63 -0.47
21 Quality of your life - - 0.50 0.85 - - 37 - - 4.19 1.86 -0.13 -1.02
Note: E1 = exploratory dataset 1; E2 = exploratory dataset 2; C = confirmatory dataset; Raters’ score potential range = 10-50; descriptive characteristics were obtained from the full sample
Page 6 of 11
Vargová et al. BMC Psychology (2024) 12:208 Page 7 of 11

Table 4 Correlations of the three factors with external variables

1 2 3 4 5 6 7 8 9 10 11
1 Society stressors - 0.77 0.75 0.16 0.25 0.20 0.20 0.17 − 0.11 0.30 0.49
2 Person stressors 0.78 - 0.76 0.28 0.37 0.26 0.31 0.31 − 0.19 0.42 0.61
3 Safety stressors 0.75 0.76 - 0.19 0.30 0.23 0.20 0.18 − 0.17 0.34 0.48
4 Depression 0.18 0.31 0.26 - 0.71 0.59 0.62 0.42 − 0.38 0.40 0.38
5 Anxiety 0.26 0.40 0.35 0.69 - 0.69 0.60 0.42 − 0.36 0.51 0.48
6 Stress 0.19 0.24 0.23 0.57 0.63 - 0.48 0.34 − 0.21 0.40 0.44
7 Insomnia 0.24 0.35 0.30 0.60 0.56 0.44 - 0.37 − 0.27 0.44 0.44
8 Loneliness 0.11 0.25 0.14 0.44 0.44 0.40 0.40 - − 0.26 0.35 0.43
9 Resilience − 0.14 − 0.29 − 0.22 − 0.37 − 0.38 − 0.18 − 0.30 − 0.19 - − 0.30 0.23
10 COVID-anxiety 0.34 0.48 0.45 0.48 0.53 0.41 0.48 0.32 − 0.32 - 0.66
11 COVID-stress 0.42 0.57 0.47 0.42 0.48 0.39 0.39 0.37 − 0.28 0.63 -
Note: Correlations from the exploratory dataset are below the diagonal, while correlations from the confirmatory dataset are above it

Table 5 Correlations of the single factor (short form) with external variables
1 2 3 4 5 6 7 8 9
1 War stress - 0.19 0.30 0.23 0.21 0.18 − 0.16 0.34 0.50
2 Depression 0.34 - 0.71 0.59 0.62 0.42 − 0.38 0.40 0.38
3 Anxiety 0.38 0.71 - 0.69 0.60 0.42 − 0.36 0.51 0.48
4 Stress 0.33 0.56 0.66 - 0.48 0.34 − 0.21 0.40 0.44
5 Insomnia 0.27 0.59 0.52 0.48 - 0.37 − 0.27 0.44 0.44
6 Loneliness 0.24 0.48 0.43 0.37 0.37 - − 0.26 0.35 0.43
7 Resilience − 0.24 − 0.42 − 0.44 − 0.26 − 0.25 − 0.28 - − 0.30 − 0.23
8 COVID-anxiety 0.39 0.44 0.43 0.36 0.42 0.31 − 0.31 - 0.66
9 COVID-stress 0.46 0.45 0.46 0.39 0.47 0.39 − 0.32 0.68 -
Note: Correlations from the exploratory dataset are below the diagonal, while correlations from the confirmatory dataset are above it

Fig. 1 Visualization of the network of the full scale and items’ strength. A = Visualization of the network; B = Items’ strength (raw values); C = Differences in
items’ strength; D = Stability of the strength indicator
Vargová et al. BMC Psychology (2024) 12:208 Page 8 of 11

Fig. 2 Visualization of the network of the short-form scale and items’ strength. A = Visualization of the network; B = Items’ strength (raw values); C = Differ-
ences in items’ strength; D = Stability of the strength indicator

