Accessibility to urban parks and resilience among adults in Spain: gender inequalities from the DESVELA study Authors: Irene Marcilla-Toribio, Maria Martinez-Andres, Irene Gómez-Gómez, Mónica Miranda-de la Maza, Leticia Sierra-Martínez, Irene Repiso-Gento, Ana Belén Castro-Rivero, Fátima Méndez-López, Oana Bulilete, María José Fernández-Domínguez, Ruth Martí-Lluch, Rafel Ramos, Lluís Zacarías-Pons, Blanca Notario-Pacheco Published: 2026 ABSTRACT AbstractIntroductionWhile associations between greenness and mental health have been studied, resilience, understood as the ability to adapt positively to stress and adversity, has received less attention. This study examined the relationship between the distance to the nearest park and resilience, considering potential gender differences.MethodsA cross-sectional study was conducted among 3,852 Spanish adults assigned to Primary Healthcare centers of nine autonomous communities. Resilience was assessed using the Connor-Davidson Resilience Scale. Unadjusted and adjusted differences in resilience and distance to the nearest park were assessed using Analysis of Variance, Analysis of Covariance, and multiple linear regression models in the total sample and stratified by sex. The covariates considered in this study were age, education, monthly income, marital status, perceived social support, and mental health. Sex was included as a covariate in analyses of the total sample. This study finally comprised 2,922 individuals, 52.0% women. The mean age of the participants was 53.6 years ± 10.9 SD.ResultsParticipants living farther from parks (> 700 m) showed significantly lower resilience than those in the close (< 300 m; mean difference = 0.59, 95% CI: 0.06, 1.11, p = 0.028) and medium (300 m-700 m; mean difference = 0.82, 95% CI: 0.28, 1.35, p = 0.003) subgroups. Greater distance to the nearest park was inversely associated with resilience in the total sample (β = -0.040, 95% CI: -0.075, -0.003, p = 0.030). Stratified analyses revealed this association was significant among men (β = -0.060, 95% CI: -0.110, -0.009, p = 0.020), but not among women (β = -0.018, 95% CI: -0.068, 0.030, p = 0.463).ConclusionsGreater residential distance to the nearest park was associated with lower resilience in adults, especially in men. These findings highlight the potential importance of urban green space accessibility for promoting mental well-being, underscoring the need to consider gender in public health research and interventions.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s12889-026-27985-y.Keywords: Greenness, Park, Resilience, Mental health, Spain, Gender Introduction The integration of green spaces into urban planning is increasingly recognised as a crucial component in promoting healthy and sustainable environments [1, 2]. Indeed, the distribution and accessibility of these spaces are not merely responsive to aesthetic or recreational criteria; rather, they serve as a critical determinant of health equity [3, 4]. A substantial body of research has demonstrated that exposure to urban green environments is associated with numerous physical and mental health benefits [5–8]. Regarding physical health, evidence shows that proximity to urban vegetation reduces risk of mortality [9–12] and morbidity, including a lower risk of cardiovascular disease [13, 14], obesity [15, 16], type 2 diabetes [17, 18], and high cholesterol levels [16, 18]. Furthermore, a number of studies have shown that residing in proximity to greenness is associated with increased levels of physical activity [19, 20]. Beyond physical health, the relationship between greenery and mental health has been studied [5, 7, 8]. Exposure to green spaces has been linked to reduced levels of stress, anxiety, and depression, as well as improvements in mood and overall psychological well-being [21, 22]. Among urban green spaces, parks represent a particularly relevant and widely accessible type. Unlike more natural or less structured green environments, parks are typically designed for public use and are characterized by the presence of recreational and service facilities, pedestrian pathways, and diverse vegetation layers [23]. These features may facilitate health-promoting behaviors such as physical activity and social interaction, while also providing opportunities for psychological restoration. Furthermore, their accessibility and integration within the urban fabric may make them especially relevant for population health compared to other types of green spaces, such as more natural or less accessible environments [24, 25]. In addition, to the best of our knowledge, no studies have specifically examined the relationship between green spaces and resilience. Resilience is a dynamic and multidimensional construct that refers to the ability to adapt positively and recover in the face of adversity, stress, or significant life challenges [26]. This process is regarded as dynamic, as opposed to being considered a fixed trait. It is influenced by both individual characteristics, including coping strategies, optimism, and perceived social support, as well as contextual factors, which encompass the physical and social environment [27–29]. Higher levels of resilience have been associated with better mental health outcomes, greater emotional stability, and improved overall well-being in adulthood [30]. Given its