We reviewed 2,605 titles and abstracts from the electronic search, and identified an additional 16 studies from reviewing reference lists and consulting technical experts. After applying inclusion/exclusion criteria at the abstract level, 388 full-text articles were reviewed, as shown in Figure 1. Of the full-text articles, we rejected 331 that did not meet our inclusion criteria.

Figure 1. Literature Flow – Mobile Applications for Caregivers.

Figure 1

Literature Flow – Mobile Applications for Caregivers.

We included 31 publications reporting on 22 CHIT interventions that were being developed, piloted, or evaluated for their effects on caregiver outcomes, patient outcomes, healthcare utilization, or process measures.1443 Of these, there were five RCTs in seven publications;16,19,20,23,26,27,35 the remaining studies consisted of feasibility studies, usability studies, pilot tests, qualitative studies, and quasi-experimental studies. We included 26 publications reporting on 22 interventions for caregivers of children with chronic illness or disability.4469 Twelve pediatric studies were RCTs.4649,55,57,59,6365,67,70

Studies on caregivers of adults and children with chronic illness or disability are shown in Tables 1 and 2, respectively. Table 3 shows the applications and functionality included in each intervention, and the phase of the interventions’ lifecycle at the time of the study. Although many interventions contained non-interactive educational and resource materials, Table 3 also lists all interactive components of the intervention according to the classification taxonomy outlined previously.

Table 1. Characteristics and findings of studies of consumer health information technology interventions to support non-professional caregivers of adults with chronic illness or disability, stratified by care-recipient population.

Table 1

Characteristics and findings of studies of consumer health information technology interventions to support non-professional caregivers of adults with chronic illness or disability, stratified by care-recipient population.

Table 2. Characteristics and findings of studies of consumer health information technology interventions to support non-professional caregivers of children with chronic illness or disability.

Table 2

Characteristics and findings of studies of consumer health information technology interventions to support non-professional caregivers of children with chronic illness or disability.

Table 3. Summary of the functionalities available in health informatics interventions to support non-professional caregivers, stratified by care-recipient population.

Table 3

Summary of the functionalities available in health informatics interventions to support non-professional caregivers, stratified by care-recipient population.

KEY QUESTION #1. How does the use of consumer health information technologies (CHIT) by non-professional caregivers of adult patients with chronic illnesses or disability, or by such patients who rely on a non-professional caregiver, affect outcomes for caregivers, patients, clinical process measures, and healthcare utilization?

Summary of Findings

The majority of articles described interventions that provided education and one of several communication modalities: online peer support groups, online access to providers through email, or general disease information and education. Users expressed a desire for more social support and studies consistently found that the online peer support groups and chat-rooms were both the most-used and most-valued components of any given website, application, or intervention. The asynchronous nature of these online communications facilitates participation in support groups by mitigating some of the barriers of travel time, geographic separation, and competing priorities. Online communications provide access to peers and clinicians that would otherwise not be available in many communities, particularly smaller town and rural environs. Anonymity was often perceived by users as an important feature of online support groups.

Several studies described how technical barriers or lack of familiarity with technology could limit accessibility of the intervention. Despite the numerous potential technical barriers, few studies reported the amount of technical assistance and training provided to users. Older caregivers may be less likely to benefit from mobile applications because they are more likely to be non-users of handheld technology. Older caregivers may therefore require training in the use of the device or application, and may also benefit from applications with special accommodations for aging vision and manual dexterity, and their own chronic illness burden. Accommodations for language preference may enhance the utility of mobile applications for immigrant caregivers.

The characteristics and findings on interventions for caregivers of adult patients are shown in Table 1.