Discussion The second goal was to develop a short version of the

After COVID-19, the war in Ukraine has been another inventory that can be used to quickly assess perceptions
crisis significantly affecting people’s mental health. This of war-related stressors. This version consists of nine
is especially the case if stressors accumulate (see [19, 20, items loaded on to one factor. These items cover fear
40, 41]). For a better understanding of the relationship for one’s safety and future, access to basic needs, poten-
between war and mental health, it is important to have tial worsening of the economic situation, and escala-
adequate measures that can capture the different aspects tion of the conflict, and the possibility of nuclear threat.
of a military conflict as accurately as possible. The results of the network analysis indicate that con-
As such, an inventory of the most-pressing war-related cern about escalation, (i.e., the spread of the conflict to
stressors (at the time and considering the context) has Europe/other countries) seems to be the central indica-
been developed and validated. The full version of the tor for this construct. As with COVID-related stressors
inventory consists of 21 items that can be further divided [26], fear of an economic crisis was also found to play an
into three factors: (1) society-related stressors (8 items); important role.
(2) person-related stressors (7 items) and (3) security- Overall, the inventory items in both the full and abbre-
related stressors (6 items). The first factor, society-related viated versions align with previous findings and high-
stressors, consists of items related to the functioning of light the importance of addressing key stressors during
the country, such as management of the economic and wartime [6, 8, 9]. As in previous research [e.g., 42], the
migration crisis and the sufficiency of energy sources. current study has demonstrated positive correlations
The second factor, person-related stressors, covers items between war-related stress and mental health issues
directly related to a person’s life (e.g., worries about the (anxiety, depression), alongside negative associations
future, personal functioning, social relationships and with resilience. Despite high levels of perceived stress,
quality of life). The third factor, security-related stressors, Kurapov et al. [43] found surprisingly low scores of anxi-
includes items related to safety issues, such as personal ety and depression among Ukrainians after 6 months of
safety, potential escalation of the conflict, and nuclear experiencing war. Notably, individuals who remained
threats. living in Ukraine achieved lower levels of anxiety and
depression compared to those who had moved abroad,
Vargová et al. BMC Psychology (2024) 12:208 Page 9 of 11

emphasizing the highly traumatic and stressful potential as well as responses to these crises, can influence the
of both leaving one’s homeland and being displaced [44]. nature and intensity of stressors across different coun-
Similarly, studies working with Ukrainian refugees have tries. In Slovakia, the challenges posed by the pandemic
also reported worsened mental health outcomes [e.g., followed closely by war, could have heightened people’s
45]. A scale which is specifically focused on measuring fears for themselves and their loved ones, as well as con-
war-related stress can assist in understanding the impact cerns about worsening economic conditions and quality
of military conflict on people’s mental well-being bet- of life. Therefore, further studies are needed to explore
ter. Similarly, it is important to identify specific stressors the generalizability of the findings beyond the context of
that are particularly important during times of war and Slovakia.
can subsequently help in targeting specific and effective
interventions. Conclusion
Concerns about war, especially its potential escalation
Limits and perspectives and the threats of nuclear conflict or economic down-
As far as we know, this study is one of the first in creating turn, can have long-term effects on overall health (e.g.,
a valid scale of war-related stress. However, there remain [49]). This scale aims to cover a variety of factors affected
several caveats in it that should be addressed by further by war and the possible repercussions on those residing
research. (1) Even though the items have been developed outside conflict zones. Due to the complexity of the issue,
and selected with regard to several aspects (see the Scale researchers are invited to further refine and adapt the
construction section), it is not an ultimate list of items scale to suit different cultural, geographical, and temporal
which can operationalize the war-related stress con- settings. The customization of the scale, especially with
struct. As such, other researchers are invited to not only consideration for the proximity of populations to conflict
replicate the factor structure using these items but also areas, can offer essential insights and broaden our com-
to refine it, add new items, and examine potential fac- prehension of the varied impacts of military conflicts. By
tor structures using an updated version of the scale. (2) enhancing our understanding of the effects of war-related
Another potential limitation is that the scale focuses on stressors on population health, it is possible to not only
individuals residing outside the conflict area. For those contribute to reducing international and national ten-
living directly within the conflict zone, the stressors are sions but also to supporting the pursuit of just and lasting
likely to be of a different nature and more specific. In par- resolutions to conflicts.
ticular, civilians and refugees in war zones face immedi-
Acknowledgements
ate threats and losses due to direct exposure to traumatic Not applicable.
events, whereas people living outside conflict areas are
more concerned with potential threats [18]. Conse- Author contributions
LV: Conceptualization, Investigation, Writing– original draft, Writing– review
quently, the origins of symptomatology, their manifesta- & editing; BJ: Conceptualization, Investigation, Writing– original draft;
tions, and the optimal interventions may vary between Writing– review & editing; ML: Investigation, Writing– review & editing; MA:
the two groups. One example of this could be direct Conceptualization, Investigation, Supervision, Formal analysis, Methodology,
Writing– review & editing. The authors contributed equally to this work. The
exposure to traumatic events leading to PTSD. In con- order of the authorship was determined based on the results of the following
trast, stressors can be more context-dependent and sub- code: set.seed(06302022); sample(c(“Lenka”, “Biba”, “Matus”), 3, replace = FALSE).
ject to change over time for individuals living outside the ML joined the author team in the later stages and provided a substantial
contribution to the paper.
conflict zone. They may diminish with the disappearance
of potential stressors or with a decrease in their salience Funding
on social media [see 5, 46]. (3) The amount and intensity This study was supported by Slovak Research and Development Agency
[APVV-17-0418; APVV-20-0319; APVV-21-0057, and APVV-22-0458], the
of stressors are likely to be more worrisome for people in Research Grant Agency of the Ministry of Education, Science, Research and
nearby countries compared to those living further away Sport of the Slovak Republic and Slovak Academy of Science [VEGA under the
[see 47]. The generalizability of the results should also be contract no. 1/0559/21], and project PRIMUS/24/SSH/017.

