The effect of digital communication technology on older adults’ formal care use in Norway: a randomized controlled trial

The effect of digital communication technology on older adults’ formal care use in Norway: a randomized controlled trial | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article The effect of digital communication technology on older adults’ formal care use in Norway: a randomized controlled trial Kristoffer Eikemo, Ira Malmberg-Heimonen, Erik Børve Rasmussen, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8378683/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 11 Apr, 2026 Read the published version in BMC Geriatrics → Version 1 posted 11 You are reading this latest preprint version Abstract Background Norway’s population is aging, increasing the pressure on already strained health and care services. In response, Norwegian policymakers have promoted a strategy of aging in place, i.e. that older adults are encouraged to live longer in their own homes. Low social contact is one of many risk factors for transitions to long-term care facilities. Digital technologies that facilitate social contact have been promoted as promising tools for increasing social contact among older adults and may also contribute to reduce strain on health and care services. In this RCT, we evaluate whether a digital communication technology, Komp, may prolong home-dwelling and reduce the need for formal care services among frail older adults. Given Komp’s documented potential to promote social connectedness and facilitate informal care, we hypothesize that Komp prolongs home-dwelling and reduces reliance on formal care services. Methods We compared an intervention group offered to try Komp (n = 516) with a control group receiving services as usual (n = 595). 150 Komp units were delivered. Outcomes were assessed 28 months after randomization. We obtained administrative data from Statistics Norway and service use data from the boroughs’ journal records. We analyzed 1,099 participants (mean age: 84.6 years, range: 67–98; 64.7% [ n = 711 ] female) using cox proportional hazards regression (primary outcomes) and ZINB and ZIP regression models (secondary outcomes). Results The intervention group had a higher, though non-significant, risk of moving to long-term care institutions compared to the control group (HR = 1.162 [.944, 1.431]). ZINB regression models indicated that the intervention group experienced a reduction of five days of short-term institutional stays (IRR = .833 [.728, .952]). There were no differences between groups in the amount of other formal care services received. Discussion and Implications The implementation of Komp did not prolong home-dwelling among frail older adults. However, the results showed a reduction in formal care services, indicating that Komp may reduce some strain on health and care services. Future studies should evaluate the effect of other types of digital communication technologies on objective outcomes. Trial registration ClinicalTrials.gov: NCT05919355. Date of registration: 16.06.2023. Age in place Survival analysis Home and community based care and services Long-term Care Intervention study design/analysis Figures Figure 1 Figure 2 Figure 3 Background Like in all European countries, Norway’s population is ageing. By 2050, one in five Norwegians will be over the age of 70 [ 1 ], similar to projections across the OECD [ 2 ]. This demographic shift increases pressure on already strained health and care services. Norwegian policymakers have responded to the challenge by promoting a strategy of ‘ageing in place’, i.e. that older adults primarily should receive long-term care in their own homes rather than in long-term care facilities [ 3 ]. While ageing in place may promote independence and autonomy among many older adults [ 4 ], it is not feasible for all. The reasons why some older adults transition to long-term care are multifaceted and complex. Functional impairments and caregiver burden have been identified as key factors [ 5 ], along with low social contact [ 6 ]. Low social contact is further linked to lower quality of life [ 7 , 8 ], cognitive decline [ 9 , 10 ], and poorer overall physical health [ 5 ], which are further associated with greater reliance on formal care services [ 11 ]. Promoting social connectedness is therefore recognized as essential to ageing in place [ 12 – 14 ]. Digital communication technologies (DCTs) that allow for synchronous communication with formal caregivers, family, and friends [ 15 ], have been highlighted as promising tools to support healthy ageing [ 16 , 17 ]. Previous research indicates that DCTs may enhance social connectedness, reduce loneliness and improve well-being among older adults by facilitating communication with their social network [ 18 – 22 ]. However, despite promising findings, we lack studies with robust causal designs to evaluate their effects [ 23 – 27 ]. Many randomized controlled trials (RCTs) have evaluated DCT interventions among home-dwelling older adults on subjective outcomes such as loneliness and social isolation. However, most have focused on clinical content delivered through technology, such as telehealth [ 28 ] or cognitive therapy [ 29 ], rather than the social function of the intervention itself. To our knowledge, only two RCTs have examined DCTs as standalone interventions aimed at facilitating social connectedness among older adults. Czaja et al. [ 30 ] conducted a multisite trial in Florida and Georgia, United States, among individuals aged 65 or older living alone in independent housing. Participants received either PRISM, a computer system designed for older adults that included communication tools (e.g., photo sharing), or a control condition that provided similar information in a paper binder. Results showed that participants receiving PRISM experienced a significant increase in perceived social support and well-being, as well as a reduction in loneliness, compared to controls. In a follow-up RCT, Czaja et al. [ 31 ] evaluated PRISM 2.0, an updated tablet-based version with expanded social features, among individuals aged 65 years and above living in senior housing, rural locations, and assisted living communities in Florida and Georgia. The intervention group was compared to a control group using a tablet with standard applications. Both the PRISM 2.0 and standard tablet groups living in senior housing and rural areas reported reductions in loneliness and social isolation and improvements in quality of life and social support. To the best of our knowledge, no RCT-studies has investigated the effect of the social function of DCTs on objective service outcomes, such as formal care use and transitions to long-term care. The present study addresses these research gaps by evaluating the effect of the DCT device Komp on formal care use among home-dwelling older adults aged 67 years and above. Komp was designed for individuals who are unfamiliar with, or struggle with, conventional digital technologies [ 32 ]. All interaction beyond turning the device on or off and adjusting the volume is managed by the user’s selected social network through a connected app. This network can initiate video calls and send photos and text messages, which appear automatically on the Komp screen. Findings from previous research on Komp indicate that this tool does increase feelings of safety, support meaningful digital contact between users and their social network, and thus facilitates informal care [ 15 , 32 – 35 ]. Based on the indications that Komp does support social contact and increase feelings of safety, our trial’s program theory proposes the following mechanisms. First, increased social contact and perceived safety may mitigate factors such as loneliness and insecurity that increase the likelihood of moving to a long-term care facility and greater service needs. Second, by strengthening informal care availability and everyday reassurance, Komp may delay transitions to long-term care and reduce reliance on home-based services. Thus, although not providing clinical content, we hypothesize that Komp prolongs home-dwelling and reduces reliance on formal care services among older adults. In accordance with the trial protocol (NCT05919355), our primary objective is to assess whether access to Komp prolongs the service recipients’ home-dwelling by delaying moving to long-term care facilities. Our secondary objective is to evaluate whether access to Komp reduces the use of other formal care services: home nursing care, short-term institutional care, practical assistance, and safety alarms. The original implementation plan also included municipal care staff use of Komp, which ultimately did not occur due to delays. This deviation has implications for our program theory and expected outcomes. While, according to our program theory, the intervention’s ability to address social needs and strengthen informal care should still influence the use of formal services, the anticipated effects are likely to be weaker when the device is not simultaneously adopted by formal care providers. Methods Study Design This study is designed as a field trial, with a parallel group assignment, and an intention-to-treat (ITT) approach. It is conducted as part of the BoVel project – a collaboration between Oslo Metropolitan University, Oslo Municipality, and Abilia, which is a private company that develops and sells Komp. Recruitment began on December 21, 2022, with data collection spanning from October 13, 2022, to March 1, 2025. Our reporting follows the CONSORT checklist [37] as far as applicable. Study Context Our study was conducted in three boroughs of Oslo, Norway: Nordstrand, St. Hanshaugen, and Østensjø. In international comparison, Norway has a comprehensive welfare state with redistributive benefits and universal health services, including long-term care services for older adults [38]. Long-term care is largely a public responsibility, with municipalities covering 90 percent of the costs [39]. Many municipalities in Norway, including Oslo, have waiting lists for long-term care facilities [40], making home-based services, such as home nursing care, the primary means of supporting older adults in need of care [41]. Nursing homes are generally reserved for those with the most complex needs, as reflected in the fact that the average length of stay in a Norwegian nursing home is approximately two years [42]. Participants We identified potential participants among 9,317 municipal health and care service records from three Oslo boroughs. Eligible participants had to (i) be 67 years or older; (ii) dwell in a private home (i.e., not in a permanent care or nursing home); (iii) have a registered address in one of three participating boroughs of Oslo, Norway; and (iv) be registered recipients of municipal home nursing care. Individuals were