Detailed findings

Our literature search identified 31 articles describing 22 CHIT interventions for caregivers supporting patients with a myriad of medical conditions. Interventions for caregivers of patients with Alzheimer’s disease or dementia made up the largest group of studies (7 interventions published in 10 reports, of which 2 of these were RCTs examining intervention effects on caregiver outcomes).1421,71,72 The needs of caregivers for patients after cerebrovascular accident were targeted in six articles (1 intervention published in 1 RCT, 1 secondary-analysis, 1 observational study, and 3 feasibility studies);2631 caregivers of cancer patients in four articles (an RCT with lung cancer patient-caregiver dyads, and observational studies of pancreatic, prostate, and cancer in general);2225 caregivers of the frail elderly in four articles regarding one intervention,3941,73 with an additional one article addressing adults with “disabilities” that was not otherwise specified. 42 Two articles each addressed schizophrenia.33,35 Two articles addressed traumatic brain injury.34,36 The needs of caregivers for patients who were diagnosed with depression, receiving home-based parenteral nutrition, post cardiac transplant, or currently undergoing surgery were addressed in one article each.32,37,38,43 Only one intervention was specifically evaluated in a VA population. 33 Details of the studies are found in Table 1.

The most commonly included applications and functionalities included education modules (12 studies); self-care aids (9 studies); peer-to-peer communication tools (13 studies); tools designed to improve caregiver-to-healthcare provider communication (11 studies). Only two studies examined email or text message reminder systems, two studies included tools to facilitate patient-caregiver communication, and no studies evaluated transactional tools such as online medication refills or appointment requests (Table 3).

In this section, we highlight studies that evaluated the effects on caregiver outcomes or evaluated specific populations relevant to the VA. We subsequently describe the common characteristics and themes that emerged across studies and disease categories.

Caregiver outcomes: Dementia population

One small RCT found that ComputerLink, a computer-based support network, reduced strain among caregivers.20 ComputerLink provided a four-module electronic encyclopedia on Alzheimer’s disease, but the more widely used feature was a communication component that included a public bulletin board, private mail, a nurse-facilitated Q&A segment, and private email communication with the nurse for clinical expertise. The study investigators installed computer equipment in the participants’ homes and provided a 90-minute training session. After 12 months, ComputerLink was associated with reduced relationship strain among spousal caregivers compared with controls. Reduced strain was also observed among those caregivers who were initially more stressed and used ComputerLink communication features more frequently. Although the amount of care received did not differ between the treatment and control groups at the end of the study, caregivers perceived benefit from being able to communicate with peers and professionals at times of day that were convenient.19

AlzOnline, an Internet- and telephone-based education and support network, was associated with increases in perceived self-efficacy and reduced caregiver burden in a pre-/post-test study of 21 caregivers.71 AlzOnline includes an Internet library, message board, expert forum, and a series of six live web- and telephone-based interactive classes on positive caregiving using cognitive-behavioral techniques. Caregivers who owned computer equipment that met the technical requirements for the interactive classes were eligible to participate. The average time to complete the series of classes was 16 weeks. Significant improvements were found on all three subscales of the Caregiver Self-Efficacy Scale (CSES), and the subjective emotional burden scale of the Caregiver Appraisal Inventory (CAI). No significant changes were found on the positive aspects or time burden subscales of the CAI, or on the Stress-Related Growth Scale (SRG). The study authors suggest that these differential findings may in part reflect the content of the intervention. The Positive Caregiving series of classes focused primarily on increasing skills in relaxation and the management of challenging caregiving situations, as well as enhancing perceptions of caregiver mastery and self-efficacy, with little emphasis on the rewards and emotional benefits of caregiving.71

Caregiver’s Friend, a worksite-based support program, reduced caregiver strain in an RCT of 299 family caregivers who were also employed outside the home.16 The Caregiver’s Friend intervention was an Internet-based multimedia program that provided tailored text material and videos that modeled positive caregiving and coping strategies in a variety of scenarios. Individual tailoring of the online content was achievable via an interactive questionnaire that led to a menu of links based on the patient’s level of dementia and the caregiver’s specific concerns. Caregivers in the treatment group (n=150) were given access to Caregiver’s Friend for 30 days. Pre-test and 30-day assessments were compared between caregivers in the treatment group and a waitlist control group (n=149). Compared with controls, caregivers in the treatment group reported significant improvements in self-efficacy; intention to get support; and caregiver gain, stress, strain, depression, and anxiety. Average exposure to the program was 32 minutes, and composite gain score was significantly correlated with time spent in the program (r=30, p<0.001).16