relevance to public health, resilience has become a central concept in understanding how individuals maintain mental health throughout their life course [31]. Addressing the gender perspective in health is essential, as men and women face different social conditions. These disparities not only affect the impact of health risks on individuals but also their ability to cope and recover [32, 33]. In the domain of mental health, women reported higher anxiety and depression prevalences [34]. These health disparities are also attributed to social determinants of health. Instead, they are deeply rooted in gender roles, care responsibilities, social discrimination, and societal expectations that shape how each group experiences stress and vulnerability [32]. Furthermore, revious research has also highlighted the presence of gender inequalities in resilience, with studies suggesting that men and women may differ in both levels of resilience and in the resources and mechanisms that support it, partly due to social roles, structural inequalities, and differential exposure to stressors [31, 35, 36]. The integration of a gender perspective is instrumental in facilitating a more profound comprehension of health patterns. Regarding green spaces, there is a paucity of research that explicitly considers gender. It is imperative to acknowledge integrating gender perspectives into research on green spaces to shed light on how access to green spaces may influence mental health differently in men and women. In addition, this approach would allow for more equitable urban policies and environments that are sensitive to the diverse needs of the population [37]. The objective of this study is to analyse the relationship between the distance to the nearest park and the resilience in the Spanish adult population, while exploring potential differences by sex. Methods Study design This cross-sectional study was conducted between 2021 and 2023 within the framework of a multicenter project designed to establish a prospective cohort of individuals, the DESVELA Study, registered in primary healthcare centres across nine autonomous communities in Spain (Cataluña, País Vasco, Castilla y León, Aragón, Galicia, Islas Baleares, Castilla-La Mancha, Andalucía and Madrid) [38]. Participants The study population comprised adults who were randomly selected from the assigned healthcare centers in each participating autonomous community. Individuals with a terminal condition or who were institutionalised at the time of recruitment, those with intellectual disabilities, dementia, or language difficulties, and those intending to relocate from Spain within five years of the study’s commencement were excluded from participation. Sampling and sample size The sample size was computed via the GRANMO sample size calculation. We considered the estimation of the relative risk for a cohort study using the Poisson approximation, with 10 years of follow-up and a rate of loss to follow-up of 30%, accepting an alpha risk of 0.05 and a beta risk of 0.2 in a bilateral contrast [38, 39]. Based on these parameters, a sample size of 3,083 subjects is required. A population-based ramdom sampling design of adults aged 25 to 74 years old that were referred to health centers in each of the nine autonomous communities was employed. After obtaining ethical approval, a comprehensive list of all patients meeting the inclusion criteria was extracted from the official administrative records of the participating Primary Care Centers. To ensure representativeness and account for potential non-response, a random selection was performed from this census, including a 30% oversampling for participant replacement if necessary. Finally, 3,852 individuals agreed to participate (response rate of 57%) in the DESVELA study. However, those who did not provide a postal address (1.5%) or had incomplete data, including missing resilience outcome (2.6%) and covariate information (19.9%), were subsequently excluded from this study, resulting in a final sample of 2,928 individuals. Ethics approval Research Ethics Committees of each participating Autonomous Community approved the DESVELA study. The research was conducted in accordance with the regulations of these committees and the principles of the Declaration of Helsinki. The studies involving human participants were reviewed and approved by Fundació Institut Universitari per a la recerca a l’Atenció Primària de Salut Jordi Gol i Gurina (reference number 19/150-P); Comité de Ética de la Investigación con medicamentos del Área de Salud de Salamanca (reference number PI 2020 02424); Andalusian Ministry of Health, Spain (reference number: 1260-M1-21); Comité de Ética de la Investigación de medicamentos de Euskadi (CEIm-E; reference number: PI2020185); Hospital Virgen de la Luz Clinical Reasearch Ethics Comiitte, Cuenca, Spain (Reference number 2019/PI2119). Research Central Commission of the Primary Care Assistance Management, Madrid, Spain. (Reference Number 07/21); Comité de Ética de la Investigación de la Comunidad Autónoma de Aragón (reference number: PI20/302); Galician Ministry of health, Spain: high impact study authorization (Reference number: 2021/047). All study participants were informed of the study's objectives, and written informed consent was obtained prior to individual participation. Study variables In this