Information needs and information seeking behaviour of dementia patients and their caregivers: a scoping review

Background Patients with dementia often require full-time caregivers especially in the later stages of their condition. Patients and caregivers’ access to reliable information on dementia is essential as it may have an important impact on patient care and quality of life. This study aims to provide an overview of the information needs and information seeking behaviour of dementia patients and their caregivers. Methods We conducted a scoping review of the literature and searched four electronic databases for eligible studies published up to August 2018. Two reviewers independently screened studies and extracted data. Information needs were classied according to emerging themes in the literature, and information seeking behaviour was categorized using Wilson’s model of information behaviour. Results Twenty studies with a total of 4,140 participants, were included in this review. Reported information needs focused on: (i) disease; (ii) patient care provision; (iii) healthcare services; and (iv) caregiver self-care. The most commonly reported information need was on healthcare service-related information. Characteristics found to inuence information needs were the severity of dementia as well as patient and caregiver status. Dementia patients and caregivers mainly displayed active searching information seeking behaviour and preferred using electronic sources to obtain health information. Conclusion Current dementia information sources available in English are extensive in the information they offer, but more emphasis needs to be placed on healthcare service-related information. All studies originated from high income countries and focused on information needs of caregiver only. The only variable found to signicantly inuence information needs was severity of dementia condition. The information needs identied in this review can be used to inform development and design of future dementia resources for patients and their caregivers.

Introduction known information (15). We de ned information seeking behaviour as "the purposive seeking of information as a consequence of a need to satisfy some goal", following the de nition by Wilson (16). We only included studies in English.
We excluded studies that included patients with certain diagnoses (i.e. AIDS dementia complex, primary progressive aphasia, Creutzfeldt-Jakob syndrome, primary progressive non-uent aphasia, Huntington disease, Kluver-Bucy syndrome), with comorbidities (e.g. dementia and Down's syndrome), as the information needs of these patients and their caregivers may differ from those with more common dementia syndromes. We excluded studies that included patients without a de nitive diagnosis of dementia, and caregivers of these patients. We excluded studies that explored the information needs of healthcare professionals as well as those focusing on general needs (such as service needs, psychological needs etc.) without mention of informational needs. We excluded studies that involved the evaluation of interventions for dementia patients and caregivers.
Data sources, collection, analysis A comprehensive search of the literature was conducted across the following databases: Medline (Ovid), Embase (Ovid), PsycINFO (Ebsco) and CINAHL (Ebsco). We searched the databases in August 2018 for relevant studies. The search terms within our search strategy included: information seeking behaviour, help seeking behaviour, needs assessment, health service needs, information needs, caregivers, spouse caregivers, family caregivers, dementia and Alzheimer's disease. We searched reference lists of related systematic reviews to ensure that relevant articles were not omitted. A search of grey literature, including various sites such as OpenGrey, Google and openDOAR was also performed (See Appendix 1).
The search results were imported into EndNote X8.2 (Thomson and Reuters 2018) to form a single combined library. After duplicates were removed, two reviewers independently screened the collated titles and abstracts. The full texts of potential studies were retrieved and screened for their eligibility.
Disagreements between the reviewers were resolved through discussion. The two reviewers independently extracted data using a data extraction form. Data for each study was extracted as follows: Study reference (author, year of publication, country where study was conducted, study design, study duration, study aims) Demographics of study population (number of participants, gender, age range, caregiver's relationship to patient with dementia, type/severity/duration of dementia, duration of caregiving, income and occupation, education level of caregiver, association with formal dementia care services)

Types of information seeking behaviour
Data measurement (data collection procedure, instrument used) Information needs identi ed Information sources (current, preferred sources of information) Other needs as speci ed by the patient and/or caregiver Inter-rater agreement between the two reviewers was calculated using Cohen's kappa, and good agreement was determined to be achieved (kappa = 0.80) between the two reviewers. Discrepancies in the extracted data were resolved through discussion between the two reviewers. A third reviewer acted as a mediator where differences could not be resolved. Study authors were contacted where information was missing or incomplete.
No quality assessment was performed on the studies in this review, as the main aim of this scoping review was not to emphasise methodological inadequacies within the studies but to present a broad scope of the existing literature on the information needs and information seeking behaviour of patients with dementia and their caregivers.

