It is currently considered that racial discrimination – a cumulative experience that degrades mental health, erodes the sense of belonging and leads to loneliness – constitutes a key determinant of health (Leigh-Hunt et al., 2017). Among racialised migrants – in other words, migrants exposed to racial discrimination – this pressure is combined with migration-related grief, precarity, and cultural distance, creating a context of high psychosocial vulnerability (Gabarrell-Pascuet et al., 2023). The international literature documents a cycle of exclusion in which racism erodes the well-being and mental health of newcomers, and this deterioration, in turn, hinders their social inclusion (WHO, 2017; El-Shaarawi & Larchanché, 2022; Cristóbal Narváez et al., 2025).

Social mentoring programmes – widely developed as a strategy for the inclusion of young migrants and refugees (Prieto-Flores & Feu, 2021) — are based on the assumption that strengthening bonds of trust can act as a protective shield. This hypothesis is evaluated in the “SAFESPACES” project, measuring whether mentoring improves well-being and, furthermore, whether it also modifies the correlation between racism and well-being. For this purpose, it analysed eight dimensions – mental health, hope, self-esteem, loneliness, social support, belonging, quality of life, and perceived discrimination – along with the composition of the personal network of trust.

The main results of the study, which included a total of 157 young racialised migrants, mainly from Africa and some from Eastern Europe and Latin America, indicate that, although overall well-being evolved similarly in both the mentoring group and the non-intervention group, a difference emerges in how the two groups cope with racism. 

At first sight, the results may seem disappointing: with the exception of initial hope, no well-being scale shows significant differences between the groups at any wave, and both groups improve similarly. However, this interpretation overlooks the essential point: the impact of mentoring does not consist so much in increasing well-being in an abstract way as in changing how migrants respond to racism. In other words, it is not a case of improving their quality of life when they are not exposed to discrimination, but of protecting it when they are. 

Social mentoring acts as a protective factor in the correlation between discrimination and loneliness. Moderation analysis reveals one of the central mechanisms in the study (see figure 1). In the non-intervention group, each increase in perceived discrimination translates into an increase in loneliness. This correlation does not arise in the mentoring group, where discrimination no longer predicts loneliness and interaction is statistically significant. Although mentoring does not eliminate racism, it does neutralise one of its main consequences: social isolation. This pattern is consistent with previous research that shows that a sense of community and supportive bonds can moderate the impact that discrimination has on migrants’ well-being (García-Cid et al., 2020). The project’s advisory group provides a key for interpretation: when people suffer discrimination on a daily basis, it tends to become normalised, but when the experience is shared with someone alarmed by the injustice of it, this means that, in the words of one of the group’s members, “you get the feeling that the situation is not normalised and you don’t feel so alone”.



 

How does mentoring protect migrant and refugee well-being? The responses that predict an increase in hope three months after the end of the programme are those related to emotional support – feeling listened to, feeling safe with their mentor, or feeling that the mentor helps them cope with distress – whereas items related to practical support – good ideas, useful information, specific solutions – do not reach statistical significance. In the face of racism, it is not the knowledge, advice, or solutions provided by the mentor that help most, but their emotional presence, in line with various studies that distinguish emotional support from instrumental support in mentoring relationships (Brady, Dolan & Canavan, 2015). The qualitative phase of the “SAFESPACES” project, based on focus groups with mentors and mentees, confirmed this pattern: emotional accompaniment and validation are factors that sustain well-being in the face of discrimination (Prats-Brugat, Alarcón & Prieto-Flores, 2026). This finding has direct implications for programme design: when selecting mentors, high levels of emotional warmth matter more than professional or cultural capital.

Discrimination, furthermore, has another effect that the analysis of support networks shows clearly: it usually encourages people to relate more closely to their own community. While this response – seeking support where one feels recognised – is understandable, it can also entail negative consequences: this dynamic reduces access to the resources and opportunities that the host society offers and, over time, weakens inter-ethnic support networks (Strang & Quinn, 2021). This trend is significant in the control group: during the study, their ties with local people decreased (from 45% to 32%), whereas those who participated in mentoring programmes maintained a diverse network, even when the environment was hostile. 



 

One of the risks associated with psychosocial programmes is that their effects dissipate once the programme ends. The data, however, suggest that the opposite is true: three months after the programme ended, 81% of participants still include their mentor in their trusted network, and 90% report ongoing contact with that person. Moreover, mentoring generates unexpected social capital among peers: 58% maintain contact with people they met through the programme, compared to 30% of the control group. The advisory group confirms this interpretation from own experience: “We’ve been working together for four years now, we’re still in touch and I know his whole family”, said one member. Another summed it up like this: “Trust is the basis of everything; once that bond exists, there’s no need to schedule meetings anymore”.



 

The analysis identifies two red flags that should be taken into consideration for the management of social mentoring programmes.

Firstly, reactivation of post-traumatic distress during the intervention. Over the course of the programme, the mentoring group developed more symptoms of trauma than the control group, a difference that was confirmed upon conclusion of the programme. One plausible explanation is that the space of trust created by the mentor enables the emergence of an emotional distress latent up to that point, caused both by experiences of racism in the host society and by trauma experienced before arrival: the persecution or violence that prompted flight and the dangers of the migration journey itself. The fact that this distress has emerged does not mean it has worsened – the differences detected upon completion of the programme are no longer observed several months later –, but it does indicate that the programme can shine a light on especially sensitive experiences and emotions. The advisory group supports this interpretation. One participant expressed it in the following way: “It’s telling the story over and over, reliving situations that you don’t want to relive; on many occasions I felt revictimized”. Thus, it is advisable to address this issue both in mentor training and, where necessary, in psychological support.