considered with caution due to geographical proximity. Data availability

Proximity can amplify anticipatory worries about esca- The data and analytic code are freely available at https://osf.io/jv2np/.
lation, and fear can be heightened by the influx of refu-
gees as these countries closely witness the impact of war Declarations
on human lives [16]. (4) Cultural and community differ-
Ethics approval and consent to participate
ences should also be considered in future (replication) Ethical permission was granted by the Ethics Committee at the Faculty of
studies. In recent years, the global population has faced Arts, University of Presov. All the participants provided written consent to
a variety of crises such as wars, the COVID-19 pandemic participate in the study. The study was conducted in accordance with the
Declaration of Helsinki.
and poverty, among others [48]. Factors such as pre-
paredness, political, economic and societal conditions,
Vargová et al. BMC Psychology (2024) 12:208 Page 10 of 11

Consent for publication 19. Jawaid A, Gomolka M, Timmer A. Neuroscience of trauma and the Russian
Not applicable. invasion of Ukraine. Nat Hum Behav. 2022. https://doi.org/10.1038/s41562-
022-01344-4. 1–2.
Competing interests 20. Yen CF, Lin YH, Hsiao RC, Chen YY, Chen YL. Associations of China’s military
The authors declare no competing interests. activities in the peripheries of Taiwan with suicide death and internet
searches for depression, suicide, and emigration among individuals in
Received: 3 June 2023 / Accepted: 26 March 2024 Taiwan. Asian J Psychiatry. 2024;92(103889):103889. https://doi.org/10.1016/j.
ajp.2023.103889.
21. Bai W, Cai H, Sha S, Ng CH, Javed A, Mari J, et al. A joint international col-
laboration to address the inevitable mental health crisis in Ukraine. Nat Med.
2022;1–2. https://doi.org/10.1038/s41591-022-01828-w.
References 22. Kardas P, Babicki M, Krawczyk J, Mastalerz-Migas A. War in Ukraine and the
1. American Psychiatric Association. Diagnostic and statistical manual of mental challenges it brings to the Polish healthcare system. Lancet Reg Health Eur.
disorders. 5th ed. Washington, DC: American Psychiatric Association; 2013. 2022;15(100365):100365. https://doi.org/10.1016/j.lanepe.2022.100365.
2. Charlson F, van Ommeren M, Flaxman A, Cornett J, Whiteford H, Saxena S. 23. Lapid Pickman L, Gelkopf M, Greene T. Do positive and negative emotional
New WHO prevalence estimates of mental disorders in conflict settings: a reactions during war predict subsequent symptomatology? A prospec-
systematic review and meta-analysis. Lancet. 2019;394(10194):240–8. https:// tive experience sampling study. J Anxiety Disord. 2021;84(102492):102492.
doi.org/10.1016/S0140-6736(19)30934-1. https://doi.org/10.1016/j.janxdis.2021.102492.
3. Doyle ST. Mental weakness and the failures of military psychiatry. 2022. 24. DeVellis RF, Thorpe CT. Scale development: theory and applications. 5th ed.
https://doi.org/10.31234/osf.io/2kgjc. SAGE; 2024.
4. Dadi AF. The mental health consequences of war in northern Ethiopia: why 25. Lieberoth A, Lin S-Y, Stöckli S, Han H, Kowal M, Gelpi R, et al. Stress and
we should be concerned. Lancet Psychiatry. 2022;9(3):194–5. https://doi. worry in the 2020 coronavirus pandemic: relationships to trust and compli-