Data synthesis and analysis
We performed a thematic analysis and categorised the data according to the emerging themes in the included studies. This was done independently by two reviewers in parallel. Discrepancies in thematic analysis were discussed between the study authors. Categories were developed using a 'bottom-up' approach, and were re ned throughout the review process. Within each category of information need mentioned, we listed the reported information needs according to frequency mentioned. Information needs were sorted into four main themes: dementia-speci c information, patient care-related information, service information and caregiver self-care related information. We evaluated the frequency of information needs and other needs by calculating the percentage of studies that were included in each information need contributing to the main information theme. The percentages for each theme was calculated upon 100%.
We sought to evaluate if variables such as sociodemographic characteristics of the population and dementia severity in uenced information needs and information seeking behaviour and to what extent.
We identi ed information seeking behaviour in the studies through patients' and/or caregivers' information sources, and their actions as speci ed in the article. For example, if a caregiver stated they often turned to their family doctor for more information on dementia, we would classify that as 'active searching' behaviour. Information seeking behaviour was grouped according to the four categories speci ed in Wilson's model of information behaviour: active searching, passive searching, ongoing searching and passive attention (17). The four categories were de ned as follows: Active searching was de ned as the individual actively seeking out information. Passive searching was de ned as "the acquisition of information that happens to be relevant to the individual". Ongoing searching is "where active searching has already established the basic framework of knowledge, but where occasional continuing search is carried out to update or expand one's framework". Passive attention is where "information acquisition may take place without intentional seeking" (18). We selected Wilson's model over other models of information behaviour as it offered a broad summary of the context of the information need of the seeker, the type of searching employed and the information sources utilised, and was not focused on the information seeking process. For current and preferred sources of information, the sources mentioned by studies in the literature were listed according to the frequency mentioned.
Our scoping review aimed to provide an outline of the information needs of patients with dementia and their informal caregivers, and we did not conduct further statistical analyses to investigate the relationship between sociodemographic variables and information needs or preferences.

Results
We identi ed a total of 2,447 abstracts in our preliminary search. After duplicates were removed, we screened 1,848 articles and eventually included 20 studies in this scoping study ( Figure 1). After screening for titles and abstracts, we retrieved the full texts for 38 articles that explored the information needs or information seeking behaviour of patients with dementia and their caregivers. From these 38 articles, we excluded seven studies that did not meet our inclusion criteria. 20 studies from 21 reports, were eventually included in this review ( Figure 1).

Figure 1: Study ow diagram
The 20 studies comprised of ten cross-sectional studies, eight qualitative studies, and two mixed methods studies with 4,140 participants (Appendix 2). One of the cross-sectional studies was reported twice. All studies were presented in English and published between 2002 and 2018. Four studies examined both caregiver and patient perspectives while the rest examined only the caregiver's. With the exception of four studies that did not report caregiver-patient relationship, the majority of the caregivers were the spouses or children of the patient. All studies were from high income countries, with eight studies from the USA. Sample sizes ranged from 9 to 1181, apart from one study where sample size could not be determined due to study methodology. Majority of caregivers were female and had mean ages between 55 to 70 years. Mean ages of the patients was between 70 to 90 years. Education level of participants was stated in eleven studies, where most of the participants had completed at least high school education.
Of the types of dementia speci ed in this review, 35% of studies included dementia (19)(20)(21)(22)(23)(24)(25), 20% of studies included LBD (22,(25)(26)(27), 15% of studies included AD and other non-speci c forms of dementia (19,22,25). However, 45% of studies did not specify the type of dementia (28-36). All 20 studies identi ed information needs irrespective of dementia types, however information seeking behaviour was only reported in 11 studies. Four studies further illustrated barriers to information needs and information seeking behaviour which included: lack of access, time, energy, knowledge to interpret or search for information and denial of the condition (19-21, 28)

Information needs
We identi ed four main themes in relation to information needs, with a total of 39 reported information needs. The four information themes were diseasespeci c information, healthcare service-related information, patient care provision, and caregiver self-care. Information related to healthcare services was the most commonly identi ed theme, while patient care provision represented the category with the most diverse number of information needs.