study, data collection was carried out by trained staff under the internal procedures of the DESVELA cohort study to ensure standardisation and the quality of the process [38, 40, 41]. Exposure: distance to the nearest park Distance to the nearest park was calculated for each participant based on their registered postal address. First, the participants’ residential addresses were geolocated using geocoding tools based on the Google Maps Application Programming Interface (API), implemented via Google Sheets [42, 43], which returned geographic coordinates in the World Geodetic System (WGS84) format. The coordinates were subsequently entered into Google Maps [44, 45] because this platform offers high spatial resolution, accessibility, and frequent updates [46, 47]. The maps used were up to date as of 2025. The distance to the nearest park was then calculated as the network-based walking distance in meters from each participant’s residence to the nearest park using the walking mode, which accounts for the street and pedestrian pathway network [48]. Additionally, all identified destinations were verified using satellite imagery, updated as of 2024, to ensure that they corresponded to public green spaces, specifically parks, excluding other types of open spaces such as squares or predominantly paved areas. Although no strict minimum size threshold was applied, all identified parks corresponded to green areas featuring recreational and service facilities, with diverse vegetation layers [23, 25]. Google Maps data were accessed in June 2025. Outcome: resilience The Spanish version [49] of the 10-item Connor-Davidson Resilience Scale (CD-RISC) [50] was used to assess resilience. This questionnaire employs a five-point Likert-type scale, with response options ranging from 0 (never) to 4 (almost always). The total score is obtained by summing the responses to all items (range: 0–40), with higher scores indicating greater resilience. In the present study, the scale showed good internal consistency, with a Cronbach’s alpha of 0.83. Covariates The covariates considered in this study were age, education, monthly income, marital status, perceived social support, and mental health. Biological sex was collected as a self-reported variable; participants were given the option to withhold sex information. Age, when used as a categorical variable, was divided into three groups defined as young adults (≤ 39 years), middle adults (40–59 years), and older adults (≥ 60 years) [51]. Education was categorized into three groups: individuals without formal schooling were grouped with those having primary education due to the small number of participants without schooling. Monthly household income was reported in predefined categories (low, medium and high), ranging from “no income” to “more than €5,000”. Marital status was categorized in four categories including single, married, divorced and widowed. Mental health was assessed using the Spanish version [52] of the 12-Item Short Form Health Survey (SF-12), considering a score lower than 50 as “bad mental health” [53], while perceived social support was measured with the Duke-UNC Functional Social Support Questionnaire (Duke-UNC-11) Spanish version, considering a score lower than 32 as “no social support” in its categorical form [54]. Sex was included as a covariate in analyses of the total sample (i.e., without stratification), and participants were given the option to withhold this information. Statistical analysis Distance to the nearest park was analysed both as a categorical variable and a continuous variable. For categorical analyses, distances were divided into tertiles based on 33% cut-offs to ensure comparable group sizes. For continuous analyses, the variable was normalised using Blom’s rank-based inverse normal transformation to address the high variability in distances across participants [55]. Resilience, the outcome variable, was analysed as a continuous variable. Normality of continuous variable distributions was tested using the Kolmogorov–Smirnov test and normal probability plots. Categorical variables were reported as frequencies and percentages, and continuous variables as means and standard deviations (SDs). Differences in categorical variables were assessed using the chi-square test. Comparisons of continuous variables between two groups were performed using the Student’s t-test, and between more than two groups were performed using one-way analysis of variance (ANOVA). A p-value < 0.05 was considered statistically significant in sex-related comparisons. One-way ANOVA was applied to compare resilience scores across categories of distance to the nearest park. Stratified analyses by sex as well as socioeconomic and psychosocial variables were also carried out. Analysis of covariance (ANCOVA) was subsequently applied to evaluate mean differences in resilience across park distance categories, adjusting for sex, age, marital status, education level, monthly income, mental health, and perceived social support. Multiple linear regression was additionally fitted, with resilience (z-scores) as the dependent variable and park distance as the independent variable. An initial unadjusted model was estimated, followed by an adjusted model including sex, age, marital status, education level, monthly income, mental health, and perceived social support. Sex-stratified