Abstract
Background:
E-health initiatives on the Internet can be used to provide support to people with chronic diseases and to their caregivers. In 2014/2015, we created a free website called jesuisautonome.fr where older people, or their carers on their behalf, can assess their independence in daily living by filling out a simple questionnaire.
Objectives:
To evaluate the interest of the public in websites of this kind, by analysing home care plans obtained via the self-assessment questionnaire. We also describe patterns of use and visitor behaviour.
Method:
Over a period of 6 months, we analysed data from the website in terms of the basic characteristics of the user; the number of questionnaires completed; the main types of needs in terms of home support; and data from Google Analytics about the number of visitors, user behaviour and behaviour flow.
Results:
During the 6-month study period, 439 visitors to the site either viewed, part-completed or fully completed the questionnaire. A total of 190 users completed the questionnaire. Seventy-one per cent of the completed questionnaires were from family caregivers, and 29% were from senior citizens. The mean age of those receiving care was 78 ± 10.46 years. Their main needs were for domiciliary care (29.3%). Data from Google Analytics showed about 420 visits per month. Approximately 7.5% completed a questionnaire, approximately 5.3% downloaded a home care plan and there was a bounce rate of about 62%.
Conclusion:
First results from this website tend to endorse its use as a means of making practical solutions available to caregivers and older people.
Keywords
In France, about 20% of the population is over 60 years of age and 8.5% over 75 years. According to the latest estimates, the number of people over 60 years of age and living alone will increase to more than 10 million by the end of the century (INSEE, n.d.). Enabling older people to remain at home for as long as possible is a central policy across Europe and implies optimising the use of limited resources available for care (Wimo et al., 2011). Costs of both formal and informal care are an important issue because there is inevitably a greater prevalence of almost all diseases among the aging population (Hebert et al., 2013; Rosow et al., 2011; Wimo et al., 2011). Unpaid family or informal caregivers provide as much as 90% of the in-home, long-term care needed by older people (National Alliance for Caregiving and AARP, 2004). Lack of formal support for caregivers means that increased caregiver burden is becoming more prevalent (Adelman et al., 2014). Although the French national Alzheimer’s plan 2008/2012 (Social Sante France, 2013) has improved the availability of respite care and specialised personnel, the various proposals of the plan are often confusing for the public (Somme and de Stampa, 2011). The waiting time for a geriatric assessment is also quite long in France. General practitioners (GPs) are not well prepared for social assessment and are often overloaded with work (Gaboreau et al., 2013). This delay increases the physical and mental suffering of caregivers and exposes older people to increased risk from factors such as fraud, falls, malnutrition and medication errors.
The accessibility of a large amount of information on the Internet has changed the relationship between patients and practitioners (Cline and Haynes, 2001; Sadan, 2002). It is essential to healthcare practice to provide protection from misleading or inaccurate data posted for commercial or other reasons (Hesse et al., 2005; Pagliari et al., 2005). E-health, defined as the use of distance interventions based on emerging information and communication technology (Chiu and Eysenbach, 2010; Oh et al., 2005; Pagliari et al., 2005), provides support to people with chronic diseases and to their family caregivers. A great many patient education and caregiver support sites have been developed to this end. E-health offers an alternative way to support family caregivers, who are often overwhelmed with care activities (Aalbers et al., 2015; Bennett and Glasgow, 2009; Social Sante France, 2013). Its 24-h availability is of great assistance to caregivers, who are often still in regular employment. Importantly, senior citizens can also be reached with e-health programmes; several studies have underlined the rapid expansion of computer access among older people. For example, in the United States, 53% of those in the age group of 65 years and older have access to the Internet (Cline and Haynes, 2001; Wild et al., 2012). In France, nearly 50% of this age group has Internet access, with 80% of these people using the Internet on a daily basis (Andreassen et al., 2007).
In 2014/2015, we created a free website called jesuisautonome.fr, where older people and their caregivers were able to assess their independence in daily living by filling out a simple questionnaire. Website usability was initially evaluated during a geriatric memory consultation and focused on older people who needed support to continue living at home (Koskas et al., 2015). Families can become exhausted by the heavy burden of caring for their parents as they age, and the establishment of the necessary support can often be delayed. In addition, caregivers themselves sometimes lack real awareness of their parents’ situation, the result being that they do not seek help when it is needed (Gaboreau et al., 2013; Somme and de Stampa, 2011). The jesuisautonome.fr website was designed to make caregivers and older people aware of their need for assistance.