Disease-speci c information
The three most commonly mentioned needs related to dementia were general information on dementia (10 studies, 62.5%), dementia treatment (7 studies, 43.7%) and identi cation and understanding of dementia symptoms and behaviour (6 studies, 37.5%). Other information needs within this category also included dementia prognosis (6 studies, 37.5%), current dementia medication (5 studies, 31.2%), experimental drugs and clinical trials (4 studies, 25%), current research on dementia (3 studies, 18.7%) and genetic aspects of dementia (2 studies, 12.5%). The following information needs were mentioned only once (0.06%): negative impact of dementia on family and community, information speci c to different stages of dementia, chance of recovery and memory skills ( Table 2).
The need for information on "negative impact of dementia on family and community" was de ned by Hirakawa as the physical and psychosocial burden on family and community (21). It was stated that this would help the caregivers prepare for and cope with the behavioural and psychological symptoms of dementia, as these symptoms worsen through the progression of the disease.

Patient Care Provision
Patient care provision was the category with the largest number of reported information needs, comprising 14 out of the 39 unique information needs (35.8%) identi ed in this review. Seven studies (53.8%) within this category indicated a need for information on how to provide general care. General care included patient hygiene, food and nutritional information and best attitudes to adopt in caring. Ways to deal with the patient's behaviour (6 studies, 46.1%) was the second most commonly mentioned information need, followed by safety issues (5 studies, 38.4%). Safety issues encompassed information on how to improve the safety of the patient's home environment, how to keep patient safe and how to recognise fall risks and poor mobility. Other reported information needs in this category were coping on with patient hallucinations (2 studies, 15.3%), navigation of communication di culties (2 studies, 15.3%), activities for the patient (2 studies, 15.3%), rst aid and medical information (2 studies, 15.3%). Additional seven information needs were retrieved from only one study and these were: con ict resolution, patient ethics, helpful experiences of other caregivers, when to transfer the patient to the hospital, dealing with family and friends, advocating for the patient, and emergency situations.
Healthcare Service-related The top information need within this category was where and how to use services, and available help (14 studies, 93.3%). This was also the information need that was reported by the largest number of studies, among all identi ed needs. Services and help referred to facilities (e.g. geriatric hospitals and nursing homes), healthcare professionals skilled in dementia diagnosis and treatment, or support groups for caregivers. Other service-related information needs were nancial help and services (5 studies, 33.3%), legal issues (4 studies, 26.6%), application for care programs (2 studies, 13.3%), and insurance (2 studies, 13.3%). As with the other main information themes, there were several information needs that were only stated once (0.07%). For service-related information, this included home help, when to transfer a patient to the hospital, and transportation options.
Caregiver self-care Self-care for the caregiver was only mentioned by four studies. Two studies identi ed information needs for stress management (50%), while carer's pension entitlements, managing caregiver's emotions, and general caregiver self-care such as exercise, diet and medications were identi ed by one study each (25%).
This information theme had the smallest number of contributing studies and reported information needs.

Variables affecting information needs
The severity of dementia and patient/caregiver status were the only variables found to in uence information needs in this review.

Information needs at various stages of dementia
Caregivers of patients with mild dementia were more likely to look for disease-speci c information, as compared to caregivers of patients with moderate to severe dementia. Both groups of caregivers required general information on dementia, identifying and understanding dementia and current dementia medication. However, those caring for patients with mild dementia also needed information on genetic aspects of the disease, experimental drugs and current research on dementia while those caring for patients with advanced dementia were primarily concerned about dementia prognosis ( Table 3).
The severity of the patient's condition did not affect the number of patient care provision-related information needs. While both groups of caregivers had equal numbers of patient care information needs, the subcategories of information requested did not overlap. Caregivers of patients with mild dementia needed information on how to care for the patient, rst aid and when to transfer a patient to the hospital. Caregivers of patients with moderate to severe dementia desired medical information, information on how to deal with patient's behaviour and safety issues.
Carers of patients with moderate to severe dementia also needed more information on healthcare services. Both groups of carers required information on where and how to use available services, and how to apply for daycare programs. Caregivers of patients in more advanced stages of dementia indicated an additional need for information on nancial help and services.