excluded if they had full function in all of the three following Activities of Daily Living (ADL) measures, as assessed by the services: outdoor mobility, memory, and cooking. They were excluded because having full function in all these three areas meant a very low risk of moving to long-term care during the trial, and thus they were outside the intervention’s target group. This exclusion process resulted in 1,170 eligible participants. These were sent written information about the project, with the option to decline participation. Information was first distributed electronically, followed by a postal version if the electronic version was not opened within 14 days. In total, 59 individuals declined participation, leaving 1,111 participants. Figure 1 demonstrate the flow chart for the study. Assignment to intervention and control group Within RCTs, eligible participants must have a defined and equal probability of being assigned to the intervention, to create groups that are comparable with respect to both measured and unmeasured characteristics [43]. In this study, eligible participants (n = 1,111) were identified from municipal records as of December 21, 2022, with the data backdated to October 13, 2022. To ensure equal probability of allocation, we applied a form of permuted block randomization [43], treating each of the three boroughs as a separate block. Separate lists were created for each borough, and participants were randomly ordered by generating random number sequences in Stata. Local project coordinators offered Komp to names on the randomly ordered lists, working from the top down. We had 60 Komps available in St. Hanshaugen, and 120 in each of the two larger boroughs. Because many participants were not willing to receive Komp or not possible to reach during the delivery period, we reached half the lists and stopped the offers before all available Komps (n = 300) had been distributed. At this point, we had made an offer to 32 percent of the 1,111 participants. In cases where persons in the intervention group did not receive Komp, we registered them as no-shows. In accordance with the ITT approach, we kept them (n = 157) in the intervention group, as the control group would also include unavailable individuals. This resulted in 516 participants in the intervention group and 595 in the control group. Blinding was not possible due to the nature of the intervention. Although we cannot rule out the possibility that knowledge of group assignment influenced how home care services were delivered, we have no indications that this occurred. Qualitative interviews with home care professionals do not suggest any systematic differences in follow-up or service provisions [32], and any such differences would likely have been detected. After participants were offered Komp, they had about 14 days to accept. Those who accepted (n = 150) received a Komp device within one week, which they could use freely. Installation was carried out by project coordinators or other trained staff, and participants received written information about technical support. In most cases, the offer was made in person, accompanied by a demonstration of the device, and, when possible, with relatives present. Most devices were delivered over a period of 10 months, except for two cases receiving Komp 21 months after distribution commenced. At any time, participants who had received a Komp could decide to stop and return the Komp to the municipality. Twelve participants returned the device earlier, after on average, using it for 8.5 months. Participants who declined the offer continued to receive services as usual (n = 209). In accordance with the ITT approach, all individuals offered Komp were included in the intervention group. Control group members continued to receive home care services as usual. Intervention Komp is a DCT developed by the Norwegian company No Isolation to support social contact between older adults and their social network. The device consists of a screen with a single button for turning it on and off, as well as adjusting the volume. The user’s social network uses a connected app to send photos, messages, and make video calls. Calls are automatically connected after ten seconds unless the user turns the device off. While off, no calls go through, but a small blinking light indicates incoming calls. An important clarification is that the Komp user cannot initiate contact through the Komp screen; they can only receive. The design is intentionally made suitable for individuals with limited technological competence. However, it also limits opportunities for interaction, making Komp valuable only if others actively engage. See Akhtar [33] for more information on Komp. Success of Randomization We estimated significance of differences using two-sided T-tests for continuous variables and chi-squared tests for categorical variables (Table 1). The results show that the groups are balanced apart from wealth and marital status: Participants in the intervention group had slightly higher wealth, while a higher proportion of participants in the control group were married. We adjust for these differences in the effect analyses. Table 1: Baseline demographics Measures Primary outcome The primary outcome was time spent living at home, measured in days from the start of the project (October 13, 2022) until either moving to a long-term care facility or the end of the study period (March 1, 2025). Time spent living at home Data was obtained from the municipal journal system, Gerica. The date of moving was defined as the start date of the service “long-term institutional stay”. Participants who died at home were coded in three different ways in the analyses: they were (i) censored at the date of death, (ii) treated as events in the same way as moving to a long-term care facility, and (iii) treated as a competing event (see statistical analysis section). Those who remained living at home throughout the observation period were censored at the study end date. Two participants in the control group had missing data on service start. Secondary outcomes Secondary outcomes were measured as the amount and type of care services received. Home nursing care We obtained data on duration and weekly hours of home nursing care decisions from Gerica. These were operationalized as total hours per participant over the project period. As receiving home nursing care was an inclusion criterion, almost all participants had a registered value. Some (n = 24) who discontinued early were assigned zero hours. Short-term institutional stays Short-term institutions are time-limited services intended for rehabilitation, assessment of care needs, or respite for family caregivers. The aim is usually for older adults to return to their own homes after illness, hospitalization, or a period of increased care needs. These were recorded as whole days in Gerica, with no value for intensity. We coded stays as (i) total number of days in a short-term institution, (ii) total number of days capped at 200 to handle extreme outliers, and (iii) total number of unique stays at a short-term care facility. Participants without any records were assigned a value of zero. Practical assistance Practical assistance includes home support services, aimed at helping individuals manage daily activities such as financial management and general routines. Recorded with duration and weekly hours in Gerica, from which we calculated the total hours received per participant throughout the project period. We added an outcome variable with hours capped at 100 to handle outliers. Participants without records were assigned a value of zero. Safety alarms Safety alarms allow older adults to call for assistance at any time of day, in case of a fall or a sudden health problem. We obtained data on safety alarm activations from the supplier Careium. Each activation was coded as unique observations in the data set, with a reason for activation labeled by healthcare professionals. We excluded records labeled as test. Participants with no records were assigned a value of zero. Statistical power We conducted power calculations prior to the trial (Figure 3) using a two-sample proportions test in Stata (StataCorp, 2023). Following conventional thresholds of a 5 percent significance level and 80 percent power, we estimated that approximately 400 participants per group would be required to detect a 10 percentage point difference in the primary outcome. However, the realized exposure differed from these assumptions. Although 516 participants were allocated to the intervention group, only 150 received a device, and an even smaller proportion got meaningful use of it. This reduces the statistical power of the trial and must be accounted for when interpreting the results. Statistical Analysis All statistical analyses were conducted using Stata 18 [44]. For the primary outcome, time until moving to a long-term care facility, we used survival analysis. Hazard ratios (HRs) were first estimated in a crude model, before adjusting for baseline differences, using the Cox proportional hazards model [45]. Here, HRs above 1 indicate higher risk of moving, and HRs below 1 indicate a lower risk. We tested the proportional hazards assumption using Schoenfeld residuals [46]. No violations were detected. Further, we had a censoring issue. The Cox model assumes non-informative censoring, meaning that participants who are censored should drop out for reasons unrelated to the outcome [47]. In our case, deaths are directly related to the risk of moving to a long-term care facility. Thus, by censoring deaths, we violate this assumption. However, if we treat deaths as events of interest, we measure an outcome different from our primary outcome. For robustness purposes, we conducted two alternative analyses treating deaths both as censored and events. In addition, we applied a competing risks model [48] to account for deaths as a competing event, reporting subhazard ratios (SHRs). All secondary outcome variables were treated as count data. Almost all of them had issues with overdispersion (see Table 3), which violates the Poisson regression assumption of equal mean and variance [49]. To account for this, we used negative binomial regression models for home nursing care. The remaining secondary variables had, in addition, an excess of zero values. To address both overdispersion and excess zeros, we applied zero-inflated negative binomial (ZINB) regression models for days of short-term institutional stays, practical assistance, and safety alarm activations. ZINB combines a logistic component, which estimates the probability of structural zeros, with a negative binomial component, which models the count distribution among service users [50]. For the variable representing unique number of short-term stays, overdispersion was not significant. Although the