analyses were also performed. Furthermore, to formally assess effect modification, a multiplicative interaction term between park distance and sex was included in the fully adjusted model. In ANCOVA, group comparisons were conducted using the Bonferroni post hoc test, and bootstrapping [56] with 5,000 resamples was applied to obtain robust estimates. Multiple linear regression analysis were also estimated with bootstrapping (5,000 resamples). All analyses were conducted using SPSS software (IBM Corporation, V29.0.2.0 Armonk, NY, USA) [57] and R software (Foundation for Statistical Computing, V.2023.12.1 + 402, Boston, MA, United States) [58]. Results Table 1 shows the baseline characteristics of the study participants, categorised by sex. The study sample consisted of 2,922 participants, of whom 52.0% were women. The mean age of the total sample was 53.6 years ± 10.9 SD. Bivariate analyses indicated statistically significant differences by sex in marital status (p < 0.001), level of education (p = 0.002), monthly income (p < 0.001), and mental health (p < 0.001). Table S1 in the Supplementary Material presents sociodemographic and psychosocial characteristics stratified by park distance tertile groups. Table 1Study sample sociodemographic characteristicsSociodemographic characteristicsTotaln (%)MenWomenpSample size n(%)2,922 (100.0)1,404 (48.0)1,518 (52.0)-Mean age (SD)53.6 (10.9)53.9 (11.0)53.4 (10.7)0.121Marital status n(%)< 0.001* Single419 (14.3)206 (14.7)213 (14.0) Married2,163 (74.0)1,079 (76.9)1,084 (71.4) Divorced250 (8.6)106 (7.5)144 (9.5) Widowed90 (3.1)13 (0.9)77 (5.1)Level of education n(%)0.002* Primary education610 (20.9)296 (21.1)314 (20.7) Secondary education1,272 (43.5)652 (46.4)620 (40.8) University education1,040 (35.6)456 (32.5)584 (38.5)Monthly income n(%)< 0.001* Low627 (21.5)252 (17.9)375 (24.7) Medium1,348 (46.1)638 (45.4)710 (46.8) High947 (32.4)514 (36.6)433 (28.5)ResilienceMean (SD)29.2 (6.2)29.4 (6.3)29.1 (6.0)0.518Perceived social supportMean (SD)45.6 (7.5)45.5 (7.5)45.6 (7.5)0.469SF-12 (mental)Mean (SD)52.7 (7.8)53.8 (7.3)51.7 (8.2)< 0.001*Distance to the nearest parkMeters, mean (SD)804.8 (1127.6)794.8 (1125.6)814.2 (1129.8)0.973The column labelled "p" refers to the chi-square test and t-test statistics performed, and the asterisk (*) indicates p < 0.05. For the age, resilience, perceived social support, distance to the nearest park, and SF-12 mental variables, the t-test statistic was calculated. The rest of the variables, which integrated qualitative sociodemographic information, were calculated with the chi-square testAbbreviations: km kilometers, m meters, SD Standard deviation Study sample sociodemographic characteristics Resilience Mean (SD) Perceived social support Mean (SD) SF-12 (mental) Mean (SD) Distance to the nearest park Meters, mean (SD) The column labelled "p" refers to the chi-square test and t-test statistics performed, and the asterisk (*) indicates p < 0.05. For the age, resilience, perceived social support, distance to the nearest park, and SF-12 mental variables, the t-test statistic was calculated. The rest of the variables, which integrated qualitative sociodemographic information, were calculated with the chi-square test Abbreviations: km kilometers, m meters, SD Standard deviation ANOVA results indicated that mean resilience scores varied significantly across distance groups (F = 3.934, p = 0.020). Overall, participants residing at a medium distance from the nearest park reported slightly higher resilience scores (29.5 ± 6.0) compared to those residing at a far distance (28.8 ± 6.1), suggesting lower resilience among individuals living farther from green spaces. Sex-stratified ANOVA revealed that the association between distance to the nearest park and resilience was significant among men (F = 4.123, p = 0.016), but not among women (F = 0.646, p = 0.525). In men, lower resilience was generally observed in the far-distance group compared to the medium-distance group. Stratified ANOVA results by sex and sociodemographic and psychosocial characteristics are presented in Fig. 1. In young adult men (≤ 39 years), resilience was higher among those residing close to the nearest park compared to those living far away, indicating lower resilience with greater distance (F = 6.491, p = 0.002). In middle-aged men (40–59 years), resilience was higher in the medium-distance group compared to the far-distance group (F = 3.189, p = 0.042). Similarly, married men showed higher resilience in the medium-distance group compared to the far-distance group (F = 4.894, p = 0.008). Among men with primary education, resilience was higher in those living at medium versus far distance (F = 3.423, p = 0.034). Men with medium income also showed higher resilience in the medium-distance group compared to the far-distance group (F = 3.400, p = 0.034). Moreover, men with high perceived social support had higher resilience when living at medium versus far distance (F = 5.079, p = 0.006). Finally, men reporting good mental health also exhibited higher resilience in the medium-distance group compared to the far-distance group (F = 4.482, p = 0.011). Comparison of resilience means scores across distance to park groups, stratified by sex and sociodemographic and psychosocial variables Figures 2 and 3 show the adjusted marginal mean differences in resilience