org/10.1016/S2215-0366(22)00013-X. ance with preventive measures across 48 countries in the COVIDiSTRESS
5. Scharbert J, Humberg S, Kroencke L, Reiter T, Sakel S, ter Horst J, Utesch K, global survey. R Soc Open Sci. 2021;8(2):200589. https://doi.org/10.1098/
Gosling SD, Harari G, Back MD. Psychological well-being in Europe after the rsos.200589.
outbreak of war in Ukraine. Nat Commun. 2024;1202. https://doi.org/10.1038/ 26. Taylor S, Landry CA, Paluszek MM, Fergus TA, McKay D, Asmundson GJG.
s41467-024-44693-6. Development and initial validation of the COVID stress scales. J Anxiety
6. Kalcza-Janosi K, Kotta I, Marschalko EE, Szabó K. The fear of War Scale Disord. 2020;72:102232. https://doi.org/10.1016/j.janxdis.2020.102232.
(FOWARS): development and initial validation. PsyArXiv. 2022. https://doi. 27. Rush AJ, Trivedi MH, Ibrahim HM, Carmody TJ, Arnow B, Klein DN, et al. The
org/10.31234/osf.io/sfz2v. 16-Item Quick Inventory of Depressive Symptomatology (QIDS), clinician rat-
7. Boehnke K, Schwartz SH. Fear of war: relations to values, gender, and mental ing (QIDS-C), and self-report (QIDS-SR): a psychometric evaluation in patients
health in Germany and Israel. Peace Confl. 1997;3(2):149–65. https://doi. with chronic major depression. Biol Psychiatry. 2003;54(5):573–83. https://doi.
org/10.1207/s15327949pac0302_3. org/10.1016/s0006-3223(02)01866-8.
8. Mbah RE, Wasum D. Russian-Ukraine 2022 war: a review of the economic 28. Spitzer RL, Kroenke K, Williams JBW, Löwe B. A brief measure for
impact of Russian-Ukraine crisis on the USA, UK, Canada, and Europe. Adv assessing generalized anxiety disorder: the GAD-7. Arch Intern Med.
Soc Sci Res J. 2022;9(3):144–53. https://doi.org/10.14738/assrj.93.12005. 2006;166(10):1092–7. https://doi.org/10.1001/archinte.166.10.1092.
9. Choudhary OP, Saied AA, Priyanka, Ali RK, Maulud SQ. Russo-ukrainian war: 29. Cohen S, Kamarck T, Mermelstein R. A global measure of perceived stress. J
an unexpected event during the COVID-19 pandemic. Travel Med Infect Dis. Health Soc Behav. 1983;24(4):385–96. https://doi.org/10.2307/2136404.
2022;48(102346):102346. https://doi.org/10.1016/j.tmaid.2022.102346. 30. Morin CM. Insomnia: psychological assessment and management. New York:
10. Pearlman W. Narratives of fear in Syria. Perspect Polit. 2016;14(1):21–37. Guilford Press; 1993.
https://doi.org/10.1017/s1537592715003205. 31. Hays RD, DiMatteo MR. A short-form measure of loneliness. J Pers Assess.
11. Martinez-Garcia MA, Ramos-Carvajal C, Camara A, Part B. 2023, 104114, 1987;51(1):69–81. https://doi.org/10.1207/s15327752jpa5101_6.
https://doi.org/10.1016/j.resourpol.2023.104114. 32. Smith BW, Dalen J, Wiggins K, Tooley E, Christopher P, Bernard J. The brief
12. McKinsey & Company. Economic conditions outlook during turbulent times. resilience scale: assessing the ability to bounce back. Int J Behav Med.
December 2022. https://www.mckinsey.com/~/media/mckinsey/business_ 2008;15(3):194–200. https://doi.org/10.1080/10705500802222972.
functions/strategy_and_corporate_finance/our_insights/economic_condi- 33. Silva WAD, de Sampaio Brito TR, Pereira CR. COVID-19 anxiety scale (CAS):
tions_outlook_2022/december_2022/economic-conditions-outlook-during- development and psychometric properties. Curr Psychol (New Brunswick,
turbulent-times-december-2022.pdf. Published 2022. N.J.). 2020;1–10. https://doi.org/10.1007/s12144-020-01195-0.