Secondly, the risk of dependency. In general, people feel they belong to a group because they have their own support network: the more social ties they have, the greater their sense of belonging. This is precisely what is observed in the control group. In contrast, in the mentoring group, this connection disappears: participants feel that they are part of something even if their support network has not increased. The most plausible explanation is that this sense of belonging is provided by the programme itself. And therein lies the risk: if belonging is sustained in the intervention and not in each person’s own social ties, it can fade at the end of the programme while the discriminatory environment remains. For this reason, a good programme should contribute to building this sense of belonging into the migrant’s own support network, so that it can endure beyond the intervention. 

Racism degrades the well-being of racialised migrants: it closes doors to them, erodes their sense of belonging and leads them to concentrate their relationships in the community itself, thus weakening their inter-ethnic ties with the host society. The “SAFESPACES” project shows that social mentoring can help tackle this challenge: not by eliminating discrimination – which will continue to be present – but by mitigating one of its most harmful effects on well-being. It seems that the key lies in the relationship: a trusting, emotional and non-instrumental relationship can help prevent experiences of racism from leading to feelings of loneliness.

Some recommendations are derived from these results. In the training and selection of mentors, priority should be given to emotional warmth over professional or cultural capital, training should be provided in racism and migratory trauma, and there should be encouragement to integrate the bonds generated during mentoring into the migrant’s everyday relations. In programme management, it is advisable to design a specific support during the closing phase and actively foster connections among participants. With regard to funding bodies and public administration, the results highlight the importance of financing longitudinal follow-ups, of recognising mentoring as an intervention that contributes to both mental health and social cohesion – not just to labour market integration – and also of acting in environments where racism continues to manifest itself (school, health, and housing).

With respect to implications for the third sector, the results show that social mentoring can act as a psychosocial support in contexts of discrimination, not so much by increasing general well-being as by preventing its deterioration in such contexts. In addition, the effect is related to the bond of trust between mentor and participant, which highlights the importance of prioritising emotional support over instrumental support in programme design. Furthermore, programmes should anticipate the possible emergence of emotional distress during the intervention and strengthen participants’ own social support networks to sustain the effects over time.

Finally, it should be underlined that, due to the sample size and the quasi-experimental design, the results should be interpreted with due caution: the study’s limited size may imply that some real effects might not achieve statistical significance. Nevertheless, the results as a whole show a consistent pattern: eight out of every ten interactions are compatible with the hypothesis that mentoring acts as a protective factor against the effects of racism on well-being. A new study with a larger sample and longer follow-up would help determine whether mentoring could be adopted as a public health tool for populations exposed to structural discrimination.
 

BRADY, B.; P. DOLAN; J. CANAVAN (2015): «“He told me to calm down and all that”: a qualitative study of social support in youth mentoring relationships», Child & Family Social Work, 22(1), 266-274.

CRISTÓBAL-NARVÁEZ, P. et al. (2025). «Invisible Youths? Future Health and Social Care Challenges of Unaccompanied Minor Migrants in Spain», en Stanek, M., Juárez, S.P., Requena, M. (eds.) Multidisciplinary Perspectives on Immigrant Health, IMISCOE Research Series. Springer, Cham. 

EL-SHAARAWI, N.; S. LARCHANCHÉ (2022): Migration and Health: Challenging the Borders of Belonging, Care and Policy, Nueva York, Oxford: Berghahn Books. 

GABARRELL-PASCUET, A. et al. (2023): «Impact of perceived discrimination and coping strategies on well-being and mental health in newly-arrived migrants in Spain», PLOS ONE, 18(12). 

GARCÍA-CID, A. et al. (2020): «Discrimination and Psychosocial Well-Being of Migrants in Spain: The Moderating Role of Sense of Community», Frontiers in Psychology, 11:2235.

LEIGH-HUNT, N. et al. (2017): «An overview of systematic reviews on the public health consequences of social isolation and loneliness», Public Health, 152:157-171.

ORGANIZACIÓN MUNDIAL DE LA SALUD (OMS) (2017): Determinantes sociales de la salud mental. 

PRATS-BRUGAT, E.; X. ALARCÓN; Ò. PRIETO-FLORES (2026): «Reducing loneliness and promoting mental well-being among migrants and asylum seekers through social mentoring in Spain: insights from a qualitative study», BMC Public Health. 

PRIETO-FLORES, Ò.; J. FEU (eds.) (2021): Mentoring Children and Young People for Social Inclusion: Global Approaches to Empowerment, Abingdon / Nueva York: Routledge.

PUTNAM, R. D. (2000): Bowling Alone: The Collapse and Revival of American Community, Nueva York: Simon & Schuster.

STRANG, A. B.; N. QUINN (2021): «Integration or Isolation? Refugees’ Social Connections and Wellbeing», Journal of Refugee Studies, 34(1), 328-353.