Caregiver outcomes: Lung cancer population

The Comprehensive Health Enhancement Support System (CHESS) program for patients with lung cancer and their caregivers improved caregiver coping strategies as measured in a two-year RCT of 285 patient-caregiver dyads.23 The control group (N=141) received usual care, a laptop computer with Internet access, and a list of patient-oriented lung cancer and palliative care websites that the researchers felt to be of high-quality. The intervention group (N=144) was provided laptops and Internet connectivity, as well as access to the CHESS website that provides formal online discussion groups, communication with expert care, educational materials, FAQs, caregiver tips, health status tools, and decision aid tools, among others. Increased bonding among caregivers – suspected to be a result of the communication tools and online discussion groups – was found among CHESS users compared to controls (p<0.05). The increased bonding appeared to mediate improvements in both appraisal-focused coping (“cognitive efforts to define and redefine the personal meaning of the stressful situation,” p<0.05) and problem-focused coping (“to modify or eliminate stressors by handling the reality of the demands,” p<0.05) strategies as measured by a previously validated scoring tool called “Brief Cope.” The mostly female (68%) caregiver users had a mean age of 55.6 years, and 50 percent had completed at least a college degree. The amount or frequency of use was not reported for the two-year study. Interestingly, only 42 percent of dyads completed the final survey. However, because the investigators were studying a population with advanced lung cancer, 84 percent of study dropouts were due to the death of the patient; thus, the high attrition rate does not appear to be indicative of use or usability obstacles.

Caregiver outcomes: Post-surgical population

A two-group prospective, pre-/post-test intervention study of post-heart transplant patients and their caregivers demonstrated improved caregiver outcomes in those who used an Internet-based website for four months.38 The intervention group comprised 24 heart transplant recipients and their family caregivers with access to a website of educational resources and psychosocial interventions such as: online, self-paced workshops on post-transplant stress management and medication management (both of which included self-assessment tools, decision support, and tailored advice), clinician-moderated online discussion groups, email contact with transplant team members, a Q&A library, healthy living tips, and a resource library. The control group consisted of 40 heart transplant recipients and their caregivers who were enrolled in other non-Internet-based interventions at the same institution over a four-month period either immediately before or immediately after the present study. Mental health (depression, anxiety, and anger-hostility) was measured using the Symptom Checklist-90 (SCL-90), which assesses mental health over the past two weeks. Caregivers in the intervention group had significant improvements in both their anxiety and anger-hostility scores from pre-intervention to post-intervention compared to controls (p=0.05 and p=0.03, respectively). Caregiver depression scores also improved over time but were not significantly different between intervention and control groups. Health-related quality of life, as measured by the Short Form-36 (SF-36), improved for both patients and caregivers in the intervention group over time, but was not evaluated in controls. Finally, a subgroup analysis found that those patient/caregiver dyads in the intervention group who used a particular component of the website, an online workshop about medication management, had increased compliance as measured by frequency of completing recommended appointments and lab work. Similarly to other studies, the majority of caregivers were female (85%), married (85%), and Caucasian (90%); 50 percent had at least a high school education.

Caregiver outcomes: Veterans

Only one intervention was specifically evaluated in a VA population.33 This was a prospective trial of a web-based intervention for relatives of Veterans with schizophrenia or schizoaffective disorder. It provided educational and support services to relatives, and included an online professionally-facilitated multi-family support group. The 26 family members in the intervention were compared to an historical control group of 16 family members who had received other support services in the four years prior to this study. They found that there was no difference between intervention and control groups in family members’ measurement of distress using the Brief Symptom Inventory scoring tool. A measurement of stress between the family member and patient did improve over time in the intervention group, but was not measured in the control group. A trend towards decreased hospitalizations among care-recipients was noted in the intervention group compared to controls but was not significant (24% v 50%, X2 2.93, p<0.09). Family members were predominantly female, with a mean age of 57 years old and an average of 14 years of education. Of note, they were not specifically caregivers for the VA patient. Seventy-nine percent of family members completed the year-long program, 85 percent also used the online discussion boards, and there was an average attendance of 59 percent at the weekly online family support group sessions.