Patient versus caregiver information needs
Three studies investigated the needs of patient groups alongside their caregivers (25,27,38). However, two of these studies did not segregate the information needs of patients from caregivers, therefore we were unable to present any differences in information needs for these particular studies (25,27).
The remaining study, that did segregate the information needs of patients and caregiver, showed that patients and caregivers had a large overlap in 80% of their information needs (38). The topics of shared interest included identifying and understanding dementia, experimental drugs and clinical trials, current dementia medication, communication di culties and how to manage them, how to deal with patient's behaviour, and patient activities. These topics were equally distributed between disease-speci c and patient care provision themes. The topics of interest speci c to the caregivers were genetic aspects of the disease, and how to deal with family and friends. On the other hand, the topics of interest unique to patients were clinical trials and where to nd support groups. Overall, caregivers were more interested in topics speci c to dementia, while patients were more concerned about service-related topics.
Information seeking behaviour was mainly surmised from current information sources stated within the studies. These sources for patients and/or their caregivers were stated in 14 studies (19-22, 25, 27, 28, 30-33, 35, 36, 38). Of these 14 studies, six studies also mentioned desired information sources (22,27,28,30,33,35). Two studies mentioned the preferred sources of caregivers without reference to their current information sources (26, 34). The most utilised source of information was the Internet (e.g. websites and forums) (10 studies, 71.4%), followed by healthcare professionals (7 studies, 50%), and family and friends (6 studies, 42.8%). Service employees (bank employees, clergy) were also approached for information, however this was the least common source of information as it was mentioned in one study (20,36). Other sources of information utilised include the Alzheimer's Association, written information sources (eg. magazines, lea ets and books), electronic sources (e.g. email, smartphone apps, videos), support groups, care centres, social workers, other family caregivers, and volunteer groups (Table 4).

Figure 2. Percentage of Studies indicating Current and Preferred Information Sources
The Internet was indicated as both the most highly utilised source of information and the most desired source of information for patients and caregivers. Most of the participants used Internet by browsing websites and forums to look for information. However, what patients and caregivers desired when using the Internet were more speci c online resources, such as a website with frequently asked questions, how-to videos for patient care and dementia and servicerelated information available in online portals (22,30,35). Unsurprisingly, other electronic sources such as mass media, smartphones, audio-visual materials and DVDs ranked equally as high as the Internet, as a preferred source of information (26, 28, 33,35). The electronic sources requested were not dissimilar to the current electronic sources already utilised by patients and their informal caregivers. Patients and caregivers showed equal preference for written information sources and both online and face-to-face support groups (50%), despite those sources not being one of the top three current information sources (26, 27,34,35). Electronic medical records, educational conferences, government services and healthcare professionals were also reported as desired information sources, in one study each (Table 4, Figure 2).
In addition to the abovementioned sources, one study also stated additional preferences on how information should be presented (35). Having information in multiple sources and languages was preferred, in order to increase its accessibility to linguistically diverse communities. Information presented through these sources should also be individualised to the different dementia syndromes and stages of the disease, as the concerns of these patients and caregivers could vary widely.

