Approximately 16 million Americans serve as informal caregivers, providing 18.5 billion hours of care, which translates into US $234 billion in health care savings [1]. Dementia caregiving in particular is more demanding than other types of caregiving, due in part to time-consuming and unpredictable caregiving demands and associated emotional and physical exhaustion [1,2]. Informal caregivers of people with dementia frequently experience insomnia [3,4]. The number of people with dementia in the United States is projected to rise from 5.7 million to more than 14 million in the next 30 years [5]. Most people with dementia (70%) are cared for at home by a family member [1]. Thus, there is a critical need to evaluate the health of dementia caregivers and provide treatment for prevalent medical conditions such as insomnia.

In addition, because rural areas are aging faster than urban areas, the proportion of people at risk for dementia is growing faster, which translates to faster growth in the proportion of rural caregivers [6]. Compared to their urban counterparts, rural caregivers face additional challenges, including more difficulty accessing health care, a greater likelihood of social isolation, and increased vulnerability to sleep problems, stress, and depression. Rural caregivers have less access to care due to significant shortages of both primary and specialty care providers in rural areas [7,8] and are also at disproportionate risk for poverty and unemployment [2], which further limits their access. On average, they travel 144 km roundtrip to see their nearest health care provider [9]. Lengthy travel combined with the lack of public transportation in rural areas represent an additional major barrier to access. Thus, developing and evaluating treatments for medical conditions that are prevalent in caregivers (eg, insomnia) that can be accessed remotely (eg, on the internet) is a critical and currently unmet need in the health care community.

Research from our group and others verifies that caregivers have poorer sleep quality [10,11], take longer to fall asleep [10], spend more time awake during the night [3], and sleep less [10,12] than same-aged noncaregivers. Chronic insomnia (more than 3 months of difficulty initiating or maintaining sleep, early morning awakening, or nonrestorative sleep) affects up to 63% of caregivers [3,13] and tends to endure (18 years on average in our previous caregiver trial) [14,15]. Factors associated with caregiver insomnia include age-related sleep architecture changes, increased sympathetic nervous system arousal [3,15], increased vigilance [3], stress, depression, cognitive dysfunction, increased dementia risk, and person-related dementia behaviors (including nighttime wandering).

Cognitive behavioral therapy for insomnia (CBT-I) improves sleep (moderate or large effects) and mood (large effects) in adults of all ages [16] and is recommended by the American College of Physicians as a first-line treatment for chronic insomnia [17]. While research [18-20] shows that CBT-I is an efficacious treatment for the chronic insomnia experienced by caregivers, rural caregivers face particular difficulties accessing behavioral treatments, which require specially trained providers and administration over multiple sessions in a provider’s office. In a recent randomized controlled trial, our group found that brief CBT-I delivered through telehealth (videoconferencing) reduced caregiver sleep onset latency compared to sleep hygiene education in caregivers. Our CBT-I protocol was adapted with permission from McCurry and colleagues’ [3] earlier version [21] and included standard CBT-I techniques (sleep hygiene, stimulus control, and cognitive therapy), modified CBT-I techniques (sleep compression replaces sleep restriction, which is potentially burdensome for caregivers due to temporary sleep deprivation; brief hybrid relaxation is included given caregiver time constraints [22]; and muscle tensing is excluded given potential pain or musculoskeletal issues in older caregivers), additional techniques for caregivers (problem-solving, as skill deficiency can contribute to caregiver insomnia [23,24]), and stress management (respite, staying healthy, and communication). Our findings indicate CBT-I tailored for caregivers translates well to telehealth delivery and that technology holds great potential to increase rural caregiver access to behavioral interventions. However, telehealth delivery still requires considerable time commitment from trained therapists (who are already in short supply).

