Discussion

The purpose of this study was to examine the association between caregiver burden and depressive symptoms in dementia caregivers living in Puerto Rico, and to examine the role of BPSD and social support in this association. There were three major findings in our study. First, we found that caregiver burden and both BPSD factors were significantly and positively associated with depressive symptoms across all models, even after adjusting for sociodemographic characteristics, caregiving hours, and perceived social support. Second, the BPSD factor represented by perceptual disturbances/apathy significantly moderated (strengthened) the relationship between caregiver burden and depressive symptoms, and the effect appeared to be driven mainly by appetite/eating symptoms. Moderation was not observed for the factor representing emotional dysfunction/behavioral instability. Finally, perceived social support did not influence the results. These findings provide new evidence for the association between caregiver burden and depressive symptoms in Puerto Rico, highlight that BPSD, particularly appetite/eating problems, may lead to poorer mental health among caregivers, and that, in this culturally unique sample of dementia caregivers, social support may not play the expected role in the association between caregiver burden and depressive symptoms.

The finding that caregiver burden and both BPSD factors were significantly and positively associated with depressive symptoms across all models suggests that negative perceptions of the caregiving experience and the symptoms shown by the person with dementia are strong and independent predictors of depressive symptoms in caregivers. Notably, the correlation coefficient between caregiver burden and depressive symptoms (r = 0.72) was higher than the correlation (r = 0.52) reported in a systematic review (del-Pino-Casado et al., 2019), suggesting a stronger association in this sample.

We found that BPSD reflecting perceptual disturbances/apathy moderated the association between caregiver burden and depressive symptoms, with post hoc analyses indicating that this effect was largely driven by appetite/eating symptoms. These findings suggest that specific symptoms within the perceptual disturbances/apathy, particularly those related to appetite and eating, may play a pivotal role in amplifying the negative appraisal of caregiving. The finding that appetite/eating was the most significant moderator of BPSD is intriguing. We can speculate that this BPSD item may serve as a proxy for more intensive, hands-on caregiving, given the eating assistance needed to address this symptom. Regardless, appetite/eating difficulties in people living with dementia may deserve additional attention in attempts to reduce caregiver burden among Hispanic dementia caregivers.

Emotional dysfunction/behavioral instability did not moderate the association between caregiver burden and depressive symptoms. This suggests that while these symptoms might directly contribute to caregiver distress, they may not necessarily alter the caregiver’s overall negative appraisal of the caregiving experience itself.

In our study, perceived social support did not play a role in the association between caregiver burden and depressive symptoms. This finding contrasts with prior studies reporting buffering effects of perceived support on caregiver burden and depressive symptoms, and therefore warrants careful interpretation (del-Pino-Casado et al., 2018; Gutiérrez-Sánchez et al., 2023). One possible explanation is that the LSNS-6 may not have captured the construct of social support relevant to caregiving contexts, as it assesses the size of one’s social network rather than the actual receipt or use of support. Indeed, for this reason, del-Pino-Casado et al. (2018) excluded the LSNS-6 from their review. In addition, within Hispanic caregiving culture—particularly under familism and marianismo—merely having access to social ties does not necessarily translate into experienced support. Familism may create an expectation of family support that is not always realized in practice (Gelman, 2014), and marianismo may discourage women from requesting or utilizing available help by reinforcing self-sacrificial caregiving norms. Thus, even when social network resources are available, they may not manifest as meaningful support that alleviates caregiving stress. These cultural mechanisms provide a plausible explanation for why perceived support, as operationalized in this study, did not attenuate the association between burden and depressive symptoms.

Importantly, these culturally shaped caregiving norms operate within a broader structural context in Puerto Rico, where traditional family support systems have been weakened. Due to migration to the U.S. mainland, low fertility rates, and increasing life expectancy, as well as economic factors (Matos-Moreno et al., 2022), Puerto Rico is experiencing low levels of social support and high levels of social isolation (Gao et al., 2021). Thus, a combination of the gap between culturally expected and actually available family support, caregiving role norms that place primary responsibility on family members, particularly women, and broader structural conditions that limit the availability of support may jointly explain why perceived social support, as measured by network size rather than enacted support, did not attenuate the relationship between caregiving stress and depressive symptoms in this study.