13. Rozanov V, Frančišković T, Marinić I, Macarenco MM, Letica-Crepulja M, 34. Yamada Y, Ćepulić D-B, Coll-Martín T, Debove S, Gautreau G, Han H, et al.
Mužinić L, Jayatunge R, Sisask M, Vevera J, Wiederhold B, Wiederhold M, Miller COVIDiSTRESS Global Survey dataset on psychological and behavioural
I, Pagkalos G. Mental health consequences of war conflicts. Adv Psychiatry. consequences of the COVID-19 outbreak. Sci Data. 2021;8(1):3. https://doi.
2018;281–304. https://doi.org/10.1007/978-3-319-70554-5_17. org/10.1038/s41597-020-00784-9.
14. Hajek A, Kretzler B, König HH. Fear of war and mental health in Germany. Soc 35. Hayton JC, Allen DG, Scarpello V. Factor retention decisions in exploratory fac-
Psychiatry Psychiatr Epidemiol. 2023;58(7):1049–54. https://doi.org/10.1007/ tor analysis: a tutorial on parallel analysis. Organ Res Methods. 2004;7(2):191–
s00127-022-02394-9. 205. https://doi.org/10.1177/1094428104263675.
15. Regnoli GM, Tiano G, De Rosa B. Italian adaptation and validation of the 36. Borsboom D, Cramer AOJ. Network analysis: an integrative approach to the
fear of War Scale and the impact of the fear of war on young Italian adults’ structure of psychopathology. Annu Rev Clin Psychol. 2013;9(1):91–121.
mental health. Social Sci (Basel Switzerland). 2023;12(12):643. https://doi. https://doi.org/10.1146/annurev-clinpsy-050212-185608.
org/10.3390/socsci12120643. 37. Revelle W. Psych: procedures for personality and Psychological Research,
16. Mărcău FC, Peptan C, Gorun HT, Băleanu VD, Gheorman V. Analysis of the Northwestern University, Evanston, Illinois, USA. Published; 2022. https://
impact of the armed conflict in Ukraine on the population of Romania. Front CRAN.R-project.org/package=psych.
Public Health. 2022;10. https://doi.org/10.3389/fpubh.2022.964576. 38. Rosseel Y. Lavaan: an R Package for Structural equation modeling. J Stat
17. Massag J, Diexer S, Klee B, Costa D, Gottschick C, Broda A, Purschke O, Opel Softw. 2012;48(2). https://doi.org/10.18637/jss.v048.i02.
N, Binder M, Sedding D, Frese T, Girndt M, Hoell J, Moor I, Rosendahl J, Gekle 39. Epskamp S, Borsboom D, Fried EI. Estimating psychological networks and
M, Mikolajczyk R. Anxiety, depressive symptoms, and distress over the course their accuracy: a tutorial paper. Behav Res Methods. 2018;50(1):195–212.
of the war in Ukraine in three federal states in Germany. Front Psychiatry. https://doi.org/10.3758/s13428-017-0862-1.
2023;14. https://doi.org/10.3389/fpsyt.2023.1167615. 40. Brągiel A, Gambin M. Depressive symptoms and psychological pain expe-
18. Chudzicka-Czupała A, Hapon N, Chiang S-K, Żywiołek-Szeja M, Karamushka rienced by Polish adults in the context of both the war in Ukraine and the
L, Lee CT, Grabowski D, Paliga M, Rosenblat JD, Ho R, McIntyre RS, Chen YL. COVID-19 pandemic. PsyArXiv. 2022. https://doi.org/10.31234/osf.io/h2yex.
Depression, anxiety and post-traumatic stress during the 2022 russo-ukrai- 41. Moshagen M, Hilbig BE. Citizens’ Psychological Reactions following the Rus-
nian war, a comparison between populations in Poland, Ukraine, and Taiwan. sian invasion of the Ukraine: A cross-national study. PsyArXiv. 2022. https://
Sci Rep. 2023;13(1):1–14. https://doi.org/10.1038/s41598-023-28729-3. doi.org/10.31234/osf.io/teh8y.
Vargová et al. BMC Psychology (2024) 12:208 Page 11 of 11