A small but growing body of evidence shows that web interventions have been successfully implemented in rural areas using the internet and other mobile platforms and improved noninsomnia health outcomes in stroke and caregivers [25-32]. Moreover, web-based CBT-I has been implemented and found to be efficacious in noncaregiver patient populations. For example, Taylor et al [33] translated 6-week CBT-I to a web format and conducted a randomized controlled trial in military personnel (N=100; n=33 received web CBT-I, n=34 received in-person CBT-I, and n=33 received minimal contact control). Compared to control, web-based and in-person CBT-I improved sleep efficiency, sleep onset latency, and wake time after sleep onset (effects moderate for web-based, large for in-person). Internet CBT-I also improved insomnia severity and decreased sleep-related cognitive arousal [33]. Another 6-week web-based CBT-I program (SHUTi) [34] was shown to improve insomnia severity (as measured through the insomnia severity index) relative to a no-contact control (large effect size) in adults with insomnia. Noted limitations of internet-based treatments include a lack of personal contact to address treatment questions or concerns without a face-to-face or telephone component. Additionally, personal support or guided delivery has been associated with larger treatment effects for both in-person and web-based CBT-I [26]. Taken together, these findings provide support for the development of web-based CBT-I for caregivers and suggest personal support or guided delivery is important to maximize treatment adherence and efficacy. Thus, our team developed a web-based CBT-I for caregivers through an iterative process.

Following the Medical Research Council recommendations [35] for the evaluation of complex medical interventions, we conducted several steps in the initial development, pilot-testing, and evaluation of NiteCAPP CARES [36] (see Figure 1).

This paper describes stages 1a and 1b, which were conducted from 2019 to 2020. The goal of stage 1a was to translate (through feedback from caregivers and primary care providers; PCPs) an efficacious 4-session CBT-I into a web-based “NiteCAPP CARES,” a web-based behavioral treatment for insomnia for dementia caregivers that incorporates guided delivery through weekly therapist moderator feedback. A second goal of stage 1a was to conduct NiteCAPP CARES usability testing to evaluate the acceptability of its content and features. The goal of stage 1b was to pilot-test NiteCAPP CARES on several key outcomes (feasibility, favorability, satisfaction, sleep, arousal, mood, burden, and cognition) in a small sample of dementia caregivers. We hypothesized that NiteCAPP CARES would produce high ratings regarding favorability of program content. Further, we hypothesized that in the pilot study, participants would complete all 4 sessions on average and that satisfaction ratings would be at least more than 7 of 10 (highest) and usefulness ratings would be at least more than 4 of 5 (highest). Although the single-arm design of this pilot study precludes examination of efficacy, within-group effects of the treatment on sleep and related behaviors were also examined. We hypothesized that subjective sleep would improve immediately following treatment using NiteCAPP CARES. We further hypothesized that daytime functioning (ie, mood, caregiving burden, perceived stress, and cognition) would improve.

This study was approved by the University of Missouri Institutional Review Board (2017125) and registered on ClinicalTrials.gov (NCT04632628).

Feasibility and acceptability were excellent, with high average completion (>100% of sessions), adherence (76% for sleep hygiene, 80% for stimulus control, and 81% for relaxation based on daily logs), satisfaction (>8.4/10 on the Satisfaction Survey; see Table 5), and usefulness ratings (>4.3/5 on the Internet Intervention Utility Questionnaire; see Table 6). As shown in Table 7, NiteCAPP CARES led to improvements in subjective sleep (sleep onset latency, wake after sleep onset, sleep efficiency, and the Insomnia Severity Scale), arousal, mood (depression), burden, quality of life, perceived stress, and subjective cognition.

Following this pilot test, additional participant qualitative feedback (see Table 8) was used to update NiteCAPP CARES a third time. Participants stated that the written and audiovisual materials were clear, though audiovisual materials were repetitive. Comments and suggestions included moving relaxation techniques to the morning and the importance of a schedule. Therapist moderator feedback was used to refine the guided web-based support and develop a moderator manual with clear, step-by-step instructions, schedule for administration of support, and case examples (see Textbox 1).