The first way to evaluate the success of an e-health service is to assess its actual use. There are several theories about non-use and about dropout rates or attrition (Eysenbach, 2005) for health services delivered by the Internet (Chiu and Eysenbach, 2010). Predisposing factors, including demographic characteristics, health beliefs and aspects of the theory of acceptance, are often considered to be major reasons for failure (Andersen, 1995; Chiu and Eysenbach, 2010).
The main purpose of this study was to evaluate the interest of the public in websites of this kind by analysing home care plans obtained via the self-assessment questionnaire; we also describe patterns of use and visitor behaviour.
Method
Website www.jesuisautonome.fr
An algorithm of the self-assessment questionnaire was created by practitioners working at the memory clinic of the Bretonneau Hospital (Koskas et al., 2015). It was then integrated into a website to provide easy access for older people and their families. The website offered two different approaches, depending on whether the user was a senior citizen or a caregiver. In the case of a senior citizen, prior to accessing the main questionnaire, there was a preliminary screening test of five questions on fragility. Only positive responses opened the corresponding part of the questionnaire. However, when the user was a caregiver, the entire questionnaire was available.
To develop the algorithm, we first defined a list of home care needs and services generally used by older people. Depending on the different responses to questionnaire items, the algorithm identified the needs of the person and the corresponding home care services the person required. We have demonstrated the correlation between individually designed care plans (Koskas et al., 2015) produced after a comprehensive geriatric assessment (Demiris et al., 2001) and plans based on the self-evaluation questionnaire completed via the website jesuisautonome.fr.
In addition, the website provided informative articles about common medical problems affecting older people, as well as articles on caregiver well-being, information on housing options to support older people and a list of home services that may improve their daily life. These elements were intended to create traffic on the website and to motivate people to complete the independent living questionnaire.
This website received no public or private funding and was not part of a specific health service, although its founders belong to the neurogeriatric network for Paris. The website complies with French regulations on data protection and, in particular, the storage of medical data.
The website has had free online access since September 2015, after an experimental phase between March and August 2015. The home care plan obtained after someone filled out the questionnaire became more readily downloadable from December 2015, due to technical improvements at the server. However, no major changes were made to the website or the algorithm during the study period.
In December 2015, an information campaign was directed at family GPs in three boroughs of Paris (the 8th, 9th and 17th). Using the official listing of family practitioners in these neighbourhoods, a letter with a flyer was sent to them by post. The French health system gives a key role to GPs in the management of patients. (A campaign aimed directly at patients seemed to us more difficult and a lot more expensive to set up. Therefore, an advertising campaign was not carried out, and we did not purchase any online advertisements such as Google AdWords.)
Between October 2015 and March 2016 (approximately 6 months following the website’s launch), we collected quantitative and qualitative data from the website jesuisautonome.fr and from Google Analytics.
Data from the www.jesuisautonome.fr website
Users had first to enter some anonymous preliminary information into the site in order to gain access to the independent living questionnaire, including age, gender, lifestyle (single or couple) and département (county/state). In compliance with French confidentiality rules, we could not ask caregivers for personal information.
From the website data collected, we carried out an analysis of the basic characteristics of the study population. We were interested in the number of questionnaires completed (following which the visitor received a home care plan), and we compared these data with the number of visits to the website (the proportion of visitors who merely visited the site as against those who completed the questionnaire).
In this group, we analysed: Information on socio-demographic characteristics, including age, marital status and position on the groupes iso-ressources (GIR) scale as a surrogate for physical disabilities. In France, the GIR scale is used to allocate individuals to one of six levels based on an overall assessment of disability (level 6 being the least disabled and level 1 the most; see Table 1). ‘Allocation Personnalisée d’Autonomie’ (APA) is a state benefit available in France to assist old people to continue to live in their own homes. It is based on their GIR score and is payable if the GIR is 4 or below (Microsoft Word – Rapport-evaluation-plan-alzheimer-2012.) (Somme and de Stampa, 2011). Average time to fill out the questionnaire. The main types of needs in terms of home service: domiciliary care, occupational or recreational activities, safety and medical equipment, transport assistance for people with disabilities, paramedical and help for family caregivers. We also assessed if there are any correlations between the demographics of the older people who downloaded a home care plan in comparison with older people who completed the questionnaire without downloading it.
Summary of GIR categories.
GIR: groupes iso-ressources; APA: Allocation Personnalisée d’Autonomie.
Data from Google Analytics
We used Google Analytics (Google, n.d.; Techwyse, 2010) to conduct an analysis of three aspects of the visits to the website:
Visitors to the website: Number of daily active users: the number of visitors who had at least one session on the website, and its progress during the 6-month study period. Their arrival patterns.
