International students are individuals who leave their home country to pursue higher education abroad, seeking enhanced academic opportunities and expanding their career prospects (Güneş and Aydar, 2019). Recent data indicates a significant rise in internationally-mobile students, with over 6.3 million recorded globally in 2020 compared to 2 million in 2000 (UNESCO Institute for Statistics, United Nations Children's Fund, and The World Bank, 2022), marking a 70% increase over the past decade (OECD, 2022). While studying abroad offers enriching learning experiences (Bilecen et al., 2023), it also poses challenges in adapting to new educational systems and socio-cultural environments. This process, known as acculturation (Wahyuningtyas et al., 2021), can induce stress, with adverse implications for psychological and social well-being (Alshammari et al., 2023; Bender et al., 2019; Koo et al., 2021; Pang, 2020; Serrano-Sánchez et al., 2021). Acculturative stress may lead to the development of depression and anxiety symptoms (Alshammari et al., 2023; Brunsting et al., 2018; Jung et al., 2007; Koo et al., 2021; Newton et al., 2021; Pang, 2020), and this evidence is supported by numerous studies indicating that, compared to other student groups, international students exhibit a higher incidence of mental health issues (Alharbi and Smith, 2018; Koo et al., 2021; Newton et al., 2021). Additionally, the literature underscores that loneliness is one of the primary challenges arising from poor adjustment to new environments, largely due to cultural disparities and language barriers, and that international students are proved to be particularly susceptible to social isolation and loneliness (Alharbi and Smith, 2018; Bilecen et al., 2023; Serrano-Sánchez et al., 2021; Wawera and Mccamley, 2020), often exacerbated by discriminatory treatment (Alshammari et al., 2023; Sun et al., 2020).

The distressing experience of loneliness encompasses a strong sense of social pain, dysphoria, emptiness, and worthlessness (Cacioppo et al., 2014; Matthews et al., 2016), stemming from a perceived deficiency in relationships (with individuals, groups, or communities) mainly related to the qualitative aspects of their social networks (Matthews et al., 2019; Perlman and Peplau, 1981; Wigfield et al., 2022). Chronic loneliness has been associated with the development and exacerbation of depressive symptoms (Cacioppo et al., 2014; Pitman et al., 2018), anxiety, alcohol abuse, cognitive decline, dementia in older age, and an increased mortality risk (Christiansen et al., 2021; Hawkley and Cacioppo, 2010; Matthews et al., 2016; Rico-Uribe et al., 2018). Long-term studies also link loneliness to a higher risk of physical illnesses such as cardiovascular disease, hypertension, asthma, migraine, tinnitus, osteoarthritis, and slipped disk (Christiansen et al., 2021; Hawkley and Cacioppo, 2010; Matthews et al., 2019). Considering the extensive evidence, loneliness emerges as a significant public health concern, especially when considering the long-term consequences of the recent COVID-19 pandemic (Lampraki et al., 2022). Pandemic-related social restrictions contributed to increased levels of loneliness, depression, anxiety, insomnia, substance use, and suicidal attempts among young adults in general (Glowacz and Schmits, 2020; Hawes et al., 2022; Sewall et al., 2022) and college students in particular (Buizza et al., 2022; Carvalho et al., 2022; Elharake et al., 2023; Haikalis et al., 2022). Lockdowns and border closures particularly heightened the vulnerability of international students, intensifying feelings of loneliness, social isolation, and exacerbating their negative impact on well-being (Alaklabi et al., 2021; Bilecen, 2020; Zhao et al., 2022). It’s evident that urgent attention must be paid and support must be provided in order to address this distressing trend among international students. Psychological and social interventions are imperative to ensure the well-being and mental health of these students, particularly in light of the lasting effects of the recent pandemic.