This study used mixed methods of qualitative and quantitative expert (PCPs) and participant-focused (rural dementia caregivers) feedback to iteratively adapt a 4-session web-based CBT-I (NiteCAPP) into a tailored digital format with standard CBT-I content plus caregiver-focused modifications (NiteCAPP CARES). The usability and acceptability of NiteCAPP CARES were then evaluated. The website was evaluated as easily accessible and useful, and the feasibility of the web-based intervention was high, suggesting that NiteCAPP CARES is highly functional for rural caregivers and PCPs. Preliminary efficacy testing also indicated that NiteCAPP CARES improved subjective sleep as well as several key aspects of daily functioning (eg, arousal, mood, burden, and subjective cognition).

Consistent with our hypotheses, the acceptability and usability of NiteCAPP CARES were high. In comparison to other in-person or telehealth versions of CBT-I [61-64], NiteCAPP CARES may offer unique advantages for treating insomnia, particularly in dementia caregivers. For example, NiteCAPP CARES provides greater caregiver accessibility given their demanding schedules since caregivers can access the materials on their schedule without having to make any appointments. NiteCAPP CARES is also less burdensome for both the patient and the therapist moderator, as trained behavioral sleep therapists who can conduct in-person or telehealth CBT-I are in short supply [8]. Further, compared to other web-based insomnia treatments [65-69], NiteCAPP CARES provides important enhancements for older caregivers. For example, although other 5 to 6–session, web-based CBT-I platforms [65,66] provide audiovisual content, NiteCAPP CARES provides tailored audio and visual recommendations for dementia caregivers. In addition, unlike other web-based CBT-I platforms, NiteCAPP CARES also provides a help forum to provide 24-7 assistance from therapist moderators. Other web-based CBT-I platforms [67] use sleep diaries and have a clinician and patient portal but function as adjuncts to in-person treatment, whereas NiteCAPP CARES is a standalone moderated treatment that allows the caregiver to access all the materials on their schedule. Other web-based CBT-I platforms [68,69] offer 6 sessions with tailored sleep recommendations based on either sleep diaries or screening questionnaires. One web-based CBT-I [69] even provides a web-based therapist who advises the patient on restructuring their cognitions, but this feedback does not extend to the patient’s sleep. However, none of these web-based insomnia treatments provide a simpler interface tailored for older caregivers or provide tailored and moderated therapy throughout treatment. In addition, the NiteCAPP CARES website itself was rated very highly in terms of its ease of use and readability (eg, font size), which were part of its initial user-centered design that took age-related cognitive or perceptual concerns in our target population into consideration (see Methods section) [70-74]. Second, caregiver use was likely promoted through the use of nonstructured session times (caregivers were encouraged to complete the sessions on their own schedule), simplified language, videos, and pictographs.

As expected, preliminary results also reveal that NiteCAPP CARES shows promise for improving the majority of our key outcomes. The moderate to large improvements across sleep outcomes (sleep onset latency, wake after sleep onset, sleep efficiency, and insomnia severity) found for NiteCAPP CARES are similar or greater than those found for other web-based CBT-I platforms [65,66,69]. Extending findings from previous web-based CBT-I interventions, our preliminary results also provide novel findings regarding NiteCAPP CARES’ potential impact on stress or arousal and subjective cognition (a treatment outcome that has not previously been assessed). NiteCAPP CARES showed improvements in subjective arousal (moderate to large effects) and cognition (small effects), as well as burden (small to moderate effects). Unlike other web- and app-based CBT-I platforms [65-69], the use of a therapist moderator was noted as a positive feature of the intervention during pilot-testing and likely contributed to its acceptance and high completion rate. Previous completion percentages of therapist-led CBT-I treatment range from 63% to 88% [61,62] for telehealth treatment and 80% to 98% [63,64] for in-person treatment, both below NiteCAPP CARES’ 100%. Similarly, average treatment adherence for therapist-led CBT-I is 38% and digital CBT-I is 35%, both lower than NiteCAPP CARES’ 76%-81% [75]. NiteCAPP CARES suggests that moderated web-based interventions may be comparable to or surpass the completion rates for in-person and telehealth delivery of CBT-I in this population. This will provide increased accessibility for caregivers living in rural areas that are not close to health care providers.