scores between participants across park distance categories and by sex. Overall, participants living in the far (> 700 m) distance group had significantly lower resilience than those in the close (< 300 m; M = 29.35 vs 28.76; mean difference = 0.59, 95% CI: 0.06, 1.11, p = 0.028) and medium (300 m-700 m; M = 29.57 vs 28.76; mean difference = 0.82, 95% CI: 0.28, 1.35, p = 0.003) subgroups. Moreover, men’ resilience was also significantly lower in the far (> 700 m) distance subgroup compared with the close (< 300 m; M = 29.45 vs 28.56; mean difference = 0.89, 95% CI: 0.09, 1.67, p = 0.029) and medium (300 m-700 m; M = 30.02 vs 28.56; mean difference = 1.45, 95% CI: 0.68, 2.23, p = < 0.001) groups. In contrast, none of the comparisons reached statistical significance among women, although small differences were observed between the far and both the close (< 300 m; M = 29.22 vs 28.92) and medium (300 m-700 m; M = 29.17 vs 28.92) groups. Adjusted mean resilience scores by park distance and sex Adjusted mean differences in resilience scores by park distance and sex As shown in Table 2, greater distance to the nearest park was significantly associated with lower resilience in the total sample (unadjusted: β = −0.045, 95% CI: −0.082, −0.007, p = 0.019). This association remained significant after adjustment for sociodemographic and psychosocial covariates (β = −0.040, 95% CI: −0.075, −0.003, p = 0.030). When stratified by sex, a significant negative association was observed only among men. In the unadjusted model, park distance was inversely associated with resilience (β = −0.060, 95% CI: −0.113, −0.007, p = 0.026), and this association remained significant in the adjusted model (β = −0.060, 95% CI: −0.110, −0.009, p = 0.020). Among women, no significant associations were found in either the unadjusted (β = −0.030, 95% CI: −0.081, 0.021, p = 0.256) or adjusted (β = −0.018, 95% CI: −0.068, 0.030, p = 0.463) models, although the estimates suggested a negative trend. Despite these differences in stratified estimates, the multiplicative interaction between park distance and sex did not reach statistical significance (β = 0.122, p = 0.229). Z-scores for all covariates included in the adjusted linear regression models are presented in Supplementary Table S2. Table 2Standardised beta coefficient (95% confidence Intervals) from Multiple Linear Regression Analysis of park distance predicting resilience (CD-RISC-10), stratified by sexGroupModelZ-scorep-valueTotal sampleUnadjusted−0.045 (−0.082, −0.007).019Adjusted−0.040 (−0.075, −0.003).030MenUnadjusted−0.060 (−0.113, −0.007).026Adjusted−0.060 (−0.110, −0.009).020WomenUnadjusted−0.030 (−0.081, 0.021).256Adjusted−0.018 (−0.068, 0.030).463Adjusted models control for mental health (SF-12 mental), perceived social support, monthly income, education level, age, and marital status. Additionally, the adjusted total sample model controls for sex. All estimates were obtained using bootstrapping with 5,000 resamples Standardised beta coefficient (95% confidence Intervals) from Multiple Linear Regression Analysis of park distance predicting resilience (CD-RISC-10), stratified by sex Adjusted models control for mental health (SF-12 mental), perceived social support, monthly income, education level, age, and marital status. Additionally, the adjusted total sample model controls for sex. All estimates were obtained using bootstrapping with 5,000 resamples Discussion This study examined the association between distance to the nearest park and resilience in the Spanish adult population, considering potential gender differences. We found that a greater distance to parks was associated with lower resilience, exclusively in men. In the total sample, for every SD farther from a park, resilience decreased by 0.04 SDs. In men, the decrease was stronger, with resilience falling by 0.06 SDs for every one SD increase in distance. Notably, the strongest association was observed for participants residing at an intermediate distance from parks, approximately 300–700 m from the nearest green space. Our results suggest that a greater distance of the park from the residence was associated with lower resilience in adults. To our knowledge, no previous studies have examined the association between distance to the nearest park and resilience in adults. However, these findings align with a growing body of research showing that proximity to greenery is linked to better mental health outcomes, including depression, anxiety, stress, and well-being [18, 22, 30, 59, 60]. Parks have important venues for social interaction and community engagement [24], both of which are well-established protective factors of resilience [61–63]. Previous research has shown that green spaces can promote social contact and neighbourhood ties, contributing to better psychological outcomes [24, 64, 65]. In fact, social interaction has been identified as one of the key pathways through which green space exerts its beneficial effects on mental health [65]. It is important to note that physical activity may represent another key pathway linking proximity to greenery with greater resilience. Evidence has shown that accessible and attractive green spaces are conducive to walking, running, and recreational exercise [5, 66, 67]. In turn, engaging in physical activity is associated with elevated levels of resilience [68, 