variance was higher than the mean, the alpha parameter in the ZINB model was not significantly different from zero (not shown), indicating that overdispersion was not an issue [51]. Therefore, we used a zero-inflated Poisson (ZIP) model for this outcome. Model comparisons using AIC and BIC favored negative binomial models over Poisson for all secondary outcome variables [52]. Results are reported as incidence rate ratios (IRRs) for the count component, and odds ratios (ORs) for the excess zero component. Results In Table 2, we present descriptives for the primary and secondary outcomes. Over the 870-day follow-up, 45.9 percent of participants remained at home, 32.5 percent transitioned to long-term care, and 21.7 percent died while living at home. Rates of moving to care facilities were highest in the intervention group, while the control group had more deaths and more participants staying at home throughout the period. In total, the population accumulated 651,158 days of living at home, corresponding to an incidence rate of 0.0005. For all secondary outcome variables, the variance exceeds the mean, indicating overdispersion [51]. In Table 3, we present the results for the primary outcome, along with HRs from the Cox (1-4) and Fine-Gray models (5-6). Here, we assess the time to move into a long-term care facility. Across all models, the estimated HRs for the intervention group range from 1.090 to 1.162. This suggests that the intervention group had approximately 10 percent higher risk of moving to a long-term care facility compared to the control group. However, the differences between the intervention and control groups are not statistically significant. Implementation data shows that participants who had registered use on their Komp were, on average, registered with 1.1 activities per week (received video calls, images, or messages). Ten participants had fewer than ten interactions in total, and 23 had fewer than 0.5 activities per week. 20 participants never received videocalls, but only images and messages (46 activations on average among these). In Table 4, we present IRRs and ORs from models assessing group differences in formal care service use. Models 7 and 8 estimate total hours of home nursing care using negative binomial regression. Both models show no significant differences between groups. The estimated alpha parameter confirms overdispersion, as it is significantly different from zero [51], supporting the model choice. Models 9-12, report results from the ZINB models for short-term institutional stays. We find a significant reduction in the total number of days spent in short-term care for the intervention group (model 9: IRR = .833 [.728, .952]) of approximately 17 percent. Given the control group average of 31 days, this translates to a reduction of approximately five days per participant on average. The adjusted model (10) shows similar results (IRR = .828 [.725, .946]). The findings remain consistent when the dependent variable is capped at 200 days (models 11 and 12). The absence of significant effects in the excess zero component suggests that the odds of receiving the service were similar between groups. In models 13 and 14, we present results from the ZIP models for the number of unique short-term institutional stays, showing no differences between groups. Similarly, the ZINB models assessing hours of practical assistance (models 15-18) and the number of safety alarm activations (models 19-20) show no significant differences between groups. [Table 4 here] We conducted two post hoc analyses (not shown) to examine treatment effects among the treated: an as-treated analysis, comparing participants who received a Komp with controls, and a per-protocol analysis, comparing those with registered Komp use with controls (Ahn & Kang, 2023). For the primary outcome, the as-treated analysis was largely consistent with the main analysis but indicated a significantly lower risk of moving to long-term institutional care and dying at home in the analyses treating deaths as an event (HR=0.692 [0.532, 0.901]). In contrast, the per-protocol analysis consistently indicated that Komp users had a significantly lower risk of moving to long-term institutional care (HR = 0.255 [0.113-0.575]). For the secondary outcomes, both post hoc analyses showed higher use of formal care services among the treated. Significant effects were observed for home nursing care (as-treated IRR = 1.396 [1.132, 1.720]), number of short-term institutional care stays (as-treated IRR = 1.244 [1.025, 1.509]; per-protocol IRR = 1.447 [1.068, 1.960]), practical assistance (as-treated IRR = 1.400 [1.181, 1.659]; per-protocol IRR = 1.682 [1.300, 2.176]), and safety alarm activations (as-treated IRR = 1.741 [1.101, 2.752]). No significant effects were observed for the number of days in short-term institutional stays. Discussion This trial investigated the effects of Komp on the time spent living at home and the use of formal care services among older adults receiving long-term care at home. The findings did not support our hypothesis that access to Komp would prolong home-dwelling. On the contrary, the intervention group had a 10 to 16 percent higher risk of moving to a long-term care facility These findings were, however, not statistically significant. Given the reduced statistical power, these estimates cannot be taken as evidence of a true effect. The lack of significant effects in the primary outcome may be caused by the lack of any intervention effect, but are as likely to be explained by the limitations in statistical power and implementation. Due to delays in approval processes, the trial commenced six months later than planned, by which time several participants had become unavailable. As a result, only half of the intended devices were distributed, reducing the contrast between the groups. This reduced the statistical power of the trial, increasing the risk of actual effects going undetected. In addition, the intervention was implemented in a narrower form than initially intended. The trial design intended a dual use of Komp, by both family members and municipal care staff. As the municipal care services never adopted the use of Komp, we could measure only the effect of family use, which likely limits the effectiveness of the intervention. On the other hand, the absence of use from the health care services allowed us to isolate the effect of family use alone, and thus avoid conflating effects. Further, the results show that the intervention group experienced reduced usage of some formal care services, as we recorded a significant reduction in the duration of short-term institutional stays. On average, the intervention group spent approximately five fewer days in short-term care compared to controls. Notably, there were no differences between groups in hours of home nursing care, number of short-term institutional stays, hours of practical assistance, or safety alarm activations. This pattern suggests that the reduction in duration of short-term institutional stays was in fact a true reduction and not offset by increased use of other formal home-based services. However, as the intervention group showed a slight increase in long-term institutional stays, we cannot rule out that some substitution occurred between short-term and long-term care. As Komp users could bring their device to short-term care facilities, this might have facilitated contact with relatives during stays and thus supported participants to return home sooner. Meanwhile, Komp user also had the assurance of close family contact after leaving the short-term facilities, which may have provided additional support and made it easier to return home. On the other hand, the facilitated contact during the short-term stay might have allowed family members to monitor the users’ condition more closely. If so, they may have been more likely to advocate for long-term institutional care after admission to short-term care. Thus, the findings related to short-term care must be interpreted with caution. Given the lack of significant effects on other outcomes, there is also a risk that the observed reduction in short-term institutional stays may be significant at random. Our post hoc analyses complicate the interpretation of our results. We performed an as-treated analysis comparing those who received a Komp with controls, and a per-protocol analysis comparing participants with recorded Komp use to controls. In these analyses, we are no longer following a randomized design. Although they could represent a true effect of Komp, they could also be the result of selection bias. One possible explanation for the results is that frailer participants were more likely to accept the offer, as more active and technologically competent individuals have previously rejected Komp because they perceived it as too simplistic or even disempowering [ 32 ]. This could explain the higher use of home-based formal care services that we observed in both post hoc analyses. On the other hand, these findings could come as a consequence of the prolonged home-dwelling observed in the per-protocol analyses. Yet, the observed prolonged home-dwelling itself could be a result of selection bias, as participants with the most active family ties would be more likely to have registered Komp activity and to receive more informal support, as previous studies have shown [ 35 , 53 ]. Overall, these findings are best interpreted as results of selection bias rather than causal effects, and we therefore place greater weight in the ITT analyses. Still, we need more robust evidence to confidently conclude our findings. Although Komp was designed to enhance social connectedness and thus reduce reliance on formal care services, we cannot rule out the possibility of unintended harms. For example, 97 Komp devices were not used during the project period. This means that 97 participants accepted the offer, but did not have anyone reaching out through the connected app. In some of these cases, participants might have paid little attention to it, and no harm may have occurred. However, for others, the presence of an unused device could have acted as a constant reminder of limited social contact. One could, in this scenario, imagine that the device reinforced feelings of social isolation, and thus inducing harm to participants in the intervention group. One potential way to prevent this would have been to select participants who were more closely aligned with Komp’s intended target group. For example, only including individuals with digital active family members, only those with family members living outside of the municipality, or participants identified by the municipal home care services as suitable candidates. Such an approach, however, has its downsides, as it could exclude individuals who might otherwise benefit from the intervention, and