Effects on healthcare utilization: Stroke population

The Caring~Web studies were notable for evaluating healthcare utilization by care-recipients. In a three-month descriptive study of five caregivers of patients who had recently been discharged to home from a rehabilitation treatment center after a stroke, investigators measured the number of phone-calls to a clinician’s office (range 1–4), appointments (range 2–3), emergency department visits (1), and hospital admissions (1).31 In a subsequent evaluation, a one-year RCT demonstrated that self-reported, utilization of healthcare services was less compared to controls.26 The intervention group consisted of caregivers of patients with a recent stroke (n=36) who were provided access to the Caring~Web application, a website with links to educational information on strokes and caregiving, tips for caregivers, an email link to solicit expert care from the institution’s nurse specialist and a multi-disciplinary stroke rehabilitation team (consisting of a physician, pharmacist, dietician, social worker, and therapists), and an email discussion group facilitated by the nurse specialist. The control group comprised caregivers (n=37) that would not be using the web. All recruited caregivers were Internet novices. After one year, the intervention group’s care-recipients had 33 percent less emergency department visits (p=0.001), and 66 percent less hospital admissions (p=0.0005) compared to the patients of caregivers in the control group. Fewer clinician provider outpatient visits were noted at three months (156 vs 188 visits), but this had become non-significant at one year of follow-up. These findings suggest that care-recipients in the intervention group received more appropriate or timely care, thus preventing emergency department visits and subsequent hospital admissions, perhaps due to their caregivers having improved access to information or closer communication with clinician providers as a result of the Caring~Web intervention. Again, caregivers were primarily female (75%), the spouses of the care-receiver (70%), with a mean age of 55 years, and ~90 percent had achieved a high school education or greater.

Common CHIT intervention characteristics among studies

Regarding the types of caregiver needs that were addressed, the majority of articles described interventions that provided one of several communication modalities: either online peer support groups, online access to providers through email or SMS/texting, or general disease information and education. The majority of interventions provided all three of these communication modalities in a single site/application, making it difficult to identify exactly which component of the intervention accounted for the outcomes.14,19,20,23,25,27,3236,38,74

Online peer support groups and chat rooms, whether professionally moderated or not, were consistently found to be both the most-used and most-valued components of any given website, application, or intervention.23,24,34,73,74 Even preliminary feasibility studies found that users expressed a desire for more social support and inter-personal communication, that presumably would be addressed with such an intervention.31 Analysis of conversational themes from one online chat room found that users frequently acknowledged a perceived benefit both from participating in a community of peers, and knowing that other caregivers faced similar challenges.30 Multiple studies speculated that, compared to in-person support groups, online peer support groups improved access and anonymity. Clearly, online, asynchronous communication allows users to participate in a support group while mitigating some of the barriers of travel time, geographic separation, and time constraints from competing priorities. Evaluations of Caring~Web also pointed out that online communications provide access to a diversity of peers and clinicians that would otherwise not be available in many communities, particularly smaller town and rural environs.28,30 Anonymity was often considered an important feature to users of online support groups, and several authors speculated that caregivers and patients with chronic medical illnesses may have feelings of stigma or actual communication disabilities (hearing difficulties, slowed cognitive processes, etc.) that are overcome in faceless, asynchronous communication.24,28,34,35,42

CHIT interventions commonly allowed users to pose questions to a clinician, variously called “Ask an Expert/Nurse,” “Expert Care,” “Q&A,” or “Frequently Asked Questions.” Such communication with clinicians allowed users the opportunity to ask questions and, in many cases, the intervention included an online “library” of previously asked questions with the related responses. All of the interventions in adult caregivers and patients provided this functionality via email from the website; none used alternative technology such as SMS/text messagings. In the majority of interventions, the email was sent to a nurse specialist; although email access to other clinician experts was also described, including an Alzheimer’s disease counselor and case manager,14 occupational therapist and social worker.17,18 Two interventions described in detail how a multidisciplinary team provided the expert care: the email went to a nurse coordinator who then consulted with the cardiac transplant team in one case,38 or a team consisting of a physician, pharmacist, dietician, social worker, and therapists.26,31 In the pediatrics literature, use of SMS/text messaging in the evaluated intervention was quite common perhaps due to researchers’ perception that parents of young children were likely to be familiar with current mobile phone and computing technologies.