Loneliness, while not an inherent trait, is shaped by life experiences, and studies showed the effectiveness of psychological and psychosocial interventions in mitigating loneliness (Bessaha et al., 2020). However, understanding the factors determining their benefits is challenging due to diverse intervention designs, theoretical foundations, delivery modalities, focal points, and the strategies employed (Eccles and Qualter, 2021; Hickin et al., 2021; Masi et al., 2011; Osborn et al., 2021; Zagic et al., 2022). Despite controversies, the current literature suggests that when compared to individual sessions, group interventions can be more successful in improving the quality of social connections by providing the opportunity to increase the number of social contacts through direct in-person experiences (Eccles and Qualter, 2021; Masi et al., 2011; Zagic et al., 2022). They provide opportunities to access sources of reciprocal social support and motivate lonely people to promote their own well-being while mediating stress responses (Bessaha et al., 2020; Zagic et al., 2022). Social support, defined as the assistance, protection and supportive behavior provided to others (i.e., individuals, groups, and the broader community) through social ties (Brunsting et al., 2021; Langford et al., 1997; Ozbay et al., 2007; Wilson et al., 2020), stands as significant protective factors in overcoming acculturative stress, playing a crucial role in the acculturation process (Bender et al., 2019; Billedo et al., 2019; Cipolletta et al., 2022; Hofhuis et al., 2019; Nakao, 2019; Pang, 2020). According to the literature, a combination of psychoeducation, counseling, and social support intervention in group delivery modality appears to be the most suitable approach for supporting lonely international students (Carr et al., 2003; Dipeolu et al., 2007). Group interventions targeting this specific group can provide a sense of belonging and shared social identification (Bessaha et al., 2020; Gold et al., 2019; Haslam et al., 2019; Masi et al., 2011; Zagic et al., 2022) by normalizing common experiences and alleviating experiences of isolation (Carr et al., 2003; Yakunina et al., 2011), although still dealing with different disadvantages such as problems with the group composition and engagement, cultural and linguistic differences leading to conflicts and misunderstandings and low participation potentially connected to high levels of social anxiety or loneliness. There are also some examples of online interventions for university (Charbonnier et al., 2022) and international college students (Balci et al., 2023) showing significant improvements in overall well-being and anxiety, while at the same time pointing out potential limitations such as high drop-out rate (50%), low adherence (only 40% completed the intervention), unexplored long-term effects, and the need for further explorations. As digital and group intervention -if separately implemented- show different disadvantages, blended interventions (combining online and face-to-face modalities) could represent an innovative and resourceful opportunity to respond to international students’ needs and challenges, facilitating help-seeking behaviors and adherence, and reducing barriers to mental healthcare (De Witte et al., 2021; Evans-Lacko et al., 2022), thus combining the advantages of online and face-to-face modalities while minimizing their disadvantages (Erbe et al., 2017).

This pilot study is the first to have tested a blended intervention, comprising online self-help activities and face-to-face group sessions to improve well-being and reduce the increasing experiences of loneliness and social isolation among international students. Given the inherent complexity in both the experience of loneliness and the situation international students face, a blended intervention is proposed, consisting of two different components: participation in face-to-face group activities conducted by a professional, and the use of online self-help materials. The theoretical framework with regard to group activities and self-help materials which has been developed is that of Personal Construct Psychology (PCP) (Kelly, 1955). This framework advances a perspective that centers on the subjective construction of wellbeing rather than a pre-defined and normative notion of wellbeing. The proposed activities aim to explore personal beliefs and identify the personal constructs that can promote wellbeing and social connections, while experiencing new ways of interacting with others within a group. To comprehensively evaluate the effects of these activities, the study introduced and compared two other active conditions: the self-help intervention involving access solely to self-help materials, and peer-to-peer intervention, involving participation in group activities facilitated by peers, plus an inactive control group.

The study aims to evaluate the impact of the blended intervention on loneliness, social support and well-being among international students at an Italian university. Additionally, it seeks to investigate which intervention modality (blended, face-to-face or online) and characteristics might be most effective within this population. Lastly, given its pilot nature, the study assesses the acceptability and relevance of the proposed interventions. In line with the literature presented above, the authors hypothesize that the blended intervention will be evaluated as more useful and accessible than the other kinds of interventions, and that the blended and peer-to-peer interventions will be more effective than the online and no intervention approach in promoting participants’ social connectedness and wellbeing.

The present study followed the methodology of a randomized controlled trial (RCT). However, due to the small sample size and difficulties recruiting participants, participant randomization was not possible, leading to an acknowledgement of the potential weaknesses inherently present in non-randomized study for psychological research. The study is therefore a non-randomized pilot study reporting a group comparison.