69]. Interestingly, our study also found that the strongest association between park distance and resilience occurred for participants residing at an intermediate distance from parks, approximately 300 m-700 m. One possible explanation is that parks located very close to home may expose residents to higher levels of noise, crowding, or other disturbances, which could impact mental health [70, 71]. In contrast, an intermediate distance requires a short walk to reach the park, providing an opportunity for cardiovascular physical activity, which itself can enhance mood and resilience [72]. Thus, when green spaces are located farther from residences, opportunities for casual encounters, recreational activities, and community bonding may be reduced, which could explain the negative association between distance to parks and resilience observed in our study [4, 73, 74]. However, the physical activity interpretation may not apply equally to all individuals. People with physical limitations or restrictions in daily life activities, such as walking, may encounter barriers to accessing parks even at relatively short distances. For these individuals, increased distance may disproportionately limit opportunities for physical activity, social interaction, and psychological restoration. Therefore, functional status and mobility should be considered as relevant social determinants when examining the relationship between park accessibility and resilience [75, 76]. Although our data suggest a tendency for women residing closer to parks to report higher resilience, the negative association between distance to parks and resilience reach statistical significance in men. Regarding gender differences of mental health and urban greenness, there is a discrepancy in the literature as some studies suggest greater benefits for women, while others report more significant effects for men [21]. Previous research suggests that men and women may use parks differently. Men are more likely to engage in leisure activities or physical exercise, whereas women’ visits are more often linked to caregiving roles, such as accompanying children [73, 77]. Furthermore, studies in various cultural contexts worldwide have found that women perceived urban green spaces, such as parks, as potentially unsafe more often than men. Safety plays an important role, as this perception can restrict their use of parks [78, 79]. Indeed, this negative women perception has led to the attempt to resolve the problem with women-only parks in Oslo [80]. Several age and gender related factors may influence resilience and mental health patterns in women. The average age of participants was approximately fifty, a stage of life in which many women experience substantial caregiving demands, often simultaneously for children and ageing parents. With regard to these circumstances confronting women, career interruptions related to caregiving may increase stress and financial insecurity [81, 82]. Furthermore, around midlife, the transition through menopause may affect mental health. The hormonal changes in this period play important roles in regulating mood, sleep, and stress reactivity. Fluctuations in hormones combined with psychosocial stressors may exacerbate mental health challenges. Taken together, such factors could diminish the potential benefits that residential proximity to parks might otherwise confer on resilience among women in our sample [83–85]. Another relevant aspect of this research was that our findings reveal an increase in resilience as socioeconomic status and education level increased, as reported in previous studies [31, 36]. In addition, resilience tended to be lower among individuals who lived farther from parks, regardless of their socioeconomic status or educational background. This trend was observed in both men and women although the statistical significance was only in men. This suggests that residing closer to parks seems to provide an independent positive effect on resilience to adult men and women. Regarding age, younger and middle-aged men appear more affected by far distances. This could be explained by the fact that at these stages of life, study or work commitments can be demanding, meaning that parks located further away from home are used less frequently [74, 86]. Controversially, the greater the distance from the park, the greater the resilience in women. This life stage often coincides with entering the labour market, where work environments and professional networks can provide important sources of social support [87]. In addition, this is a period when young adults typically move out of the family home and establish independence, which may involve relocating to neighbourhoods that are farther from their previous social circles. This fact may necessitate a visit to a park located at a considerable distance from the place of residence, where can access a social support network, a phenomenon that is particularly salient among women, who appear to be more affected by this process [88–90]. This study has several strengths. Firstly, it addresses a novel research area by examining the relationship between residential proximity to parks and resilience in adults, a subject that has not been studied. Secondly, using objective measures of park proximity enhances the study’s methodological rigour compared to relying on self-reported