42. Surzykiewicz J, Skalski SB, Niesiobędzka M, Gladysh O, Konaszewski K. Brief 47. Jain N, Prasad S, Bordeniuc A, Tanasov A, Shirinskaya AV, Béla B, et al.
screening measures of mental health for war-related dysfunctional anxiety European countries step-up humanitarian and medical assistance
and negative persistent thinking. Pers Indiv Differ. 2022;195(111697):111697. to Ukraine as the conflict continues. J Prim Care Community Health.
https://doi.org/10.1016/j.paid.2022.111697. 2022;13:21501319221095360. https://doi.org/10.1177/21501319221095358.
43. Kurapov A, Kalaitzaki A, Keller V, Danyliuk I, Kowatsch T. The mental health 48. WHO. Conflict and crisis reveal the tip of the iceberg the world’s vulnerable
impact of the ongoing russian-ukrainian war 6 months after the Russian face in accessing their right to health. https://www.who.int/news-room/com-
invasion of Ukraine. Front Psychiatry. 2023;14. https://doi.org/10.3389/ mentaries/detail/conflict-and-crisis-reveal-the-tip-of-the-iceberg-the-world-
fpsyt.2023.1134780. s-vulnerable-face-in-accessing-their-right-to-health. Published 2023.
44. Maier K, Konaszewski K, Skalski SB, Büssing A, Surzykiewicz J. Spiritual 49. Poikolainen K, Aalto-Setälä T, Tuulio-Henriksson A, Marttunen M, Lönnqvist J.
needs, religious coping and mental wellbeing: a cross-sectional study Fear of Nuclear War Increases the Risk of Common Mental Disorders among
among migrants and refugees in Germany. Int J Environ Res Public Health. Young Adults: A Five-Year Follow-Up Study. BMC Pub Health. https://doi.
2022;19(6):3415. https://doi.org/10.3390/ijerph19063415. org/10.1186/1471-2458-4-42. https://doi.org/10.1186/1471-2458-4-42
45. Konstantinov V, Reznik A, Isralowitz R. Depression and quality of life among
Ukrainian adults relocated to Russia. J Loss Trauma. 2023;28(6):493–503.
https://doi.org/10.1080/15325024.2023.2216986. Publisher’s Note
46. Asmundson GH, Rachor G, Drakes SH, Boehme BAE, Poluszek MM, Taylor S. Springer Nature remains neutral with regard to jurisdictional claims in
How does COVID stress vary across the anxiety-related disorders? Assess- published maps and institutional affiliations.
ing factorial invariance and changes in COVID stress scale scores during
the pandemic. J Anxiety Disord. 2022;102554. https://doi.org/10.1016/j.
janxdis.2022.102554.