The present research project was registered on ClinicalTrial (Protocol N. NCT05867758) and was approved by the Ethical Committee for the Psychological Research of the University of Padova (Protocol N. 5112).

The present study aimed to test the methods and procedures of an intervention to reduce social isolation and loneliness in international students, with the primary objective of evaluating its feasibility and relevance for participants, exploring potential effects and experiences that warrant further investigation in a subsequent, more comprehensive study (Thabane et al., 2010). The intervention conditions were specifically developed to meet international students’ needs and to allow them to face the challenges of living in a foreign country. Activities and materials were developed based on the conceptual framework of the PCP (Kelly, 1955) and by considering social support as a fundamental construct for well-being, especially in terms of meaningful relationships (Lee and Goldstein, 2016; Umberson and Karas Montez, 2010; Ryff, 1989).

The pilot study proposed three different types of intervention (blended, online self-help or peer-to-peer) encompassing group activities and self-help materials. The qualitative results show that the interventions including in-presence group activities provided a social context in which the participants could experience social connectedness, and enter in relationship with peers, while the online self-help intervention left the reflection process to the individuals and their environment, limiting the potential for experiencing social connections.

The comparison of the blended, self-help, and peer-to-peer interventions with the control condition allowed for an assessment of the feasibility and effectiveness of each intervention modality in addressing social isolation and loneliness among international students, and in fostering their well-being. The participants’ satisfaction with the interventions, and the individual relevance of the content and activities, are essential for assessing their potential impact and feasibility for larger-scale implementation (Thabane et al., 2010).

The findings reveal that the interventions that included in-person group activities (i.e., the blended and peer-to-peer interventions) were perceived as more relevant and more functionally organized compared to the intervention that only involved online self-help materials. Furthermore, the participants reported that the topics proposed and discussed in the blended and peer-to-peer interventions, which included group activities, were relevant to their lives, and encouraged stimulating reflections on their experiences, leading to the development of new perspectives.

On the other hand, the participants in the self-help group condition reported to have found the online materials useful and relevant when it came to addressing many challenges and fostering their well-being (Schotanus-Dijkstra et al., 2015; Walsh et al., 2018), but complained about the absence of face-to-face interaction and suggested integrating the intervention with in-person sessions, while acknowledging the significance of offline relationships in strengthening well-being and social support (Nowland et al., 2018).

Notably, in terms of satisfaction with the interventions, the results show that the blended group evaluated the self-help materials as being relevant, satisfactory, and organized in a more effective way compared to the self-help group. These findings emphasize the necessity of transforming self-help materials’ insights and feedback into active and direct experiences, thereby increasing the materials’ relevance and increasing the beneficial impact on well-being. This observation is consistent with Esposito and colleagues’ findings (Esposito et al., 2017), who claimed that reflexive processes proposed by self-help materials have only limited effects when exclusively used for individual thinking, but they can have greater potential when involving a process of shared meaning co-construction among participants in a collective context.

Comparing the blended and peer-to-peer groups, no significant differences are noted in terms of satisfaction, relevance, improvement in connection and well-being, a result that is possibly due to the same structure of activities and contents being involved. Concerning the peer-to-peer group, open-ended responses focused on the possibility of new social connections, which could enhance feelings of self-acceptance and social support, positively impacting on depression, anxiety, and feelings of helplessness (Bessaha et al., 2020; Kotwal et al., 2021). Blended group participants’ responses however, stressed the relevance of the non-judgmental atmosphere of the group, an element that supported them in discussing significant topics and sharing personal experiences.

The blended and peer-to-peer intervention facilitated the free exchange of experiences and ideas in a secure and cooperative environment, leading to the establishment of new social connections rooted in the common experience of being international students. Uniting individuals with shared experiences, indeed, reduces feelings of loneliness and validates individual constructs (Bessaha et al., 2020; Carr et al., 2003; Yakunina et al., 2011). Moreover, working in groups offered participants a chance to reshape their meaning systems and construe novel perspectives, helping them to enrich their constructs or elaborate new anticipations and resources. The present findings support Garcia-Martínez et al.’s (2021) hypothesis that report how the process of construing shared meanings can enhance communication and foster transformative interactions.