thus create a double burden for those already at risk of being left out. To the best of our knowledge, no previous trials have evaluated the effects of DCTs on similar outcomes. We therefore cannot directly compare our results to prior trials. However, as our program theory suggests that Komp may promote social connectedness, and thus prolong home-dwelling, we turn to studies on DCTs impact on subjective outcomes. As previously mentioned, only two full-scale RCT have evaluated DCT as standalone interventions. Czaja et al.’s [ 30 ] computer system (PRISM) designed specifically for older adults were compared to Binder – a notebook that contained paper content similar to the content in PRISM. They found that PRISM significantly reduced loneliness and increased social support, and showed tendencies to reduce social isolation after six months. These improvements, however, were not maintained at 12 months. Czaja et al.’s [ 31 ] follow-up trial found significant reductions in loneliness, social isolation, and improvements in social support, quality of life, and health related quality of life. These improvements were only seen among participants in rural areas and senior housing locations – not for those residing in assisted living communities. The follow-up trial also included a mediation analysis, indicating that the reduction in social isolation and loneliness were key pathways to improved quality of life. In sum, these trials indicate that DCTs may facilitate social connectedness in some contexts. Turning to interventions with clinical content, a meta-analysis of six RCTs on smart-phone based video calls and computer-based training aimed at reducing loneliness in older adults showed little to no effect [ 27 ]. Thus, the evidence for significant large-scale effects of DCTs remains limited. From an economic perspective, this raises questions about the long-term sustainability of widespread DCT implementation in health and social care. However, when viewing the evidence of DCTs improving social connectedness and quality of life [ 30 , 31 ], in light of our findings of potential reductions in formal care use and associated cost savings, a more tangible picture takes form. Specifically, the observed reduction in the duration of short-term institutional stays equals $ 1,500 per participant in a Norwegian context. Seen together, these findings suggest that while DCTs may hold promise as supportive tools in health and social care, the current knowledge base is still too weak to draw firm conclusions. Our trial is the first RCT to evaluate the effect of a DCT on the use of formal care services among older adults. We have done so with a relatively large sample size, which proved necessary to be able to detect any differences between groups, in what turned out to be a trial marked by non-compliance. Moreover, the use of high-quality registry data available from Statistics Norway, allowed us to detect any skewness between groups on a broad range of variables with high accuracy. As our trial was conducted in a realistic municipal care setting within a conservative ITT approach, we have also enhanced the external validity and likely reduced the risk of unforeseen factors influencing the effects in a potential implementation at scale. Our study has several limitations. First, the uptake of our intervention was low: of the 516 participants in the intervention group, 150 (29%) received a Komp device. This corresponds to 42 percent of those offered Komp (n = 359). Among those who received a device, 53 participants were registered with use. This corresponds to 10 percent of the entire intervention group, 15 percent of those offered Komp, and a third of those who accepted one. Among those who were registered with Komp use, the usage differed substantially. As we follow an ITT-approach, the estimates should be interpreted as the effect of “belonging to the intervention group” rather than the isolated effect of the intervention [ 54 ]. While this reflects what could be expected in a large-scale implementation, it likely underestimates the potential positive or negative effects of the intervention itself. Second, the statistical power of the trial was limited. Calculations showed that we would not likely be able to detect smaller effects, which means that the lack of significant findings in both the primary and secondary outcomes should be interpreted with caution, as meaningful effects may have gone undetected. Third, there is a potential SUTVA violation, as two control group members resided with participants in the intervention group who received a Komp device. One control group member was also mistakenly offered a Komp. Given the low number of contaminated participants, it is unlikely to have influenced the results. Fourth, there is always a risk of unobserved variables having influenced the results. However, access to comprehensive background data reduces endogeneity concerns, as it let us adjust for differences in wealth and marital status. Lastly, the generalizability of our findings may be limited. Previous trials have demonstrated that contextual factors matters, as a DCT only proved effective among older adults living outside assisted living communities [ 31 ]. Similarly, we must account for factors that restrict our findings to the context of our trial, e.g. that it was conducted in an urban area. We recommend that future studies continue to evaluate the effect of DCTs on formal care use in both similar and other contexts. These studies should also plan for investigating substitution effects between formal care services. Meanwhile, more trials assessing standalone DCT interventions, which are easier to implement at large scale, are needed. Additionally, it would be valuable for future research to explore more targeted approaches, as mentioned previously, which could both help prevent unintended harms and better reflect large scale implementations. Further, we recommend evaluations of how Komp, and other DCTs designed for individuals with less digital competence, affect users’ social contact and well-being, to better understand the underlying mechanisms behind our findings. Finally, we recommend researchers to consider quasi-experimental approaches, such as propensity score matching, in trials where compliance is low. In sum, more robust evaluations of DCTs across a broader range of outcomes are necessary to strengthen the evidence base for the digitalization of welfare states. Conclusion In this trial, the results indicate that Komp, a digital communication technology, does not prolong home-dwelling among Norwegian older adults above 67 years. The results further indicate that Komp reduces reliance on some formal care services, with short-term institutional stays reduced with an average of five days in the intervention group. We recommend further research to investigate how Komp affects users’ quality of life and well-being, to better understand the underlying mechanisms. We also recommend more robust evaluations of other types of digital communication technologies on various outcomes to strengthen the evidence base for the digitalization of welfare services. Abbreviations OECD Organization for Economic Co-operation and Development DCT Digital Communication Technology RCT Randomized Controlled Trial PRISM Personal Reminder Information and Social Management ITT Intention-to-Treat CONSORT Consolidated Standards of Reporting Trials ADL Activities of Daily Living SD Standard Deviation HR Hazard Ratio SHR Sub Hazard Ratio ZINB Zero-Inflated Negative Binomial ZIP Zero-Inflated Poisson AIC Akaike Information Criterion BIC Bayesian Information Criterion IRR Incidence Rate Ratio OR Odds Ratio SUTVA Stable Unit Treatment Value Assumption Declarations Ethics approval and consent to participate We have institutional approval from the GDPR contact at the Faculty of Social Sciences, Oslo Metropolitan University. Ethical approval was obtained from the Regional Committees for Medical and Health Research Ethics (REK, 516796) on October 10, 2022, and from the Norwegian Agency for Shared Services in Education and Research (SIKT, 332463) on December 21, 2022. In accordance with the approval from REK, active informed consent was not required for the use of anonymized background data [55]. All participants were informed about the study and were given the opportunity to opt out of the use of their data. Informed consent to receive the intervention was obtained from all participants who were offered the intervention by healthcare professionals. All procedures were conducted in accordance with the ethical principles of the Declaration of Helsinki [56]. Consent for publication Not applicable. Availability of data and materials The data that support the findings of this study are not publicly available without further approval from Oslo Municipality and SIKT. The do-files can be accessed by contacting the authors. The trial is registered at ClinicalTrials.gov (NCT05919355). The trial protocol is publicly available [57]. Competing interests The authors declare that they have no competing interests. Funding This work was supported by the Research Council of Norway [grant number 331810]. Authors’ contributions EBR led the research project. EBR obtained funding with support from KAW and BSAT, and data with support from KAW, BSAT and KE. KE led the work with the present paper, analyzed and interpreted the data, and drafted the manuscript. KE, IMH, EBR, BSAT, AGT and KAW contributed to critically revising the paper and agreed to be accountable for all aspects of the work. Acknowledgements The authors would like to thank Anna-Stina Slattum, Research & Impact manager in Abilia, who led the BoVel-project; Professor Anne Lund and Martin Vinther Bavngaard for their contribution to the development of, and participation in the project; the local project coordinators for making it possible to implement Komp; and Anett Günther Warnstrøm and Irene Oksdøl for their assistance related to the Gerica data, and initiating the idea to study safety alarm data. This work would not have been possible without your contributions. References Nasjonale befolkningsframskrivinger [https://www.ssb.no/befolkning/befolkningsframskrivinger/statistikk/nasjonale-befolkningsframskrivinger/artikler/vi-blir-flere-men-ogsa-mye-eldre] OECD: Health at a Glance: Europe 2024: State of Health in the EU Cycle : OECD publishing; 2024. HOD: Fellesskap og meistring: Bu trygt heime . In . Edited by Services MoHaC; 2022-2023. Haak M, Fänge A, Iwarsson S, Dahlin Ivanoff S: Home as a signification of independence and autonomy: experiences among very old Swedish people . Scandinavian journal of occupational therapy 2007, 14 (1):16-24. Tate K, Bailey S, Deschenes S, Grabusic C, Cummings GG: Factors influencing older persons’ transitions to facility-based care settings: a scoping review . The Gerontologist 2023, 