Discussion
We identified 20 studies (4,140 participants) reporting the information needs and information seeking behaviour of dementia patients and their caregivers. Studies mostly focused on the information needs and information seeking behaviour of dementia caregivers, were conducted in high income countries and involved predominantly female caregivers. Dementia health services related information was the most commonly reported information need in the included studies. The most diverse information needs related to patient care. Severity of dementia and patient/caregiver status were the only variables observed to affect information needs of dementia caregivers and patients. Information seeking behaviour was reported in 14 studies, and active searching behaviour was the most commonly reported type of information behaviour.
Currently, there is a lack of evidence focused on the information needs and information seeking of dementia patients, and patients with different types of dementia. More research is also required for populations in low-and middle-income countries, and for comparison of the impact of different variables on information needs and information seeking behaviour.
The results of this current review suggest that the type of information most often requested by dementia patients and their caregivers was where to find and use available dementia-focused healthcare services (14 out of 20 studies; Table 2). Surprisingly, this contradicted most of the available literature that we have seen. Systematic reviews exploring the needs of informal dementia care providers indicated that the most frequently reported information need was information on the disease, followed by patient care information (11,40). There could be several reasons for this inconsistency. We included several studies that were published after the dates of these reviews. As such, studies that were more recently published could have implied there was a shift in the focus of information needs away from the disease and towards available help and services. At the point of diagnosis, the patients and their caregivers could have been counselled on the condition to help them understand and accept the condition. However, information about formal care services may not always be adequately provided at the point of diagnosis, and this information becomes more pertinent as the patient's condition declines (41). As seen in our results, the severity of the patient's dementia condition also affects the type of information required by their caregivers and the patients themselves. While we are unable to establish if this was a definite contributor to the difference in information needs as both reviews did not provide the stages of dementia for their included studies, this could have also affected the information needs reported as caregivers usually need more help in the later stages of dementia, following the worsening of behavioural and psychological symptoms of dementia. Another possible reason would be that the focus of these systematic reviews was not on information needs, but rather on general needs, and thus did not include some of the studies and information needs that were mentioned in the current review.
In order to better understand the information preferences of dementia patients and their informal caregivers, we also evaluated their information seeking behaviour and information sources utilised. It was reported that the most desired source of information was the one that patients and caregivers most frequently utilised (i.e. the Internet). This suggests that their information needs were unmet while using the Internet, as a result of the inability of these groups to find the information they need or the inadequacy of information available on specific websites or portals that were accessed (42,43). Our results seem to suggest that the latter was a more likely reason, as participants indicated specific resources they wished to find on the Internet, such as videos and frequently asked questions (FAQs) websites (see Table 4). The sentiment on being unable to find the specific types of information they were looking for across various information sources has also been echoed in existing literature (11). The inability to find specific information may not be due to a lack of information, but in certain cases it could also be presented in a form that was difficult to understand by patients and their caregivers. Information on the disease could have been presented using medical terms that people were not familiar with. Often, the amount of information available could be overwhelming for patients and their care providers to sort through, leading them to miss what they were looking for even if it was readily available. Despite only one study in our review emphasising the need for targeted and timely information to dementia patients and their caregivers, this need is not unique based on types and severity of dementia. This was also previously iterated in two reviews (11,44). Future interventions focusing on information provision should take these into account when developing information resources.
An interesting finding from our review was that the top sources of information that dementia patients and caregivers preferred were electronic sources such as the Internet, mass media and smartphones. This was unexpected given that the mean ages of caregivers were between 55 to 70 years, and between 70 to 90 years for the patients. Existing literature to date suggests that electronic sources were not preferred as health information sources due to a lack of awareness about what these sources could offer, rather than cost or access-related reasons (45). As internet adoption among older adults have steadily increased over the years, the purposes for which they used the Internet for could also have changed. In 2003, the use of the Internet in older adults were mainly confined to email and general information seeking but more recent analyses in 2009 show that in addition to those purposes, health information seeking was placed as one of the more common online activities (46). Electronic sources are able to offer a variety of information independent of time and location, which could be the reason for their growing popularity as a health information source. This growing trend of information retrieval from electronic sources is not limited to the Internet but could also encompass apps for dementia care on smartphones and programmes on mass media sources (47,48). Our results also identified active searching behaviour as the most common information searching behaviour amongst all included studies, highlighting the growing acceptance of new information sources (such as smartphones and the Internet) that were previously not used by dementia patients and caregivers.
Patients and caregivers also showed a preference for written information and support groups as information sources over healthcare professionals. Healthcare professionals were usually the first information source that patients and caregivers would encounter, with information about the condition being provided at point of diagnosis. However, several studies within our review expressed that the information provided was insufficient and that further information was also not provided at subsequent follow up appointments (22,28,31,34). The information provided by healthcare professionals could be of a medical nature (such as medications to manage the condition), while patients and caregivers may desire information regarding the prognosis of the condition or available community care resources and entitlements. This finding again highlights the mismatch between information needs and information sources, where healthcare professionals may not necessarily be unable to provide the information required but could be unaware of these other information needs of the patients and caregivers. Support groups and written information sources could also be more highly preferred due to their ability to provide an array of information beyond medical information.