User-centered design

According to the Usability Professionals Association, User-centered Design is the approach to design that grounds the process in information about the people who will use the product.75 The process focuses on the usability of the product or service, making sure that users can readily and effectively accomplish their tasks as intended. Usability is considered part of a wider concept of user experience, or the way an individual (user) feels about using a product or service. User experience highlights the experiential and affective value or other meaningful aspects of human interaction; and also includes perceptions of utility, ease of use and efficiency of the system. User experience can be dynamic and change over time as the circumstances change.


A limited number of studies assessed usability using formal tools and iterative design methods.32,39 Chambers et al. used the Website Analysis and Measurement Inventory (WAMMI) to assess the initial implementation of the Assisting Carers using Telematics Interventions to meet Older Persons’ Needs (ACTION) intervention.39 They found that usability problems related to difficulty understanding terminology used on the website (15%), confusing screen layout (11%), challenges learning to use the application, and problems with navigation. More telling, 77 percent felt it took too many steps to reach information they desired. In an exploratory study of a Swedish website application to support the informal caregivers of patients with depression, Stjernsward et al. found that the 20 pilot users gave the application a mean System Usability Score of 78 (range 43–98), suggesting good usability overall (>70). However, notably 20 percent of users dropped-out of this exploratory study, which may indicate that some users were unable to continue using the application due to usability issues.

One pediatric study evaluated the usability of cystic fibrosis, diabetes, or arthritis portals by novice users.54 Their approach illustrates some of the very simple and practical information such testing can yield. For instance, parents pointed out use of jargon and abbreviations as barriers to understanding information, and they wanted confirmation that communication with healthcare professionals was received.

The Caregiver Support Service, developed for Chinese Canadians and associated with decreased caregiver burden among frequent users, examined language preferences among 28 caregivers of patients with dementia.17,18 Users could access a bilingual portal for caregiver information or email an occupational therapist and social worker. Most caregivers resided with the patient; the typical caregiver was female, aged 40 to 60, worked full-time, had a college education, was born outside of Canada and immigrated to Canada 10 to 20 years prior. Most care-recipients were the parents who co-resided with the CG. All spousal CGs invited to participate declined use of the email support service, and the study authors speculated that this may be a cohort effect associated with age.

A handheld device called PocketBuddy was developed for use by aging adult caregivers, with accommodations for poor vision, cognitive function, motor skills and hearing.15 Features of the device include landscape mode for greater screen space, optional large font to improve readability, button lists, finger tapping, enlarged keyboard, audible cues, and multimodal support options. Various functions include assisting in caregiving activities, monitoring the well-being of both caregiver and care-recipient, and networking the caregiver with family and friends to actively involve others in the daily life of the caregiver and care-recipient. The device was pilot-tested by eight older adults (aged 65–89 years), including two spousal Alzheimer’s disease caregivers. The caregivers expressed satisfaction with the system and its potential usefulness, though further details were not reported.72

ACTION caregiver users had a mean age of 70 years (range 47–88 years) and the technology was specifically designed to meet the needs of older adults, including large print to accommodate vision deficits.39,41 The Prostate Interactive Education System (PIES) intervention to assist diagnosis and treatment decision-making found that 89 percent of users had no vision difficulties impacting ability to use technology, and 83 percent had no difficulty with using a mouse.25

In general, these studies highlight some notable barriers and accommodations necessary when considering use of technology interventions for older caregivers, whether deployed on mobile computer platforms or not. A demographic survey of cell phone Internet usage conducted by the Pew Research Center determined that seniors (aged 65+) had the lowest usage rate among any major demographic group. Although 70 percent of persons aged 65+ currently own a cell phone, only 16 percent of them use their phones to go online.76 Older caregivers may require training in the use of the device or application, and may also benefit from applications with special accommodations for aging vision and manual dexterity. However, one study did note that streaming video applications that might accommodate older caregivers with vision or attention deficits was the component with which users were most likely to experience technical difficulties. 33 Accommodations for language preference may enhance the utility of mobile applications for immigrant caregivers.