These findings, considered in the broader sociocultural and demographic context of Puerto Rico, suggest several potential directions for caregiver support. First, appetite and eating difficulties may reflect a set of caregiving demands that are highly hands-on and time-intensive. Interventions that are symptom-focused and provide both skills training and opportunities for task sharing or respite may help reduce the extent to which these symptoms shape caregiving appraisal negatively, thereby mitigating adverse mental health outcomes. Second, although this study could not determine whether actual received support reduces depressive symptoms, the demographic reality of Puerto Rico, where shrinking kin networks and large-scale outmigration have reduced the availability of family care, suggests that informal support alone may no longer be sufficient. In this context, improving the accessibility of formal or community-based services may be essential to supplement reduced family caregiving capacity and to strengthen practical sources of support for dementia caregivers.

The results should be interpreted with caution due to several limitations. First, while studies using the SSC model incorporate measures such as familism to capture culturally specific values and explain variation across ethnic groups, this study did not include such items or comparative ethnic groups. Future research should incorporate measures of cultural values to better capture potential within-group and between-group variation among Puerto Rican caregivers. Although we could not empirically test these cultural values with our data, they can help interpret the findings within the sociocultural context of dementia caregiving in Puerto Rico, where the majority of the population identifies as Hispanic. Second, while the LSNS-6 assesses network size and the number of available support resources, it may not capture the actual amount of support caregivers receive or their perceived satisfaction with that support (Haber et al., 2007). Future research may more effectively measure the construct of perceived social support by using instruments that capture caregivers’ experiences of received support, allowing for a more accurate understanding of the effects of social support.

Third, we were not able to examine differences by caregiver-care recipient relationship type due to the relatively small sample size and the fact that most caregivers in our sample were adult children. Prior evidence shows that spousal caregivers are generally more vulnerable to burden than adult children because they are viewed as the primary attachment figure, tend to assume a higher level of responsibility, are more likely to co-reside with the care recipient, and often have poorer physical health, all of which heighten the overall caregiving burden (Pinquart & Sörensen, 2011). Future research should therefore include the type of caregiver–care recipient relationship and, further, incorporate qualitative indicators that capture the characteristics of the relationship (e.g., marital satisfaction, emotional closeness). Including these variables would allow for a more nuanced understanding of how relationship characteristics shape caregiving appraisal and, in turn, influence mental health outcomes. Finally, since this study used cross-sectional data, it is difficult to determine the directionality or causality of the relationships among the variables.

Our study has several strengths. First, it addresses a significant gap in the literature by focusing on dementia caregivers in Puerto Rico, an underrepresented population in caregiving and mental health research. Second, it extends prior research that identified caregiver burden and BPSD as independent predictors of depressive symptoms by providing empirical evidence that perceptual disturbances/apathy, particularly symptoms related to appetite/eating, can magnify the association between caregiver burden and depressive symptoms. Third, it highlights the possibility that appetite/eating-related symptoms, a symptom often overlooked in previous BPSD studies, may be a key contributor to depressive vulnerability among Puerto Rican dementia caregivers.

In conclusion, we found that both caregiver burden and BPSD were strongly associated with depressive symptoms among dementia caregivers in Puerto Rico, even after adjusting for sociodemographic characteristics, caregiving hours, and perceived social support as covariates. We also found that one BPSD factor, perceptual disturbances/apathy, moderated this association, and further analyses suggested that symptoms related to appetite and eating primarily drove this result. Finally, we observed that perceived social support did not reduce the magnitude of caregiver burden on depressive symptoms, likely because social network availability did not translate into enacted support within the sociocultural and demographic context of Puerto Rico. Therefore, improving access to formal or community-based services may be critical to offset limited family caregiving capacity and provide more tangible support for dementia caregivers, particularly for those facing intensive demands related to BPSD.