access, with a key strength of this study as the use of network-based walking distance. This approach provides a more realistic estimate of accessibility compared to commonly used Euclidean distance measures, as it accounts for the actual street and pedestrian pathway network. Furthermore, the manual verification of destinations using satellite imagery ensured accurate identification of public green spaces, reducing the risk of misclassification. Thirdly, analysing sex differences provides valuable insight into subgroup-specific patterns and controlling for relevant covariates such as age, socioeconomic status, and education increases the reliability of the observed associations. Finally, the large and geographically diverse sample from multiple autonomous communities in Spain, including more urban and more rural areas, enhances the representativeness of our findings. This study is not without limitations. One the one hand, the cross-sectional design precludes causal inferences and raises the possibility of reverse causation; for instance, individuals with higher incomes may preferentially choose to reside in areas with better designs and access to green spaces, such as proximity to parks. On the other hand, the analyses of this study did not account for certain neighbourhood-level characteristics, such as the neighbourhood deprivation index, which may act as relevant confounders in the relationship between green space exposure and mental health outcomes. Other unmeasured factors, such as park quality, safety perception, frequency of park use, noise, and even physical limitations or functional status, may have influenced the results. In addition, resilience data were collected via a self-reported questionnaire, although validated, these responses may be subject to bias or inaccurate reporting. Moreover, while multiple linear regression is a valid and robust approach for this research, our models did not account for the potential clustering effects of Autonomous Communities; therefore, future studies should consider multilevel modeling to further refine these findings. Furthermore, despite the substantial sample size of the present study, the observed sex differences should be interpreted with caution due to the lack of a statistically significant multiplicative interaction. Additionally, a proportion of participants were excluded due to missing data and future studies could consider the use of multiple imputation or other advanced missing data techniques to assess the robustness of these findings. Finally, although distances were calculated using Google Maps, a widely used and practical tool, the results may be influenced by potential variability in routing algorithms and periodic updates to the underlying map data. Moreover, while this approach captures the street and pedestrian network, it does not account for other factors that may influence accessibility, such as temporary barriers, access restrictions, or the quality of the walking environment. These findings suggest that promoting access to nearby parks from residences could be a practical strategy to enhance resilience in adults. From a public health perspective, policies could focus on improving the availability, safety, and quality of parks, particularly in urban areas, and designing community programs that promote inclusive and regular use of these spaces [4, 91]. Such an initiative could help reduce mental health disparities and foster resilience at the adult population level. Finally, qualitative studies are needed to better understand how men and women use parks differently. Insights from such studies could inform interventions and guide park design and programming, applying a socio-ecological perspective to ensure that women can also fully benefit. These findings highlight the potential importance of green space accessibility for promoting mental well-being. Furthermore, studying the sex-specific differences underscores the need to consider gender in public health research and interventions, thereby ensuring that both men and women can reap the maximum benefits from residential greenness. Conclusion In conclusion, the present study suggests that a greater distance from residence to the nearest park is associated with lower resilience in adults. The association was statistically significant only in men, although a similar trend was observed in women. The association was most pronounced at intermediate distances (300 m-700 m) from participants’ residential addresses, resulting in a walking time of between five and ten minutes to reach the park. Supplementary Information Supplementary Material 1. Acknowledgements Andalusia—Loyola Andalucía University: Emma Motrico and Irene Gómez-Gómez and Instituto de Investigación Biomédica de Málaga: Patricia, Moreno-Peral, Sonia Conejo-Cerón, and Juan Ángel Bellón. Aragon—Fundación Instituto de Investigación Sanitaria Aragón: Rosa Magallón-Botaya, Fátima Méndez-López, Alejandra Aguilar-Latorre, Maria Beltrán-Ruiz, Bárbara Oliván-Blázquez, Marta Domínguez-García, María Isabel Rabanaque Hernández, and Eva María Andrés Esteban. Castilla La Mancha—Centro de estudios sociosanitarios: Blanca Notario Pacheco, Montserrat Solera Martínez, Lidia Lucas-de la Cruz, Miriam Garrido Miguel, María Martínez Andrés, María Eugenia Visier Alfonso, and Irene Marcilla Toribio. Castilla y León—Salamanca