User experience

A number of studies described the usage of the CHIT interventions through the lens of the caregiver or patient users. Many of these evaluations or qualitative descriptions were not primary goals of the study, yet offer important information related to their drivers or barriers to use. A few examples are illustrated here, across the adult and the pediatric literature.

The modality of tool or type of functionality offered was perceived to be important in a few studies. Among teens with traumatic brain injury and family members, the availability of audio recording enhanced the users’ ability to relax and comprehend the written content.67 Parents given the choice of a DVD or an Internet-based intervention after a child’s injury felt that the web tool offered greater accessibility.69

Many studies offered peer-to-peer communication, which as described previously, achieved high satisfaction and use. Individual comments about peer communication tools provide glimpses into reasons for this user experience. Children with diabetes58 and juvenile arthritis,45 and their parents using a portal to communicate with peers received social support by “not feeling alone” and perceiving greater universality in their journey. Parents of children with cancer were positive about the convenience of communicating at any time, the ability to “vent” their feelings through words, and feeling less isolated knowing others were experiencing the same condition and issues.50,51 Caregivers of patients with Alzheimer dementia reported that use of peer-peer communication was reassuring and led to greater self-efficacy as a caregiver.18,20 While views of peer online peer support were overwhelmingly positive, some users described negative experiences. Parents of children with cancer felt stressed when another child died;51 others were less satisfied if discussions “went off topic,”51 or there were insufficient numbers of participants to achieve “critical mass”.32,50 Of note, although some groups were moderated by staff and others were not moderated, there were no comments about the effect of health professional moderation.

Some studies reviewed for this report also described the process and/or results of soliciting user participation in the development and design process. Parents and teens with juvenile arthritis provided valuable input during development, leading to improvements in website navigation and display of information, such as reducing ambiguity of medication information presented.45 Parents of children with diabetes also pointed out problems with medical jargon and lack of information about normal values of laboratory data.54 Caregivers of patients with depression found it important for developers to ensure there was clear descriptions of functions so as users would understand their value. These caregivers also pointed out that using real names on peer communication violated privacy; when reminded by the study team that real names were optional and described in the user instructions, users remarked that instructions usually were not read.32

Technical challenges

Technical barriers or lack of familiarity with technology were described in several studies as potentially limiting accessibility of the intervention. Dew found that 16 percent of would-be users were unable to login and thus never able to access the intervention.38 Five of 34 families never used the ACTION intervention site due to technical difficulties.40 Multiple studies documented that intended users of the implementation frequently did not have a computer, let alone Internet access; those who did were only minimally experienced in their use, and sometimes the caregivers’ or care-recipients’ hesitancy to use an unfamiliar technology could not be overcome;25,26,32,34,38 however, these articles are through 2009 and thus might not represent current users’ computer and Internet experiences. Both the Caring~Web26,27 and ACTION 39,40 interventions were designed for use with a standard television and remote control (using Microsoft’s WebTV product) in order to increase user familiarity and overcome barriers, such as computer and Internet literacy. Of note, no studies found that security or privacy concerns were a barrier to use of technology.