Behaviour of the user: New versus returning sessions: average percentage of first-time visitors (a good website will have a solid mix of new and returning visitors). Sessions: the period of time for which the user was actively engaged at the website. Page views in each session: the average number of pages viewed during a session.
Behaviour flow and page view analytics: site content (landing and exit page), unique page views, to eliminate the factor of multiple views of the same page within a single session (i.e. every time the page was viewed in an individual session as a single event), average time on page as the average length of visitors’ sessions (longer sessions indicated that users were more engaged), ‘bounce rate’ as the percentage of visits that were single page only (i.e. people who visited one page and left; typically a high bounce rate is a sign that people are leaving the website because they are not finding what they are looking for).
Results
Data from website
During the 6-month study period, 439 visitors to the site viewed, partly completed or fully completed the questionnaire; 249 questionnaires were started but not completed; and 190 users completed the questionnaire. Of these 190 users, 133 downloaded a home care plan (see Table 2). About 71% of the users were family caregivers, and 29% were senior citizens. The mean age of people in receipt of care for whom the questionnaire was completed and downloaded was 78 ± 10.46 years (66% female; 46.6% lived alone; and 53.4% lived with their spouse). The GIR score was 3.47 ± 1.2; 49 (36.8%) were at GIR 2 level, GIR 3 for 22 (16.5%), GIR 4 for 35 (26.3%), GIR 5 for 21 (15.8%) and GIR 6 for 6 (0.04%).
Data from Google Analytics and the jesuisautonome.fr website.
We found no specific correlation between the demographics of the older people and the likelihood of their downloading a care plan (see Table 3). Their main needs were for home care (29.3%), safety and medical equipment (18.6%), occupational and recreational activities (18.6%), transport assistance for the disabled (17.3%), paramedical (9.3%) and help for family caregivers (6.6%). The average time to fill out the questionnaire was about 7 min.
Correlations between the demographics of the elderly people and the likelihood that a care plan is downloaded.
GIR: groupes iso-ressources; NS: non significant.
aStudent’s t-test or χ 2, p significant < 0.05.
Data from Google Analytics
Table 2 shows that the number of visitors per month increased moderately during the 6-month study period, from 378 users or visitors and 385 sessions in October to 499 users and 599 sessions in March; and that 43.1% landed on the website’s home page, and about 20% on informational articles. Approximately 20% were new visitors. Conversion as ‘organic traffic’ (e.g. from search engines) averaged around 8% throughout. The referral traffic (coming from other websites) was too dependent on ‘Google Images’ and was not taken into account to avoid misleading results. User behaviour was characterised by low rates of pages per session (about 2.66) and page views, but a high rate of returning visitors (about 77%). The unique page views rate was about 74%. The bounce rate (percentage of sessions with viewing of only a single page or where the user left the site from the entrance page without having interacted with it) increased slowly from 58.18% in October to 68.95% in March.
Discussion
Our results underlined three main findings: (a) The majority of those who visited the website were family caregivers; (b) the proportion of users who completed a care plan was acceptable given that we conducted a limited information campaign in three districts of Paris and no paid advertising campaign on the Internet had been carried out; and (c) Google Analytics data provided information on user behaviour.
These results show that it is essentially the caregivers (71%) who used the website. There are several possible explanations for this. First, we think that family caregivers, and especially children, are more accustomed than older people to using the Internet (Hone et al., 2016; Kernisan et al., 2010). Second, studies have highlighted the frequent failure by older people to organise support at home, partly from failure to recognise their difficulties and partly because they prefer to receive aid from their family (Adelman et al., 2014; Rosow et al., 2011). Third, older adults with poorer health and more perceptual and cognitive deficits have greater difficulty using the website (Demiris et al., 2001). In designing a web system for use by older people, the user’s functional impairments and inexperience with computers need to be taken into account. Jesuisautonome.fr followed specific guidelines (Demiris et al., 2001) to maximise the number of potential visitors by using an accessible format for those with reduced sensory, motor and cognitive abilities and for older people in general. However, about 29% of users were older adults who filled out the questionnaire for themselves. Jesuisautonome.fr provided consumer information specifically related to healthcare, financial factors and the daily living needs of older people. Physicians are often unable to meet the needs of older patients and their families for this kind of information (Gaboreau et al., 2013; Sadan, 2002).
We found no difference in the characteristics of people downloading the care plan compared with those completing the questionnaire and not downloading the care plan. In France, a GIR score of 4 or less makes the subject eligible for financial aid. We noted that the mean GIR was less than four for these questionnaires, and we hypothesised that older people or their carers would naturally want to know if the recipient was eligible for this financial support.
