Fifty million people are estimated to live with dementia around the globe (1). This group of disorders is a main cause of dependence among older adults (2). Worldwide, families are the backbone of long-term care provided to people with dementia (PwD) and carry much of the burden of the disease (3).

Throughout the caregiving trajectory, informal dementia caregivers are at greater risk of experiencing depression and anxiety disorders, when compared both to the general population and to caregivers of people living with other chronic diseases (4). Informal caregivers tend to report poor diet, unhealthy sleep and poor exercise habits and often face financial strain, isolation, and discrimination (5, 6).

Higher levels of stress are thought to be experienced by caregivers who perceive their coping resources as being beneath their demands (7). To deal with caregiving, caregivers may rely on a plethora of information and support sources, including care professionals, community services, printed information materials, and internet-based resources (8).

The increasing attention given to internet interventions aimed at dementia caregivers has put in perspective the role that web resources may play in the upcoming years. Such resources may diverge in complexity, ranging from static informational web platforms to interactive and customizable web programs/apps providing psychoeducation, training, or health support (9). Internet-based resources may provide accessible and affordable opportunities to tackle barriers faced by caregivers in using in-person psychosocial services. Geographic inequalities concerning service offers, 24/7 or very intensive care, transportation issues, or experiencing stigma, are among the reasons hampering the access of caregivers to in-person support (2). Dementia caregivers have shown interest in using internet resources to find support, connect with other caregivers, and seek information about dementia, financial, emotional, and practical aspects of caregiving (10–12).

Only a few studies have explored the factors associated with the use of internet-based health and caregiving-related resources by dementia caregivers (5, 11–20). Most studies resort to heterogeneous samples of caregivers (caring in different chronic diseases), and findings were inconsistent on the factors associated with the use of health and caregiving-related internet resources. The use of such resources was associated in some studies (but not in others) with caregivers' age (5, 13, 15), education (5, 13, 15), income (15), hours spent caring (15), relationship with the care recipient (15), emotional stress (15, 16), and perceived health status (5). Qualitative research has stressed the role of burden of care (20), stage of the disease (14), years spent caring, social support, and self-efficacy (17). Conflicting literature findings may result, among other reasons, from a substantial heterogeneity in the inquiry process that supports the categorization of individuals as users or non-users of internet resources. Most studies base such categorization on generic questions about health or caregiving-related internet use, with no further specification of such purposes. Much less provide information on the use of the internet for specific purposes (13). Online health, dementia, or caregiving-related resources are highly diverse regarding the information they convey and pose different user interaction demands (passive as, e.g., in an informational website, vs. active, e.g., in a psychoeducational program). The motivations and the profiles of caregivers may differ among those searching for information, looking for training or support on practical care, or searching for resources on self-care. A disaggregated analysis of how different factors may relate to different internet uses is missing in the literature. Moreover, caregivers are usually categorized according to the dichotomy of being users or non-users of the internet for health, dementia, or caregiving-related purposes, a grouping that presents many limitations in explaining caregivers' resource seeking behavior (15). In a digital era, non-users of the internet, both in general and for health-specific purposes (at a very minimum, basic online searches), are becoming less prevalent in developed countries. At least for research conducted in those countries, non-users of health-related internet resources are becoming circumscribed to the most economically, socially, and digitally excluded groups in society. Research comparing users and non-users of health resources on the web is at risk of confusing factors associated with such use, with those related to digital literacy. A more fine-grained analysis is required, focusing on individuals who can use the internet, as well as on the factors associated with the frequency of using—rather than the use/non-use- of dementia and caregiving-related internet resources.

Another caveat of studies in this field is the scarce attention given to “enabling” or “need” variables (21) such as caregivers' service utilization history or physical/psychological health, contrasting to the great focus on predisposing variables (15).

In tackling current research gaps, this cross-sectional study investigates the information and support-seeking behavior of dementia caregivers focusing on the utilization of internet resources. Our objective is threefold: 1) describe the use of dementia and caregiving-related internet resources in a sample of informal dementia caregivers who can use the internet autonomously, regarding the frequency and purposes of use; 2) compare dementia caregivers who characterize differently on predisposing, enabling, and need variables on their frequency of using dementia and caregiving-related internet resources (for eight different purposes); and 3) identify predisposing, enabling, and need variables which are associated with the overall frequency (composite measure) of using dementia and caregiving-related internet resources. The study was conducted in Portugal, a country with a high prevalence of dementia (21 cases per 1,000 inhabitants) (22), and high rates of informal co-residential care (23).