Although further testing of NiteCAPP CARES is currently underway [76], these preliminary findings potentially suggest several clinical implications. Improving caregiver sleep is critical given that caregiver insomnia is associated with negative changes in central stress processing, resulting in poor physical health (altered heart rate and heart-rate variability) [77] and mood (anxiety and depression) [18,78]. Insomnia is also associated with poor caregiver cognition (processing speed [79,80], attention [79,81,82], executive functioning [82], and memory [82,83]) and neurodegenerative biomarkers associated with Alzheimer disease [83]. Poor mental and physical health makes it harder for caregivers to serve as caregivers and poor caregiver sleep contributes to early institutionalization [84]. Thus, treating caregivers’ insomnia may allow them to continue to provide care longer. Further, NiteCAPP CARES is a moderated, web-based intervention, as opposed to more traditional in-person or telehealth interventions that involve therapist-led sessions [85,86]. Thus, this intervention can be disseminated broadly, and the therapist moderator manual will enable other health care providers (eg, nurses and PCPs) to learn the treatment quickly and use NiteCAPP CARES in various treatment settings.

This study provides a foundation for future sleep research in rural dementia caregiver populations, which have largely been ignored. Our study has several limitations. First, the small number of individuals in our focus group and pilot test may not generalize to a broader population of caregivers and PCPs. Similarly, our small sample may have inflated our effect sizes; however, our previous studies in similar populations have shown these effect sizes to be robust [18,20,36]. Future studies should include a larger sample size that includes individuals from many different facets of life. For instance, it would be beneficial to gain insight from people with dementia, as NiteCAPP CARES treatment modules can be expanded to include resources specific to the care recipient. Furthermore, it would also be important to seek more evaluation from health care administration teams, who are likely to be the ones to give patients step-by-step instructions on how to log in and navigate NiteCAPP CARES. Future studies should focus on gaining feedback from dementia patient advocates, who have a considerable amount of insight on how to effectively care for and work with people with dementia and caregivers. The promising results from this preliminary development and initial evaluation study suggest a future randomized controlled trial that examines the efficacy of NiteCAPP CARES compared to an active web-based control in a larger sample of rural dementia caregivers is an important and warranted next step. Such future work may also benefit from a tailored, dyadic approach that involves the person with dementia in the treatment to the extent they are capable. Following further development and the establishment of efficacy, additional future steps include an effectiveness trial and broader implementation and dissemination efforts, as well as the development of a web-based, stand-alone sleep treatment for dementia caregivers. Specifically, as noted above, NiteCAPP CARES’s moderated web-based intervention provided excellent adherence and engagement compared to completion rates for other web-based treatments [58-61]. While this preliminary study’s results support our use of a moderated approach, a web-based, stand-alone option may be preferred by some caregivers and would provide them with an additional access option for obtaining CBT-I.

A brief, web-based CBT-I platform called NiteCAPP CARES was created for caregivers to address caregiver sleep issues. Initial website feedback was accessed by asking rural caregivers and PCPs their opinions on NiteCAPP CARES. They thought the site was easy to understand, engaging, and well designed. Preliminary pilot-test data indicated that caregivers used the site (high treatment adherence and session completion), found it useful, and reported high satisfaction. NiteCAPP CARES also improved subjective sleep and daily functioning. These promising outcomes suggest that NiteCAPP CARES can be a beneficial sleep tool for both rural dementia caregivers and health care providers. Underserved populations who have difficulty obtaining evidence-based health care may particularly benefit from NiteCAPP CARES given its web-based delivery, which can be accessed anywhere the internet is available. A future randomized controlled trial to evaluate NiteCAPP CARES compared to an active web control in a larger sample is needed.