The interactive nature of the website Jesuisautonome.fr provided tailored information to users with a focus on the quality of daily life, but we could not assess the quality criteria proposed in other studies (Demiris et al., 2001). However, we suggest that the average time taken to complete the independent living questionnaire would remain stable at about 7 min.
Our results suggested that the proportion of visitors who completed the questionnaire was 17.4%, and that 5.3% of users both completed the self-assessment questionnaire and downloaded the home care plan. The World Health Organization (WHO) recommended the use of self-administered questionnaires to evaluate one’s own health and functional status, and several studies have underlined the benefits of active participation on the part of the individual to improve the status of their health (Hone et al., 2016; Kernisan et al., 2010).
The main needs reported by those who completed the questionnaire were focused on improving quality of life in terms of home care (29.3%), safety and medical equipment (18.6%), and integration into society through the facilitation of travel with assistance for people with disabilities (17.3%), or requests for occupational and recreational activities (18.6%). The proposed home care plan addressed many aspects of daily life that are essential for senior citizens and their family caregivers, such as the presence of home help for meals (e.g. housekeeper with psychosocial training, ‘meals on wheels’ or home meal delivery), administrative assistance or remote home monitoring. We highlight the fact that the website also provided online support.
We noticed that users often wanted to know which institution was behind the site, although a presentation of the method, the team, disclosures and references was available on the website (Eysenbach et al., 2004).
During the study period, jesuisautonome.fr had about 420 visits per month. This relatively low figure must be qualified because no large-scale information or advertising campaign was carried out. We noticed that the proportion of website visitors fell from December to January, and then, there was a recovery in the number of visits. Probably the information campaign conducted among GPs increased the number of visits in December, and the period between late December and mid-January is traditionally particularly quiet in France. On the other hand, it is important to note that in over 420 monthly visits, approximately 5.3% resulted in the user downloading a home care plan.
About 79% of visits to the site were new sessions, with a page view of 1.24 min, and average time per session of 1.57 min. These figures remained stable during the follow-up. We surmised that a large number of users may have arrived at the website but did not stay because they did not find what they were looking for, resulting in a lower rate of pages, page views and unique page views. People usually arrived via a search engine but the reference of the website may have been inaccurate.
Data from Google Analytics and from the website (see Table 2) allowed a comparison between the number of visitors and the number of people who completed the questionnaire and obtained a home care plan, which constitutes the target of our website. Visitors to a website targeting adults caring for aging parents reported seeking both general information on caregiving and specific assistance with the complex aspects of caregiving (Andreassen et al., 2007; Sadan, 2002). Visitors can arrive at a health information website in many different ways, such as through a search engine or via a portal if they follow the link to an appealing-sounding article. Google Analytics data for the website confirmed this, with 20% of arrivals on the website being via articles on older people’s health or practical advice such as the layout of a bathroom, or information about remote home monitoring. However, we think that we need to engage in more targeted advertising to promote jesuisautonome.fr. Other websites for caregivers have exchange forums, where, for example, family caregivers often seek communication with and support from other caregivers.
The bounce rate was 62.42% during the study period. This rate is quite also commonly found on other sites (Techwyse, 2010). Generally, a bounce rate of between 56% and 70% is acceptable, but above 70% is a concern (Techwyse, 2010). The specific characteristics of the website may assist to explain this bounce rate (Kernisan et al., 2010). Unlike an online sales site, users aim to pick out specific health information. They remain on the site until they find a specific element (Cline and Haynes, 2001), but user demand in the field of health sites is characterised by high volatility in research (Atkinson et al., 2009; Cline and Haynes, 2001). This may also explain the high number of unique views. This bounce rate should be matched with the conversion rate on the home care plan of about 5%, which is the main aim of our website.
Our study had several limitations. We cannot be confident about the origin of the users as many identified themselves as caregivers and, in order to comply with French confidentiality rules, we could not ask them for personal information. The process by which users arrived at our website was complex (Atkinson et al., 2009; Eysenbach and Köhler, 2002), so it is difficult to know how these factors may have affected our sample. The study period was relatively short in regard to the different factors of attrition evoked by Eysenbach (2005). However, to the best of our knowledge, it is the first study of a website that focused on a home care plan produced through self-assessment.
Many people who provide unpaid care to an older person use the Internet as a resource for education and support. They look for specific advice on the practical aspects of managing the daily living needs of their parents with dementia or other frailties. Knowing what to expect and how to plan for caregiving emerged as the central need. Our work emphasises the importance of providing these caregivers with practical answers, which could quickly lead to an improvement in the quality of their lives.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.