Limitations and future research directions
Our review identified several limitations. We were unable to ascertain the severity of dementia for the patients, and if the participants included in some studies were truly the primary caregivers for dementia patients, as some surveys were returned through the post or online.
While we did not set restrictions on the geographical regions of where studies were conducted, all studies were conducted in high income countries such as the USA and UK, therefore the findings may not be applicable to other geographical regions or even rural areas within these countries. Furthermore, the search strategy was limited to studies published in English and we may have missed relevant articles that were published in other languages. Most of the studies included addressed mainly the information needs and information seeking behaviour of the caregivers, and only four studies addressed the needs of both patients and caregivers. Some studies recruited participants through selfselection, therefore they may already be more highly educated, have higher income levels and increased informational needs and already be more active in searching for information.
Our scoping review highlights that the mismatch between information needs and information sources is not due to access, but rather the inability to find specific information across various sources. Rather than finding new ways and platforms to deliver information to patients and their caregivers, which would result in further fragmentation of available information, effort should be made to consolidate information on specific sites or platforms that are already in use (e.g. Alzheimer's Association website). Future interventions targeted at information provision should focus on utilising familiar platforms to deliver information pertaining to healthcare services while ensuring that the information is available in different languages.
Most of the studies included in the review used mainly cross-sectional (21, 23-27, 29, 32, 36, 38, 49) or qualitative study designs (20,22,28,30,(33)(34)(35)37). Future research should consider the use of longitudinal study designs, in order to further identify information needs of patients and caregivers at various stages of their disease -from diagnosis where disease related information may be more desired, to the more advanced stages where help may be sought on providing adequate care to the patient. With the advances in technologies, future research should explore how information can be delivered in a targeted manner based on different types of digital technologies, such as through smartphones. The use of mobile phone technologies for health professions education has seen great effectiveness and is also a cost-effective and accessible alternative to printed information sources (50). Since there is already an indicated preference for electronic sources, this approach could similarly be adapted to deliver information to dementia patients and caregivers.

Conclusions
This scoping review provides an overview of the literature on the information needs and information seeking behaviour of dementia patients and caregivers. Information needs of patients and caregivers can be categorised into four themes -disease related information, healthcare services-related information, patient care provision and caregiver self-care. Information on healthcare services need to be emphasised by healthcare professionals at consultations and through preferred electronic and written sources. Caregiver self-care was also another information need that was frequently overlooked, and future research should focus on evaluation of this information need. The only variable found to significantly influence information needs was the severity of dementia. Information needs between caregivers and patients were found to be largely consistent, however, more research is required on this area as this was only highlighted by one study. Patients and caregivers actively sought for information and preferred using the Internet to look for desired information. Future interventions on information delivery to dementia patients and caregivers should focus on having required information on central platforms such as the Alzheimer's Association website and emphasise the use of preferred electronic sources. Future research should aim to evaluate the information needs of patients and caregivers at specific stages, across various dementia syndromes and focus on variables that affect these information needs.

List of abbreviations
Alzheimer's disease = AD, frontotemporal dementia = FTD, Lewy Body dementia = LBD, Electronic medical records = EMR, digital versatile disc = DVD, frequently asked questions = FAQs, General practitioner = GP, European Union = EU Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.

Availability of data and material
This scoping review included the data extracted from the primary studies. The whole set of data extraction sheet is available upon request.

Competing interests
The authors declare that they have no competing interests.

Funding
Lee Kong Chian School of Medicine, Nanyang Technological University Singapore, Singapore

Start-Up Grant
Authors' contributions LTC conceived the idea for the project. AS and AST wrote the paper. LTC provided the methodological guidance and critical inputs for the review. LTC, AST, BMK and TYL reviewed the article and provided the comments.  Negative impact of dementia on family and community 1 (0.06%) Information specific to different stages of dementia 1 (0.06%) Chance of recovery 1 (0.06%) 1 Percentage of studies refer to the percentage of studies in each reported information need contributing towards the main theme.