The statistical analyses run to verify the impact of each intervention on the standardized questionnaires did not reveal significant results. Nevertheless, the effect size was large enough to suggest an increase of perceived social support and wellbeing and a decrease of anxiety, depression in the blended group and an increase of perceived social support and a decrease of anxiety, depression and loneliness in the peer-to-peer group. On the contrary, the control group reported an increase of anxiety, depression and loneliness and did not report any effects of change on social support and wellbeing. The group that only used self-help material reported an increase of perceived social support and well-being, but also an increase of anxiety and depression.

Only perceived social support increased significantly over time across all groups, including the control group, but there was no evidence to attribute such effect to the type of intervention. A hypothesis of interpretation could be found in the evolutionary function of loneliness (Cacioppo et al., 2014), suggesting that loneliness can motivate reconnection with others, leading to improved perceived social support over time without the need for specific interventions (Masi et al., 2011). Other hypotheses might be linked to the specific context of the study: first, being involved in the study might in itself have made participants feel more supported than before the participation in the study, and second, the period when the study was conducted (2022–2023) was a moment of transition following the restoration of social and in-person academic activities post-COVID19 pandemic when students perceived high social support (Bersia et al., 2024).

Another potential explanation of the results obtained in the measure of perceived social support, but also in the loneliness questionnaire, regards the nature of interventions to address loneliness, which are typically focused and designed to increase social connections and reduce social isolation, rather than specifically focusing on reducing experiences of loneliness (Wigfield et al., 2022). The results of the present study suggest that the proposed interventions contributed to increase the perception of social support whereas only the peer-to-peer intervention contributed to decrease the feelings of loneliness but not to increase perceived social support. This is an interesting data that deserves to be further explored and that suggests that peer-to-peer intervention might be the modality of choice to address loneliness.

The lack of statistical significance in the other questionnaire scores raises questions about the reasons behind this outcome if compared to the subjectively reported change in participants’ wellbeing and social connectedness, and to the differences observed in the descriptive analyses, also confirmed by the effect size. Hypothetical reasons that have been acknowledged and discussed within the research team include the small sample size and the high non-completers’ rate, which may have contributed to the lack of statistical significance. Furthermore, the PCP theoretical framework (Kelly, 1955) based on which the interventions were designed, focuses on the exploration and elaboration of participants’ personal experiences and meanings, and, by its nature, the framework avoids standardization. This may have led to difficulties in using standardized questionnaires to measure such experiences, as they may oversimplify complex experiences. In this sense, the PCP framework may have constrained the capacity of the administered tests to measure intervention-induced changes. This limitation could arise from the standardized nature of the tests: the questionnaires used could not assess changes in the personal processing and elaboration of the constructs they analyze, which constituted a pivotal focus of the intervention.

Moreover, the language of administration of the tests and their cultural sensitivity may have played a crucial role. All of the standardized tools employed in fact were validated involving Anglo-American populations, while the pilot study participants who completed these measures, were mainly non-native English speakers and non-Anglo-American. Such cultural specificity, and potential difficulties in understanding and correctly interpreting the questions provided in a vehicular language, could have resulted in the tests’ lack of sensitivity to the respondents’ cultural diversity and language ability, compromising the results of the tests. Thus, while the collaborative interactions among students could have helped them overcome communication barriers during the group activities, potential language barriers may have influenced the results and compromised the assessments. These results raise different questions regarding the consistency and feasibility of the standardized instruments adopted, and the possibility of considering alternative measures for future evaluations.

Moreover, due to the close timing of the questionnaires’ administration (within approximately 2 months), the recorded scores could have been influenced by the test–retest effect and memory bias, potentially underestimating the actual change. Another important element to consider is that the majority of the participants in the present study were enrolled in psychology degree programs, potentially influencing the results and scores attributed due to familiarity bias. In fact, familiarity with professional psychological interventions and psychological tests may have influenced their completion and the evaluation of the intervention impact itself.