This exploratory study sought to investigate online information and support-seeking behaviors of informal dementia caregivers who can use the internet autonomously. No similar study was carried out in Portugal and, internationally, the research topic is scantly researched. A disaggregated analysis of internet use for eight different caregiving-related purposes, together with a focus on the frequency of use rather than on the dichotomy of use/non-use are two distinctive elements of our study design compared to research produced to date. We have concluded for a moderate frequency of using dementia and caregiving-related internet resources. Across all the eight specific purposes investigated, a “frequent” internet use was never observed in more than 46% of the sample. The average score of the composite measure of dementia and caregiving-related internet use was 20.8 in 32 possible points. In an American study, Kim et al. have concluded that only 59.1% of dementia caregivers were health-related internet users (15); while in a South-European study, 47.5% of caregivers have received dementia services online (13). In our study, the caregivers using internet resources, regardless of the frequency, range from 49% (for using internet interventions) to 93% (to gather information on dementia). Hence, the prevalence of caregiving-related internet use is higher than the prevalence reported in previous studies, and higher than health-related internet use estimated for the general Portuguese population [49% (33)]. However, there is a high variability on the measurement of health, dementia, and/or caregiving-related internet use [e.g., a single item in (15) on internet resources “related with being a caregiver”; vs. asking about receiving dementia services online in (13)]. Therefore, no direct comparisons can be made. Moreover, internet usage trends are rapidly evolving, and distinct study timeframes may be influential. Our finding may also relate to the fact that we have studied a sample of frequent internet users.

In describing internet use for distinct purposes, we concluded that caregivers show less information and support-seeking behaviors for their own benefit than for directly benefiting the person in care. Caregivers may perceive the use of internet resources as an additional caregiving task rather than a help-seeking behavior (34). Moreover, reluctance in help-seeking for mental health is a well-known issue and may contracyclically be more expressed by the persons who are more in need (35). Using the internet to share experiences with other caregivers was infrequently reported in our sample, in line with findings from another European study (13).

This study also sought to identify variables associated with the frequency of using dementia and caregiving-related internet resources. First, we compared dementia caregivers who characterize differently on predisposing, enabling, and need variables on the frequency of using the internet for each of the eight dementia and caregiving-related purposes under study. We concluded that associations of predisposing, enabling, and need variables varied across specific caregiving-related internet use purposes (Supplementary Table). Second, we resorted to linear regression models to assess the impact of predisposing, enabling, and need variables on a composite measure of such usage frequency. The final multivariable linear regression model suggests that being younger, not having a usual source of support to provide care, having used a face-to-face psychosocial intervention at some point, being employed, and perceiving the own physical health as worse when compared to counterparts, are associated to a higher frequency of using dementia and caregiving-related internet resources. Perceived physical health and lifetime access to psychosocial interventions were the variables with the strongest unique contribution to explaining such frequency. These results highlight the relevance of including enabling and need variables in future research. The association of lifetime access to psychosocial interventions with caregiving-related internet use was a new addition from our study to the field.

Our findings confirmed the associations of age (5, 13, 15), relationship with the person in care (in the univariable model) (13, 15), and perceived health status (5), with caregiving-related internet use (however, in (16), there was no relationship between self-rated health and health-related internet use). We do not confirm the role of formal education (5, 13, 15), caregiving duration (13, 17), or number of hours spent caring (15), even though we found the last to be associated with the frequency of internet use to find legal/fiscal information (r_s = 0.185). Contrary to the conclusions of other studies (14, 15) and to what we have previously hypothesized, we did not find an association of using internet resources with the degree of dependence of the person in care. Studies concluding for such relationship showed either more (14) or less (15) use of health-related internet resources by caregivers of severely dependent persons.