63 (7):1211-1227. Luppa M, Luck T, Weyerer S, König H-H, Brähler E, Riedel-Heller SG: Prediction of institutionalization in the elderly. A systematic review . Age and ageing 2010, 39 (1):31-38. Blakely KK, Baker NR: Loneliness in community-dwelling, older adults: an integrative review . The Journal for Nurse Practitioners 2023, 19 (1):104390. Steiner JF, Ross C, Stiefel M, Mosen D, Banegas MP, Wall AE, Martin C, Kelly TS, Paolino AR, Zeng C: Association between changes in loneliness identified through screening and changes in depression or anxiety in older adults . Journal of the American Geriatrics Society 2022, 70 (12):3458-3468. Donovan NJ, Wu Q, Rentz DM, Sperling RA, Marshall GA, Glymour MM: Loneliness, depression and cognitive function in older US adults . International journal of geriatric psychiatry 2017, 32 (5):564-573. Griffin SC, Mezuk B, Williams AB, Perrin PB, Rybarczyk BD: Isolation, not loneliness or cynical hostility, predicts cognitive decline in older Americans . Journal of aging and health 2020, 32 (1-2):52-60. Lambotte D, De Donder L, Van Regenmortel S, Fret B, Dury S, Smetcoren A-S, Dierckx E, De Witte N, Verté D, Kardol MJM: Frailty differences in older adults’ use of informal and formal care . Archives of Gerontology and Geriatrics 2018, 79 :69-77. Bigonnesse C, Chaudhury H: The Landscape of “Aging in Place” in Gerontology Literature: Emergence, Theoretical Perspectives, and Influencing Factors . Journal of Aging and Environment 2020, 34 (3):233-251. Emlet CA, Moceri JT: The Importance of Social Connectedness in Building Age-Friendly Communities . Journal of Aging Research 2012, 2012 (1):173247. Steels S: Key characteristics of age-friendly cities and communities: A review . Cities 2015, 47 :45-52. Bavngaard MV, Lund A, Thordardottir B, Rasmussen EB: The Uses and Experiences of Synchronous Communication Technology for Home-Dwelling Older Adults in a Home Care Services Context: Qualitative Systematic Review . Journal of Medical Internet Research 2024, 26 . Digital Technologies Can Help Older Persons Maintain Healthy, Productive Lives [https://www.un.org/en/un-chronicle/digital-technologies-can-help-older-persons-maintain-healthy-productive-lives] ITU: Ageing in a digital world - from vulnerable to valuable : ITU Publications; 2021. DesChâtelets JR, Khowaja AR, Mechelse K, Koning H, Ventresca D: Exploring the access and use of social technologies by older adults in support of their mental health during the COVID-19 pandemic: A rapid review . Canadian Journal on Aging/La Revue canadienne du vieillissement 2023, 42 (4):728-743. Ambagtsheer R, Borg K, Townsin L, De Plaza MP, O'Brien L, Kunwar R, Lawless M: The effectiveness of technology interventions in reducing social isolation and loneliness among community-dwelling older people: A mixed methods systematic review . Archives of Gerontology and Geriatrics Plus 2024, 1 (1). Yutong T, Yan Z, Qingyun C, Lixue M, Mengke G, Shanshan W: Information and Communication Technology Based Integrated Care for Older Adults: A Scoping Review . Int J Integr Care 2023, 23 (2):2. Noone C, McSharry J, Smalle M, Burns A, Dwan K, Devane D, Morrissey E: Video calls for reducing social isolation and loneliness in older people: a rapid review . Cochrane Database of Systematic Reviews 2020, 5 . Petersen B, Khalili-Mahani N, Murphy C, Sawchuk K, Phillips N, Li KZ, Hebblethwaite S: The association between information and communication technologies, loneliness and social connectedness: A scoping review . Frontiers in Psychology 2023, 14 :1063146. Choi HK, Lee SH: Trends and Effectiveness of ICT Interventions for the Elderly to Reduce Loneliness: A Systematic Review . Healthcare 2021, 9 (3):293. Forsman AK, Nordmyr J, Matosevic T, Park A-L, Wahlbeck K, McDaid D: Promoting mental wellbeing among older people: technology-based interventions . Health promotion international 2018, 33 (6):1042-1054. Hansen T, Nes RB, Hynek K, Nilsen TS, Reneflot A, Stene-Larsen K, Tornes RA, Bidonde J: Tackling social disconnection: an umbrella review of RCT-based interventions targeting social isolation and loneliness . BMC Public Health 2024, 24 (1):1917. Heins P, Boots LM, Koh WQ, Neven A, Verhey FR, de Vugt ME: The effects of technological interventions on social participation of community-dwelling older adults with and without dementia: a systematic review . Journal of clinical medicine 2021, 10 (11):2308. Jin W, Liu Y, Yuan S, Bai R, Li X, Bai Z: The effectiveness of technology-based interventions for reducing loneliness in older adults: A systematic review and meta-analysis of randomized controlled trials . Frontiers in Psychology 2021, 12 :711030. Teo AR, Benton MC, Hooker ER, Zaccari B, Hidalgo NJ, Newell S, Tuepker A, Greaves DW, Nicosia FM: Effect of telehealth yoga on loneliness and social isolation among rural older adults: a randomized controlled trial . Aging & Mental Health 2025:1-9. Jarvis MA, Padmanabhanunni A, Chipps J: An evaluation of a low-intensity cognitive behavioral therapy mHealth-supported intervention to reduce loneliness in older people . International journal of environmental research and public health 2019, 16 (7):1305. Czaja SJ, Boot WR, Charness N, Rogers WA, Sharit J: Improving Social Support for Older Adults Through Technology: Findings From the PRISM Randomized Controlled Trial . The Gerontologist 2018, 58 (3):467-477. Czaja SJ, Charness N, Rogers WA, Sharit J, Moxley JH, Boot WR: The benefits of technology for engaging aging adults: Findings from the PRISM 2.0 trial . Innovation in aging 2024, 8 (6). Rasmussen EB, Askheim C, Oppedal B, Haldar M: Varm teknologi mot ensomhet blant eldre? En skriptanalyse av kommunikasjonsløsningen KOMP . 2021. Akhtar SA: Socially connected while apart: the use of technology to increase social connection between nursing home residents and their relatives . Frontiers in Public Health 2024, 12 . Rasmussen EB, Akhtar S: Enkel og trygg pårørendekontakt: Forskningsrapport fra et pandemiprosjekt om bruk av sosial teknologi for eldre på langtidshjem . Skriftserien 2022:71-71. Berg A, Myren GES: Digital teknologi for sosial kontakt: En deskriptiv studie av de eldste eldres erfaringer med KOMP . Nordisk tidsskrift for helseforskning 2021, 17 (2). Funnell SC, Rogers PJ: Purposeful program theory: Effective use of theories of change and logic models : John Wiley & Sons; 2011. Hopewell S, Chan A-W, Collins GS, Hróbjartsson A, Moher D, Schulz KF, Tunn R, Aggarwal R, Berkwits M, Berlin JA: CONSORT 2025 statement: updated guideline for reporting randomised trials . The Lancet 2025, 405 (10489):1633-1640. Christensen K, Wærness K: Long-term care services in Norway: a historical sociological perspective . In: The Routledge handbook of social care work around the world. Routledge; 2017: 15-28. Colombo F, Llena-Nozal A, Mercier J, Tjadens F: Help wanted? Providing and paying for long-term care. Paris: OECD; 2011. Antall personer på venteliste for langtidsplass på sykehjem [https://www.helsedirektoratet.no/statistikk/venteliste-for-langtidsplass-pa-sykehjem] Sykehjem og hjemmetjenesten i Norge [https://www.ssb.no/helse/helsetjenester/artikler/sykehjem-og-hjemmetjenesten-i-norge] Gjennomsnittlig botid på langtidsopphold i institusjon [https://www.helsedirektoratet.no/rapporter/kommunale-helse-og-omsorgstjenester-i-2023/utvalgte-temaer/gjennomsnittlig-botid-pa-langtidsopphold-i-institusjon] Broglio K: Randomization in clinical trials: permuted blocks and stratification . Jama 2018, 319 (21):2223-2224. StataCorp: Stata Statistical Software: Release 18 . In . : StataCorp LLC.; 2023. Cox DR: Regression models and life ‐tables . Journal of the Royal Statistical Society: Series B (Methodological) 1972, 34 (2):187-202. Kleinbaum DG, Klein M: Survival analysis a self-learning text : Springer; 1996. Ranganathan P, Pramesh C: Censoring in survival analysis: potential for bias . Perspectives in clinical research 2012, 3 (1):40. Fine JP, Gray RJ: A proportional hazards model for the subdistribution of a competing risk . Journal of the American statistical association 1999, 94 (446):496-509. Hilbe JM: Modeling count data . In: International encyclopedia of statistical science. Springer; 2011: 836-839. Harris T, Hilbe JM, Hardin JW: Modeling count data with generalized distributions . The Stata Journal 2014, 14 (3):562-579. Hardin JW, Hilbe JM: Regression models for count data based on the negative binomial (p) distribution . The Stata Journal 2014, 14 (2):280-291. Burnham KP, Anderson DR: Multimodel inference: understanding AIC and BIC in model selection . Sociological methods & research 2004, 33 (2):261-304. Oppedal B, Askheim C, Haldar M: Erfaringer med KOMP som velferdsteknologi . Skriftserien 2019:56-56. Feinman RD: Intention-to-treat. What is the question? Nutrition & Metabolism 2009, 6 (1):1. NESH: Guidelines for Research Ethics in the Social Sciences and the Humanities . In . , 5th edn: National Committee for Research Ethics in the Social Sciences and the Humanities (NESH); 2022. WMA: World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects . Jama 2013, 310 (20):2191-2194. Research protocol for the BoVEL project [https://static1.squarespace.com/static/6315e99c979ac470513caea9/t/6405b982ee0eef5b7d402eb6/1678096770598/Forskningsprotokoll+BoVEL.pdf] Tables Table 1 to 5 are available in the Supplementary Files section. Additional Declarations No competing interests reported. Supplementary Files CONSORT2025checklist.docx Tables.docx Cite Share Download PDF Status: Published Journal Publication published 11 Apr, 2026 Read the published version in BMC Geriatrics → Version 1 posted Editorial decision: Revision requested 20 Feb, 2026 Reviews received at journal 01 Feb, 2026 Reviews received at journal 26 Jan, 2026 Reviewers agreed at journal 20 Jan, 2026 Reviewers agreed at journal 19 Jan, 2026 Reviewers agreed at journal 19 Jan, 2026 Reviewers invited by journal 07 Jan, 2026 Editor assigned by journal 07 Jan, 2026 Editor invited by journal 26 Dec, 2025 Submission checks completed at journal 23 Dec, 2025 First submitted to journal 23 Dec, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8378683","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":572557510,"identity":"b5d2955a-bb5e-4308-80c6-e0ee6a9910bf","order_by":0,"name":"Kristoffer Eikemo","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABH0lEQVRIie2SsWrDMBCGTxjiRY3XMwb3FRQMmUrzIF1kAp6q3VNrMKiLaOeQl/DU2cbQSQ+Q4sWQF1DpkqHQyimEDHa6dtA33CGhD+4/BOBw/EMQSGFbHBxPvLbFtxf9X0oNSVicFGoLH/qkYjkpMCjILyvhti0/Tc4wiGrS9/rmLgj3reH5a7yCq2ZswGieSqw1w/CZe4zvMrHZZhly3SUU5ms2osSUSGjk1wPTMENuWlF190tIZZcqoEscV0rTSIYrDf6Bm29RvevEWOVxSokoKXBQGIUZ8F0tqh1laBVOJ5RQEYnaZkFNShthLTbqN8tCtTQZy4La35t82Jjymo/D26148dvWmLy79p/UYmxjZxy/wTne5fcOh8PhmOYHuK1m7urczEoAAAAASUVORK5CYII=","orcid":"","institution":"OsloMet – Oslo Metropolitan