One conference abstract described a qualitative study analyzing reasons why an electronic care management tool for type 1 diabetics failed to meet functionality expectations. Investigators conducted key informant interviews and surveys of patients and their parents which revealed several points of failure, including lack of patient or parent time/interest, glucometer malfunction, and lack of Internet service (especially for participants in rural settings).57

Few studies reported the amount of technical assistance and training provided to users. As part of the Caring~Web intervention, a technician installed the television software required or, in some cases, a computer and Internet connectivity, instructed caregivers and patients on its use, provided a paper-based support manual, and was available for technical assistance via phone throughout the study. 26,27 Families enrolled in the Online Family Problem-Solving (OFPS) intervention for children with TBI were given a computer, printer, and webcam, and extensive in-person training on use of the equipment.66 In contrast, Chambers noted that during ACTION’s Ireland implementation, staff had insufficient time to explain the system to users.39 Not one study quantified the costs or amount of time that user training and ongoing support required; although, anecdotal experience suggests that these often surpass the initial capital investment of technology implementations.

KEY QUESTION #2. What lessons can be learned from studies evaluating consumer health information technologies (CHIT) that specifically target the parents/caregivers of children?

We found 26 studies of 22 CHIT interventions in a variety of pediatric populations describing caregiver involvement with the intervention and/or caregiver outcomes (Table 2). In all cases, parents were the caregivers being described. Cancer (4 interventions), traumatic brain injury (3 interventions), and diabetes (2 interventions) were the most common target conditions. The following sections describe examples of different intervention subtypes tested in pediatric populations.

Self-management education and enhancing caregiver-patient communication

A number of studies described interactive educational web-based interventions designed to help patients and their parents develop self-management skills for chronic illnesses such as TBI and asthma. The largest group of studies described a multi-component intervention for children with traumatic brain injury and their parents. The online family problem-solving (OFPS) intervention includes logistic and educational material presented in 12 distinct interactive web sessions. The family collectively completed each module, entering responses to questions posed by the program and receiving tailored feedback from the website. Completed exercises were stored online and reviewed by a therapist, who also conducted a synchronous online videoconferencing session with the family after each module was completed. A small, unblinded RCT found the OFPS was more effective than static Internet resources alone in reducing parental depression and anxiety, though it was impossible to separate the effects of the web intervention from the videoconferencing sessions. Of note, nearly half the parents had no more than a high school education suggesting the potential feasibility of a complex web-based intervention in populations with varied educational backgrounds.64 The same group of investigators more recently developed a version of OFPS designed for adolescents and similarly found reduced parental depression and improved parent-child communication.67,68 An earlier feasibility study of the OFPS intervention found that technical issues such as the type of webcam used and virus infection could affect intervention usability.66

One RCT of 228 children and their parents evaluated the effects of a multi-media Internet asthma education program compared to standard print/verbal education on child/parent disease knowledge, asthma symptoms and emergency room utilization.46,47 The intervention, which included 44 vignettes designed to be reviewed in about one minute – suited to a child’s attention span – increased child and parent knowledge of asthma, and was associated with significantly decreased ER utilization and asthma symptoms compared to control participants.

Peer communication

Caring for a chronically ill child can be anxiety-provoking and also isolating, as parents of healthy children may not understand their experience. Several studies examined the role of electronic peer communication strategies. Han et al. surveyed 73 parents about their experience with online support groups for parents of children with cancer.51 Over three-quarters (77%) of parents highly valued the informational aspects of the support groups including information about the cancer itself and its treatment. While they also valued sharing experiences, receiving general support, venting feelings and gaining accessibility, they identified large volumes of often off-topic mail/posts and lack of face-to-face contact as drawbacks.

Demaso et al. evaluated a web-based tool for sharing personal stories about caring for children with cancer.56 They found that on a seven-point scale, parents reported the application increased their understanding of their experience as a caregiver of a child with depression (mean rating of 5.4). In addition, 83 percent reported that the tool helped them to understand symptoms of pediatric depression. Families in this study also reported that they learned ways to cope with stressful circumstances.