While it was reported that persons facing barriers in accessing conventional services are more likely to use health-related internet resources (36), we conclude otherwise, with caregivers reporting lifetime use of psychosocial services reporting more frequent use of caregiving-related internet resources. Moreover, Kim et al. have concluded that emotional stress is associated with health-related internet use (15, 16). We used an indicator of perceived psychological health, and we could not confirm an association of this with the overall use of caregiving-related internet resources. However, more frequent use of the internet to find information on dementia, or to find strategies to provide quality care, was seen among caregivers who perceive themselves in worse mental health. In referring to research on conventional support services, the relationship of caregivers' mental health with service use is equivocal and can go both ways (higher or lower use for worse mental health) (37).

While this study is attentive to research gaps, its findings must be seen in the context of study limitations. Since this study follows a cross-sectional design, we are not able to determine causality. For instance, we could not determine whether caregivers who appraise their physical health more favorably are more likely to seek web resources to manage the psychological effects of caregiving or whether using such resources holds a positive effect on perceived physical health.

Another caveat concerns the use of a convenience sample. As participants were recruited by advertising through community projects and caregivers' associations, they may be more likely to show help and information-seeking behaviors. Also, by studying internet use and resorting to an e-survey, one must account for the influential role of education on internet use (38). Highly educated participants comprised most of our sample, which differs from the typical characterization of Portuguese informal caregivers as low educated (23). Moreover, most of our sample comprises children of the PwD (younger caregivers). As such, the frequency of internet use may be higher in this sample as if it was appraised in a sample better representing less-educated individuals and older spousal caregivers. As individuals who can use the internet tend to be more educated, have higher income, and be younger (39), study findings must not be applied to the mainstream population of caregivers but rather to a segment of similar profile to this sample.

To date, no predictive model was proposed to explain the frequency of using dementia and caregiving-related internet resources. Given the exploratory nature of this study, we cannot suggest such a model either. Our findings should be seen as hypotheses-generating, especially in informing on enabling and need variables associated with using caregiving-related internet resources.

Future work would benefit from more complex sampling methods, longitudinal design, and even more comprehensive measurements of internet use for caregiving-related purposes. An examination of additional factors is encouraged in considering the small to moderate associations found for most variables. Future research must also shed light on caregivers' eHealth literacy, i.e., on the ability to locate, analyze, and act upon internet-based health information. Evidence is missing on whether internet resources used by caregivers are understood and useful. Moreover, whether caregivers search for information online merely through search engines or use apps to help with finding information, is an open question.

This study has offered insights into which frequency and for which purposes informal dementia caregivers seek dementia and caregiving-related information and support through the internet. We concluded for a moderate frequency of using the web to reach such resources. The internet is primarily used to find general information about dementia and, to a much less extent, to find support for caregivers themselves through formal services or sharing experiences with peers. This research also identifies predisposing, enabling, and need variables associated with both the overall and the purpose-specific frequency of using dementia and caregiving-related internet resources. We offer insights on the role of under-researched enabling and need variables, which provide contextual information and may allow an in-depth understanding of online support-seeking behaviors. Our research shows that in addition to being younger, not having a usual source of caregiving support, having a history of using face-to-face psychosocial interventions, being employed, and perceiving the own physical health more negatively, are associated with a more frequent use of dementia and caregiving-related internet resources.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by the Ethics Committee for Health of the São João University Hospital Center/Faculty of Medicine of the University of Porto (reference 208/18). The patients/participants provided their written informed consent to participate in this study.

ST: conceptualization, methodology, formal analysis, investigation, writing—original draft, funding acquisition, and responsible for ensuring that the descriptions are accurate and agreed by all authors. CP: conceptualization, methodology, writing—review and editing, and supervision. CC-S: methodology and formal analysis. AF: conceptualization, resources, writing—review and editing, and supervision. All authors contributed to the article and approved the submitted version.

This publication was supported by National Funds through the Portuguese Foundation for Science and Technology, I.P. (FCT), in the scope of the project UIDP/04255/2020 [A publicação deste trabalho foi financiada por fundos nacionais através da FCT Fundação para a Ciência e a Tecnologia, I.P., no âmbito do projeto/centro de investigação “CINTESIS—Financiamento Programático; UIDP/04255/2020”]. ST was individually funded by the Portuguese Foundation for Science and Technology (FCT) (D/BD/135496/2018). The funding source was not involved in the research.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.