As a matter of fact, participants both in the blended and peer-to-peer groups, reported a willingness for longer face-to-face group activities, to help them develop solid relationships and build satisfactory social support networks that could positively impact their psychosocial wellbeing. The study was designed to encompass 8 sessions based on the existing literature (Kotwal et al., 2021; Osborn et al., 2021), which frequently suggests a duration of 6–8 sessions as optimal for implementing these intervention modalities. Despite this, as reported by the participants, and according to the literature (Brunelli et al., 2016), a longer duration could enhance the development of meaningful relationships and impact on clinically-relevant dimensions of mental health. This suggestion should be taken into account for the future implementations of interventions.

Furthermore, data show that in the blended group all participants, with one exception, had previous experiences with psychological support, whereas no one in the peer-to-peer group, and only one in the self-help group, reported previous experiences with psychological support. Taking those elements into consideration, it is plausible to hypothesize that the outcomes observed in the blended group might underestimate the potential benefits of the blended intervention. Our study suggests that previous psychological support experience could have confounded the results, potentially diminishing the specific impact of the intervention: previous interactions with mental health professionals may have already enriched participants’ psychological and social resources, which are the same resources the interventions aimed to provide. Interestingly, the blended group had a greater number of participants with prior experience of psychological support, suggesting that individuals accustomed to a specific intervention structure may have been inclined toward seeking professional-led intervention rather that peer-to-peer activities. Additionally, when psychological support is ongoing, measures of mental health and well-being might be more critical, carrying the risk of subsequently influencing the results. Future investigation, taking into account those elements, is necessary to elaborate further our results and hypothesis.

Another relevant element to be discussed relates to the high non-completers’ rate between baseline and post-intervention assessment. Comparative analyses between the non-completers’ group and the group that completed the interventions were carried out to investigate elements that could explain potential causes. Those analyses revealed significantly higher levels of loneliness, anxiety and depression raising questions about whether participants with elevated levels of anxiety and loneliness might perceive the group intervention as overly threatening, or the self-help materials as irrelevant for their needs. These findings may indicate that the treatment completers may not adequately represent the general population, while suggesting that administering a preliminary clinical screening before the start of the intervention may be pivotal for future research. Such screening could help orient different participants to more suitable modalities of intervention based on personal characteristics and psychological resources, hopefully preventing the high non-completers’ rate. Moreover, more effective recruitment strategies might be implemented such as organizing a pre-intervention meeting or call to present the aim and modalities of the intervention and allow participants to better understand the study and choose if they wanted to take part in it. Another option might be to involve the students in the recruitment process through the organizations of social initiatives where the students might participate and feel committed, thus lowering the threat of participating in an intervention promoting social connection and giving them an active role in organizing and planning the intervention also according to their needs and constrains (e.g., choosing a period of the year that is more suitable for them).

Even though not as expected, the results collected in this pilot study support the possibility that tailored interventions improve social connectedness and wellbeing, underlining the importance of providing spaces to meet other people and share relevant experiences, while dealing with specific challenges and disruptions. The present pilot study in this sense provides initial data and considerations which should be taken into account when designing and implementing future versions of tailored interventions to support international students in adapting to the new situation, fostering their social connectedness with peers and increasing their wellbeing. The results provide interesting starting points that foster the possibility of reflecting on and discussing the importance of creating and implementing tailored interventions, involving multidisciplinary activities and resources, such as the digital and group ones, responding to the actual needs of these populations.

In conclusion, this study provides valuable insights into interventions targeting social isolation and loneliness. The results emphasize the significance of group-based approaches in enhancing personal well-being and perceptions of social support. However, achieving statistical significance in self-report standardized questionnaire analyses was inconsistent, underscoring the need for further investigation.

The study highlights the importance of face-to-face group interactions in fostering a sense of belonging and connectedness among participants, as well as ways to construct and co-construct one’s personal meaning. Despite limitations such as a high non-completer’ rate and nonhomogeneous group distribution, the feasibility and relevance reported by participants suggest a beneficial impact of the group in-person interventions on interaction improvement and resource development. Future research should address these limitations by considering larger participant samples and implementing longitudinal designs to capture the long-term effects of those interventions, allowing for a nuanced exploration of their impact on well-being and social support over time. Finally, randomized controlled conditions could enhance study rigor and enrich findings.