System transactions, accessibility and efficiency

Some of the online tools reported gave 24-hour access to laboratory results, medication lists,52,62 and secure messaging with providers.53 This kind of access – for example, secure-messaging – was also seen as more efficient and less frustrating than playing telephone tag with clinicians.53

Two studies evaluated text messaging interventions. One examined the effects of an automated text message reminder system on medication adherence and transplant rejection rates in pediatric liver transplant patients.60 Participants entered contact information online for patient and caregiver, medication information, and the desired time for alerts. Text messages were initially sent to the person responsible for medication administration (parents for younger children, and the patients themselves in older age groups); the caregiver was notified if a message confirming medication intake was not received within 30 minutes. A pre-/post-study of 41 children found significant improvement in immunosuppressive drug levels, and a reduction in biopsy-proven transplant rejection over one year. Though the impact of this simple intervention on health outcomes is promising, several issues including a high attrition rate (37%) which may raise concerns about the acceptability of this type of reminder system should be noted.

One large RCT examined an intervention to increase influenza vaccination rates in an urban, low-income pediatric and adolescent population in which parents received up to five weekly text messages.59 The messages included educational information about the influenza vaccine – to reduce misconceptions – and information about Saturday flu vaccine clinics. Texts were personalized to include the child’s name in the body of the text message. Both the intervention group and the control group received usual care, which was an automated telephone reminder. A higher proportion of intervention group (42.6%) received immunizations, compared to controls (39.9%) (relative rate ratio=1.09; CI=1.04–1.15, p=.001). This study demonstrated the feasibility of reaching significant numbers of parents/caregivers using text messaging. Though the absolute increase in vaccination rates was only three percent, given the low cost of texting, this is still likely to be a cost-effective intervention. Of note, the texts used multiple strategies aimed at reducing barriers to child vaccination: one was to inform about the vaccine, intended to reduce parent concerns about possible dangers from vaccination; a second was providing practical logistic information about the time and place for the vaccine clinics; and a third was a reminder function.

User experience

This information is summarized in Table 2 and in the “User experience” section under Key Question #1.

KEY QUESTION #3. What are the major gaps in the consumer health information technology literature serving non-professional caregivers of adult patients with regards to technology development, availability, and/or evaluation?

Table 3 summarizes the interactive applications and functionality of the interventions included in this review, and indicates the current phase of the interventions’ lifecycle at the time of publication.

The CHIT literature reflects a relatively new, developing field. Most studies described interventions in early development (ten studies) or pilot-tested (five studies) on a small scale. Only seven studies were developed to evaluate health outcomes, but most of these were relatively small studies. There is a dearth of literature describing the health outcome effects of CHIT in larger populations. Some of the larger studies involved interventions, such as text messaging, which might be logistically simpler to deploy and test on a large scale. Almost no studies evaluated the actual implementation of interventions that had already been tested and found to be efficacious.

Reviewed studies were also not designed to qualitatively develop a contextual understanding of the use of the intervention technology. At this time, there is not information to assess how these interventions fit into the day-to-day lives of caregivers. Additionally, there is relatively little information about how caregiver demographic characteristics affect the user experience. These are promising areas for future research; however, longer horizons will be necessary to assess usage over time given that caregivers’ needs for information, skills, and support are likely to evolve.

Intervention content varied widely. Nearly half the interventions, interestingly, included a component designed to facilitate peer communication. On the other hand, despite the focus on supporting caregivers, relatively few interventions included components specifically designed to enhance communication between caregivers and patients. Similarly few studies analyzed outcomes separately for caregiver and patient users; notable exceptions were CHESS,23 Caring~Web,27 WECARE,34 and The Schizophrenia Guide.35 Thus, it largely remains unclear if there are differences between caregivers and patients in their interactions with and benefit derived from technology. This likely has implications for perceptions of value, usability, utility, and satisfaction, as well as usage and challenges to adoption. Studies that simultaneously enroll both the patient and caregiver as a dyad, as was performed in CHESS,23 WECARE,34 and Dew et al.38 may provide an opportunity to better understand these relationships. The question of whether technology implementations should be designed for the caregiver or the patient, or both as the end user is not answerable from the current literature, and may be best addressed by expert opinion and consensus. Other important gaps include an understanding of how these mobile interventions complement existing services. We also found very little information about the cost-effectiveness of CHIT.