Primary Care Research Unit: José A MaderueloFernández, Leticia Sierra-Martínez, Olaya Tamayo-Morales, Miriam Daniela GarcíaCubillas, Ana B. Castro-Rivero, María D. Martín-Santos, Carmen Castaño Sánchez, and Luis García-Ortiz. Catalunya—Fundació Institut Universitari per a la recerca a l’Atenció Primària de Salut Jordi Gol i Gurina: Bonaventura Bolíbar, Ruth Martí-Lluch, Rafel Ramos, Yudy Young-Silva, Marc Casajuana-Closas, Anna Berenguera, Constanza Jacques-Aviñó, Lia Alves-Cabratosa, Lluís Zacarías-Pons, Anna Ponjoan, Eva Espigulé-Ribas, Francesc Ribas-Aulinas, Jordi Blanch, Èric Tornabell-Noguera, and Anna Moleras-Serra; Sant Joan de Déu Health Park: Enric Vicens-Pons, Montserrat Gil-Girbau, Mari Carmen Olmos Palenzuela, María del Carmen Gallardo González, Mª Teresa Peñarrubia-María, and Paula Arroyo-Uriarte; and Community of Madrid, Infanta Mercedes Health Centre, Madrid Health Service: Francisco Camarelles Guillem. Euskadi—Institute for Health Research Biocruces Bizkaia: Jose María Aiarzaguena, Álvaro Sánchez Pérez, Sandra Garcia-Martinez, Usue Elizondo Alzola, Mónica Miranda de la Maza, Ainhoa Abrisketa Ullibarri, and Mikel Rueda-Etxebarria. Galicia—Galicia Sur Research Institute: Mª José Fernández Domínguez, Sabela Couso Viana, Roberto Fernández Alvarez, Ana Claveria Fontan, Ana Isabel Castaño Carou, Clara González Formoso, María Victoria Martín Miguel, and Clara Guede Fernández. Balearic Islands—Gerencia de atención primaria de Mallorca, Instituto de investigación sanitaria de las Islas Baleares: Joan Llobera Cànaves, Caterina Vicens, Maria J. Serrano-Ripoll, Laura Gallardo-Alfaro, Oana Bulilete, Christian Jean-Mairet Soler, David Medina-Bombardó, and TColl Benejam. Authors’ contributions Conception and design: IMT, MMA, BNP. Analysis and interpretation of data: IMT, MMA, BNP. Writing: IMT, MMA, BNP. Review and editing: IMT, MMA, BNP, IGG, MMM, LSM, IRG, ABCR, FML, OB, FDMJ, RML, RR, LZP. Administrative, technical, or material support: RML, RR. Funding This study has been funded by Instituto de Salud Carlos III (ISCIII) with competitive grants for the period 2019–2022 through the Fondo de Investigación Para la Salud (FIS), which is co-funded by European Regional Development Fund/European Social Fund “A way to make Europe”/“Investing in your future”. Project Grants codes are: P19/01285; P19/00997; P19/01140; P19/00147; P19/01076; P19/00434; P19/01459; P19/01314; P19/01264; and P19/00115. The coordinator group received a pre-doctoral training contract in health research (PFIS-FI20/00270) from the 2020 call of the Strategic Action in Health 2017–2020, co-funded by the European Union–Next Generation EU. Investigation groups were also funded through the Research Network in Preventive Activities and Health Promotion in Primary Care (redIAPP), RD16/0007/0001; RD16/0007/0002; RD16/0007/0003; RD16/0007/0004; RD16/0007/0005; RD16/0007/0006; RD16/0007/0008; RD16/0007/0009; RD16/0007/0010; and RD16/0007/0012; and through the research grants on the call for the creation of Health Outcomes-Oriented Cooperative Research Networks (RICORS) co-funded with European Union-Next Generation EU funds, allowing the creation of the Network for Research on Chronicity, Primary Care, and Health Promotion (RICAPPS) with the following references: RD21/0016/0001; RD21/0016/0003; RD21/0016/0005; RD21/0016/0009; RD21/0016/0010; RD21/0016/0012; RD21/0016/0018; RD21/0016/0022; RD21/0016/0025; and RD21/0016/0029,. Additional grants: Regional Gerencia Regional de Salud de Castilla y León (GRS 2306/B/21 and GRS 2356/B/21); Andalusia Ministry of Education and Science (PY20 RE 025); the VII PPIT-US. The funders had no role in the study design, writing of the report, or in the decision to submit the protocol for publication. All authors confirm that they worked independently from funders. Data availability The datasets used during the current study available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate Research Ethics Committees of each participating Autonomous Community approved the DESVELA study. The research was conducted in accordance with the regulations of these committees and the principles of the Declaration of Helsinki. The studies involving human participants were reviewed and approved by Fundació Institut Universitari per a la recerca a l’Atenció Primària de Salut Jordi Gol i Gurina (reference number 19/150-P); Comité de Ética de la Investigación con medicamentos del Área de Salud de Salamanca (reference number PI 2020 02424); Andalusian Ministry of Health, Spain (reference number: 1260-M1-21); Comité de Ética de la Investigación de medicamentos de Euskadi (CEIm-E; reference number: PI2020185); Hospital Virgen de la Luz Clinical Reasearch Ethics Comiitte, Cuenca, Spain (Reference number 2019/PI2119). Research Central Commission of the Primary Care Assistance Management, Madrid, Spain. (Reference Number 07/21); Comité de Ética de la Investigación de la Comunidad Autónoma de Aragón (reference number: PI20/302); Galician Ministry of health, Spain: high impact study authorization (Reference number: 2021/047). All study participants were informed of the study's objectives, and written informed consent was obtained prior to individual participation. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The datasets used during the current study available from the corresponding author on reasonable request.