University","correspondingAuthor":true,"prefix":"","firstName":"Kristoffer","middleName":"","lastName":"Eikemo","suffix":""},{"id":572557511,"identity":"4e4f388e-0fc6-4d82-931a-72add02745a2","order_by":1,"name":"Ira Malmberg-Heimonen","email":"","orcid":"","institution":"OsloMet – Oslo Metropolitan University","correspondingAuthor":false,"prefix":"","firstName":"Ira","middleName":"","lastName":"Malmberg-Heimonen","suffix":""},{"id":572557512,"identity":"624417fe-1b59-4d85-8b6a-2a237e7fc57b","order_by":2,"name":"Erik Børve Rasmussen","email":"","orcid":"","institution":"OsloMet – Oslo Metropolitan University","correspondingAuthor":false,"prefix":"","firstName":"Erik","middleName":"Børve","lastName":"Rasmussen","suffix":""},{"id":572557513,"identity":"54a342ac-f84c-45f2-89ca-1ba6a5b2e9fd","order_by":3,"name":"Björg Sigridur Anna Thordardottir","email":"","orcid":"","institution":"OsloMet – Oslo Metropolitan University","correspondingAuthor":false,"prefix":"","firstName":"Björg","middleName":"Sigridur Anna","lastName":"Thordardottir","suffix":""},{"id":572557514,"identity":"5f79498c-a70b-41f0-b8e6-611822f57877","order_by":4,"name":"Anne Grete Tøge","email":"","orcid":"","institution":"OsloMet – Oslo Metropolitan University","correspondingAuthor":false,"prefix":"","firstName":"Anne","middleName":"Grete","lastName":"Tøge","suffix":""},{"id":572557515,"identity":"e68c8c91-bc5c-4420-b5bb-ff992c03df32","order_by":5,"name":"Kjetil A. van der Wel","email":"","orcid":"","institution":"OsloMet – Oslo Metropolitan University","correspondingAuthor":false,"prefix":"","firstName":"Kjetil","middleName":"A. van der","lastName":"Wel","suffix":""}],"badges":[],"createdAt":"2025-12-16 17:23:35","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8378683/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8378683/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12877-026-07322-z","type":"published","date":"2026-04-11T15:59:30+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":100377969,"identity":"79904f38-5b8c-4edd-b699-20983758b3a6","added_by":"auto","created_at":"2026-01-16 08:49:15","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":264707,"visible":true,"origin":"","legend":"","description":"","filename":"Eikemoetalmanuscript3.docx","url":"https://assets-eu.researchsquare.com/files/rs-8378683/v1/485999dd65b0434a15e5923c.docx"},{"id":100342319,"identity":"32b34c3b-7397-4939-aafc-e469721d5d9e","added_by":"auto","created_at":"2026-01-16 00:05:58","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9059,"visible":true,"origin":"","legend":"","description":"","filename":"f1e4de440f7a424b91e3871b668c0318.json","url":"https://assets-eu.researchsquare.com/files/rs-8378683/v1/375ac6a1322f427608fb1e35.json"},{"id":100342323,"identity":"0c5a63f6-9daa-44f8-b5fb-6f87b9d0a07d","added_by":"auto","created_at":"2026-01-16 00:05:58","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":33564,"visible":true,"origin":"","legend":"","description":"","filename":"CONSORT2025checklist.docx","url":"https://assets-eu.researchsquare.com/files/rs-8378683/v1/0055601230f2bfff003d7fc1.docx"},{"id":100373804,"identity":"40fd9fa3-3fe7-4e1a-9667-4e1507fe0231","added_by":"auto","created_at":"2026-01-16 08:19:10","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":165800,"visible":true,"origin":"","legend":"","description":"","filename":"f1e4de440f7a424b91e3871b668c03181enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8378683/v1/d0ef20848e40cde53acb488b.xml"},{"id":100342321,"identity":"27c11572-a44f-4b7c-906e-050a8ca1a866","added_by":"auto","created_at":"2026-01-16 00:05:58","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":34355,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8378683/v1/904f9a14d234e40a5cb0e3fa.png"},{"id":100342327,"identity":"71018c83-a733-41bc-b16a-9838c7d4b98d","added_by":"auto","created_at":"2026-01-16 00:05:58","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29220,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8378683/v1/d38d8ed32d182a9ac60c3d45.png"},{"id":100342325,"identity":"c281cd75-1244-40a2-b7c6-a03b6d4469e0","added_by":"auto","created_at":"2026-01-16 00:05:58","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":66352,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8378683/v1/d6f0815d4fa5e7da24ec5c55.png"},{"id":100342330,"identity":"6fe909b2-1daf-41c5-b8cf-e783e4694c79","added_by":"auto","created_at":"2026-01-16 00:05:58","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":32407,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8378683/v1/93340108b20e8b06118e4396.png"},{"id":100373829,"identity":"c34d3895-00c2-4e39-8ecc-f18f3d680a82","added_by":"auto","created_at":"2026-01-16 08:19:21","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":13775,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8378683/v1/d95eb4e38cab226c0084015d.png"},{"id":100342329,"identity":"eeb911a0-7d00-4f37-b904-9560551bf5e1","added_by":"auto","created_at":"2026-01-16 00:05:58","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":13033,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8378683/v1/251978f4c0131b47cc794a97.png"},{"id":100342332,"identity":"b8122e49-f8a0-4c9e-8620-e0e291f5d40b","added_by":"auto","created_at":"2026-01-16 00:05:58","extension":"xml","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":163090,"visible":true,"origin":"","legend":"","description":"","filename":"f1e4de440f7a424b91e3871b668c03181structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8378683/v1/b837934c24f2750978645f2c.xml"},{"id":100342333,"identity":"4c156fd0-dc43-4c32-8128-d6c86e88c5c0","added_by":"auto","created_at":"2026-01-16 00:05:59","extension":"html","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":179468,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8378683/v1/e7bbad78627c6a36c166a9aa.html"},{"id":100342317,"identity":"3bd426c0-d7c7-4ae8-bbbf-83b72b1f7dde","added_by":"auto","created_at":"2026-01-16 00:05:58","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":34355,"visible":true,"origin":"","legend":"\u003cp\u003eLogic model (based on the pipeline-model [36])\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8378683/v1/142b3f825e03fa599f6eeb8a.png"},{"id":100373382,"identity":"5967785e-71ac-44ff-8d86-d68776bb13ad","added_by":"auto","created_at":"2026-01-16 08:14:12","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":29220,"visible":true,"origin":"","legend":"\u003cp\u003eCONSORT 2025 Flow Diagram [37]\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8378683/v1/65e1e424f7bcc4b9ed17cf92.png"},{"id":100342326,"identity":"50e7e300-04b6-4697-b414-6d4205866570","added_by":"auto","created_at":"2026-01-16 00:05:58","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":66352,"visible":true,"origin":"","legend":"\u003cp\u003eStatistical power calculations for primary outcome\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8378683/v1/0adead93e02d51adca88b611.png"},{"id":106810691,"identity":"e6bc2b24-128d-4f92-8cec-dc6c276e38a3","added_by":"auto","created_at":"2026-04-13 16:16:32","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2785532,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8378683/v1/73c311f4-6b48-42e1-9565-2bbe19e3e659.pdf"},{"id":100378076,"identity":"520a5103-5fa2-4bae-be29-5ae711f5ff4e","added_by":"auto","created_at":"2026-01-16 08:50:00","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":33564,"visible":true,"origin":"","legend":"","description":"","filename":"CONSORT2025checklist.docx","url":"https://assets-eu.researchsquare.com/files/rs-8378683/v1/72da7ab43d00be1fabcdad65.docx"},{"id":100378817,"identity":"7b143bb4-af60-4c03-80e1-d07594124b13","added_by":"auto","created_at":"2026-01-16 09:00:27","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":29150,"visible":true,"origin":"","legend":"","description":"","filename":"Tables.docx","url":"https://assets-eu.researchsquare.com/files/rs-8378683/v1/739c494d6acf5dc856ca3bdb.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"The effect of digital communication technology on older adults’ formal care use in Norway: a randomized controlled trial","fulltext":[{"header":"Background","content":"\u003cp\u003eLike in all European countries, Norway’s population is ageing. By 2050, one in five Norwegians will be over the age of 70 [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e], similar to projections across the OECD [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. This demographic shift increases pressure on already strained health and care services. Norwegian policymakers have responded to the challenge by promoting a strategy of ‘ageing in place’, i.e. that older adults primarily should receive long-term care in their own homes rather than in long-term care facilities [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eWhile ageing in place may promote independence and autonomy among many older adults [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e], it is not feasible for all. The reasons why some older adults transition to long-term care are multifaceted and complex. Functional impairments and caregiver burden have been identified as key factors [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e], along with low social contact [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Low social contact is further linked to lower quality of life [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e], cognitive decline [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e], and poorer overall physical health [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e], which are further associated with greater reliance on formal care services [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Promoting social connectedness is therefore recognized as essential to ageing in place [\u003cspan additionalcitationids=\"CR13\" citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e–\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eDigital communication technologies (DCTs) that allow for synchronous communication with formal caregivers, family, and friends [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e], have been highlighted as promising tools to support healthy ageing [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. Previous research indicates that DCTs may enhance social connectedness, reduce loneliness and improve well-being among older adults by facilitating communication with their social network [\u003cspan additionalcitationids=\"CR19 CR20 CR21\" citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e–\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. However, despite promising findings, we lack studies with robust causal designs to evaluate their effects [\u003cspan additionalcitationids=\"CR24 CR25 CR26\" citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e–\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eMany randomized controlled trials (RCTs) have evaluated DCT interventions among home-dwelling older adults on subjective outcomes such as loneliness and social isolation. However, most have focused on clinical content delivered through technology, such as telehealth [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e] or cognitive therapy [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e], rather than the social function of the intervention itself. To our knowledge, only two RCTs have examined DCTs as standalone interventions aimed at facilitating social connectedness among older adults. Czaja et al. [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e] conducted a multisite trial in Florida and Georgia, United States, among individuals aged 65 or older living alone in independent housing. Participants received either PRISM, a computer system designed for older adults that included communication tools (e.g., photo sharing), or a control condition that provided similar information in a paper binder. Results showed that participants receiving PRISM experienced a significant increase in perceived social support and well-being, as well as a reduction in loneliness, compared to controls. In a follow-up RCT, Czaja et al. [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e] evaluated PRISM 2.0, an updated tablet-based version with expanded social features, among individuals aged 65 years and above living in senior housing, rural locations, and assisted living communities in Florida and Georgia. The intervention group was compared to a control group using a tablet with standard applications. Both the PRISM 2.0 and standard tablet groups living in senior housing and rural areas reported reductions in loneliness and social isolation and improvements in quality of life and social support. To the best of our knowledge, no RCT-studies has investigated the effect of the social function of DCTs on objective service outcomes, such as formal care use and transitions to long-term care.\u003c/p\u003e \u003cp\u003eThe present study addresses these research gaps by evaluating the effect of the DCT device \u003cem\u003eKomp\u003c/em\u003e on formal care use among home-dwelling older adults aged 67 years and above. Komp was designed for individuals who are unfamiliar with, or struggle with, conventional digital technologies [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. All interaction beyond turning the device on or off and adjusting the volume is managed by the user’s selected social network through a connected app. This network can initiate video calls and send photos and text messages, which appear automatically on the Komp screen. Findings from previous research on Komp indicate that this tool does increase feelings of safety, support meaningful digital contact between users and their social network, and thus facilitates informal care [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan additionalcitationids=\"CR33 CR34\" citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e–\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eBased on the indications that Komp does support social contact and increase feelings of safety, our trial’s program theory proposes the following mechanisms. First, increased social contact and perceived safety may mitigate factors such as loneliness and insecurity that increase the likelihood of moving to a long-term care facility and greater service needs. Second, by strengthening informal care availability and everyday reassurance, Komp may delay transitions to long-term care and reduce reliance on home-based services. Thus, although not providing clinical content, we hypothesize that Komp prolongs home-dwelling and reduces reliance on formal care services among older adults.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIn accordance with the trial protocol (NCT05919355), our primary objective is to assess whether access to Komp prolongs the service recipients’ home-dwelling by delaying moving to long-term care facilities. Our secondary objective is to evaluate whether access to Komp reduces the use of other formal care services: home nursing care, short-term institutional care, practical assistance, and safety alarms. The original implementation plan also included municipal care staff use of Komp, which ultimately did not occur due to delays. This deviation has implications for our program theory and expected outcomes. While, according to our program theory, the intervention’s ability to address social needs and strengthen informal care should still influence the use of formal services, the anticipated effects are likely to be weaker when the device is not simultaneously adopted by formal care providers.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e\u003c/p\u003e \n\n \n\n\n\n\n\n "},{"header":"Methods","content":"\u003ch2\u003eStudy Design\u003c/h2\u003e\n\u003cp\u003eThis study is designed as a field trial, with a parallel group assignment, and an intention-to-treat (ITT) approach. It is conducted as part of the BoVel project \u0026ndash; a collaboration between Oslo Metropolitan University, Oslo Municipality, and Abilia, which is a private company that develops and sells Komp. Recruitment began on December 21, 2022, with data collection spanning from October 13, 2022, to March 1, 2025. Our reporting follows the CONSORT checklist [37] as far as applicable.\u003c/p\u003e\n\u003ch2\u003eStudy Context\u003c/h2\u003e\n\u003cp\u003eOur study was conducted in three boroughs of Oslo, Norway: Nordstrand, St. Hanshaugen, and \u0026Oslash;stensj\u0026oslash;. In international comparison, Norway has a comprehensive welfare state with redistributive benefits and universal health services, including long-term care services for older adults [38]. Long-term care is largely a public responsibility, with municipalities covering 90 percent of the costs [39]. Many municipalities in Norway, including Oslo, have waiting lists for long-term care facilities [40], making home-based services, such as home nursing care, the primary means of supporting older adults in need of care [41]. Nursing homes are generally reserved for those with the most complex needs, as reflected in the fact that the average length of stay in a Norwegian nursing home is approximately two years [42].\u003c/p\u003e\n\u003ch2\u003eParticipants\u003c/h2\u003e\n\u003cp\u003eWe identified potential participants among 9,317 municipal health and care service records from three Oslo boroughs. Eligible participants had to (i) be 67 years or older; (ii) dwell in a private home (i.e., not in a permanent care or nursing home); (iii) have a registered address in one of three participating boroughs of Oslo, Norway; and (iv) be registered recipients of municipal home nursing care. Individuals were excluded if they had full function in all of the three following Activities of Daily Living (ADL) measures, as assessed by the services: outdoor mobility, memory, and cooking. They were excluded because having full function in all these three areas meant a very low risk of moving to long-term care during the trial, and thus they were outside the intervention\u0026rsquo;s target group. This exclusion process resulted in 1,170 eligible participants. These were sent written information about the project, with the option to decline participation. Information was first distributed electronically, followed by a postal version if the electronic version was not opened within 14 days. In total, 59 individuals declined participation, leaving 1,111 participants. Figure 1 demonstrate the flow chart for the study.\u003c/p\u003e\n\u003ch2\u003eAssignment to intervention and control group\u003c/h2\u003e\n\u003cp\u003eWithin RCTs, eligible participants must have a defined and equal probability of being assigned to the intervention, to create groups that are comparable with respect to both measured and unmeasured characteristics [43]. In this study, eligible participants (n = 1,111) were identified from municipal records as of December 21, 2022, with the data backdated to October 13, 2022. To ensure equal probability of allocation, we applied a form of permuted block randomization [43], treating each of the three boroughs as a separate block. Separate lists were created for each borough, and participants were randomly ordered by generating random number sequences in Stata. Local project coordinators offered Komp to names on the randomly ordered lists, working from the top down. We had 60 Komps available in St. Hanshaugen, and 120 in each of the two larger boroughs. Because many participants were not willing to receive Komp or not possible to reach during the delivery period, we reached half the lists and stopped the offers before all available Komps (n = 300) had been distributed. At this point, we had made an offer to 32 percent of the 1,111 participants.\u003c/p\u003e\n\u003cp\u003eIn cases where persons in the intervention group did not receive Komp, we registered them as no-shows. In accordance with the ITT approach, we kept them (n = 157) in the intervention group, as the control group would also include unavailable individuals. This resulted in 516 participants in the intervention group and 595 in the control group. Blinding was not possible\u0026nbsp;due to the nature of the intervention. Although we cannot rule out the possibility that knowledge of group assignment influenced how home care services were delivered, we have no indications that this occurred. Qualitative interviews with home care professionals do not suggest any systematic differences in follow-up or service provisions [32], and any such differences would likely have been detected.\u003c/p\u003e\n\u003cp\u003eAfter participants were offered Komp, they had about 14 days to accept. Those who accepted (n = 150) received a Komp device within one week, which they could use freely. Installation was carried out by project coordinators or other trained staff, and participants received written information about technical support. In most cases, the offer was made in person, accompanied by a demonstration of the device, and, when possible, with relatives present. Most devices were delivered over a period of 10 months, except for two cases receiving Komp 21 months after distribution commenced. At any time, participants who had received a Komp could decide to stop and return the Komp to the municipality. Twelve participants returned the device earlier, after on average, using it for 8.5 months. Participants who declined the offer continued to receive services as usual (n = 209). In accordance with the ITT approach, all individuals offered Komp were included in the intervention group. Control group members continued to receive home care services as usual.\u003c/p\u003e\n\u003ch2\u003eIntervention\u003c/h2\u003e\n\u003cp\u003eKomp is a DCT developed by the Norwegian company No Isolation to support social contact between older adults and their social network. The device consists of a screen with a single button for turning it on and off, as well as adjusting the volume. The user\u0026rsquo;s social network uses a connected app to send photos, messages, and make video calls. Calls are automatically connected after ten seconds unless the user turns the device off. While off, no calls go through, but a small blinking light indicates incoming calls. An important clarification is that the Komp user cannot initiate contact through the Komp screen; they can only receive. The design is intentionally made suitable for individuals with limited technological competence. However, it also limits opportunities for interaction, making Komp valuable only if others actively engage. See Akhtar [33] for more information on Komp.\u003c/p\u003e\n\u003ch2\u003eSuccess of Randomization\u003c/h2\u003e\n\u003cp\u003eWe estimated significance of differences using two-sided T-tests for continuous variables and chi-squared tests for categorical variables (Table 1). The results show that the groups are balanced apart from wealth and marital status: Participants in the intervention group had slightly higher wealth, while a higher proportion of participants in the control group were married. We adjust for these differences in the effect analyses.\u003c/p\u003e\n\u003cp\u003eTable 1: Baseline demographics\u003c/p\u003e\n\u003ch2\u003eMeasures\u003c/h2\u003e\n\u003ch3\u003ePrimary outcome\u003c/h3\u003e\n\u003cp\u003eThe primary outcome was time spent living at home, measured in days from the start of the project (October 13, 2022) until either moving to a long-term care facility or the end of the study period (March 1, 2025).\u003c/p\u003e\n\u003ch4\u003eTime spent living at home\u003c/h4\u003e\n\u003cp\u003eData was obtained from the municipal journal system, Gerica. The date of moving was defined as the start date of the service \u0026ldquo;long-term institutional stay\u0026rdquo;. Participants who died at home were coded in three different ways in the analyses: they were (i) censored at the date of death, (ii) treated as events in the same way as moving to a long-term care facility, and (iii) treated as a competing event (see statistical analysis section). Those who remained living at home throughout the observation period were censored at the study end date. Two participants in the control group had missing data on service start.\u003c/p\u003e\n\u003ch3\u003eSecondary outcomes\u003c/h3\u003e\n\u003cp\u003eSecondary outcomes were measured as the amount and type of care services received.\u003c/p\u003e\n\u003ch4\u003eHome nursing care\u003c/h4\u003e\n\u003cp\u003eWe obtained data on duration and weekly hours of home nursing care decisions from Gerica. These were operationalized as total hours per participant over the project period. As receiving home nursing care was an inclusion criterion, almost all participants had a registered value. Some (n = 24) who discontinued early were assigned zero hours.\u003c/p\u003e\n\u003ch4\u003eShort-term institutional stays\u003c/h4\u003e\n\u003cp\u003eShort-term institutions are time-limited services intended for rehabilitation, assessment of care needs, or respite for family caregivers. The aim is usually for older adults to return to their own homes after illness, hospitalization, or a period of increased care needs. These were recorded as whole days in Gerica, with no value for intensity. We coded stays as (i) total number of days in a short-term institution, (ii) total number of days capped at 200 to handle extreme outliers, and (iii) total number of unique stays at a short-term care facility. Participants without any records were assigned a value of zero.\u003c/p\u003e\n\u003ch4\u003ePractical assistance\u003c/h4\u003e\n\u003cp\u003ePractical assistance includes home support services, aimed at helping individuals manage daily activities such as financial management and general routines. Recorded with duration and weekly hours in Gerica, from which we calculated the total hours received per participant throughout the project period. We added an outcome variable with hours capped at 100 to handle outliers. Participants without records were assigned a value of zero.\u003c/p\u003e\n\u003ch4\u003eSafety alarms\u003c/h4\u003e\n\u003cp\u003eSafety alarms allow older adults to call for assistance at any time of day, in case of a fall or a sudden health problem. We obtained data on safety alarm activations from the supplier Careium. Each activation was coded as unique observations in the data set, with a reason for activation labeled by healthcare professionals. We excluded records labeled as test. Participants with no records were assigned a value of zero.\u003c/p\u003e\n\u003ch2\u003eStatistical power\u003c/h2\u003e\n\u003cp\u003eWe conducted power calculations prior to the trial (Figure 3) using a two-sample proportions test in Stata (StataCorp, 2023). Following conventional thresholds of a 5 percent significance level and 80 percent power, we estimated that approximately 400 participants per group would be required to detect a 10 percentage point difference in the primary outcome.\u003c/p\u003e\n\u003cp\u003eHowever, the realized exposure differed from these assumptions. Although 516 participants were allocated to the intervention group, only 150 received a device, and an even smaller proportion got meaningful use of it. This reduces the statistical power of the trial and must be accounted for when interpreting the results.\u003c/p\u003e\n\u003ch2\u003eStatistical Analysis\u003c/h2\u003e\n\u003cp\u003eAll statistical analyses were conducted using Stata 18 [44]. For the primary outcome, time until moving to a long-term care facility, we used survival analysis. Hazard ratios (HRs) were first estimated in a crude model, before adjusting for baseline differences, using the Cox proportional hazards model [45]. Here, HRs above 1 indicate higher risk of moving, and HRs below 1 indicate a lower risk. We tested the proportional hazards assumption using Schoenfeld residuals [46]. No violations were detected. Further, we had a censoring issue. The Cox model assumes non-informative censoring, meaning that participants who are censored should drop out for reasons unrelated to the outcome [47]. In our case, deaths are directly related to the risk of moving to a long-term care facility. Thus, by censoring deaths, we violate this assumption. However, if we treat deaths as events of interest, we measure an outcome different from our primary outcome. For robustness purposes, we conducted two alternative analyses treating deaths both as censored and events. In addition, we applied a competing risks model [48] to account for deaths as a competing event, reporting subhazard ratios (SHRs).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAll secondary outcome variables were treated as count data. Almost all of them had issues with overdispersion (see Table 3), which violates the Poisson regression assumption of equal mean and variance [49]. To account for this, we used negative binomial regression models for home nursing care. The remaining secondary variables had, in addition, an excess of zero values. To address both overdispersion and excess zeros, we applied zero-inflated negative binomial (ZINB) regression models for days of short-term institutional stays, practical assistance, and safety alarm activations. ZINB combines a logistic component, which estimates the probability of structural zeros, with a negative binomial component, which models the count distribution among service users [50]. For the variable representing unique number of short-term stays, overdispersion was not significant. Although the variance was higher than the mean, the alpha parameter in the ZINB model was not significantly different from zero (not shown), indicating that overdispersion was not an issue [51]. Therefore, we used a zero-inflated Poisson (ZIP) model for this outcome.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eModel comparisons using AIC and BIC favored negative binomial models over Poisson for all secondary outcome variables [52]. Results are reported as incidence rate ratios (IRRs) for the count component, and odds ratios (ORs) for the excess zero component.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eIn Table 2, we present descriptives for the primary and secondary outcomes. Over the 870-day follow-up, 45.9 percent of participants remained at home, 32.5 percent transitioned to long-term care, and 21.7 percent died while living at home. Rates of moving to care facilities were highest in the intervention group, while the control group had more deaths and more participants staying at home throughout the period. In total, the population accumulated 651,158 days of living at home, corresponding to an incidence rate of 0.0005. For all secondary outcome variables, the variance exceeds the mean, indicating overdispersion [51].\u003c/p\u003e\n\u003cp\u003eIn Table 3, we present the results for the primary outcome, along with HRs from the Cox (1-4) and Fine-Gray models (5-6). Here, we assess the time to move into a long-term care facility. Across all models, the estimated HRs for the intervention group range from 1.090 to 1.162. This suggests that the intervention group had approximately 10 percent higher risk of moving to a long-term care facility compared to the control group. However, the differences between the intervention and control groups are not statistically significant.\u003c/p\u003e\n\u003cp\u003eImplementation data shows that participants who had registered use on their Komp were, on average, registered with 1.1 activities per week (received video calls, images, or messages). Ten participants had fewer than ten interactions in total, and 23 had fewer than 0.5 activities per week. 20 participants never received videocalls, but only images and messages (46 activations on average among these).\u003c/p\u003e\n\u003cp\u003eIn Table 4, we present IRRs and ORs from models assessing group differences in formal care service use. Models 7 and 8 estimate total hours of home nursing care using negative binomial regression. Both models show no significant differences between groups. The estimated alpha parameter confirms overdispersion, as it is significantly different from zero [51], supporting the model choice.\u003c/p\u003e\n\u003cp\u003eModels 9-12, report results from the ZINB models for short-term institutional stays. We find a significant reduction in the total number of days spent in short-term care for the intervention group (model 9: IRR = .833 [.728, .952]) of approximately 17 percent. Given the control group average of 31 days, this translates to a reduction of approximately five days per participant on average. The adjusted model (10) shows similar results (IRR = .828 [.725, .946]). The findings remain consistent when the dependent variable is capped at 200 days (models 11 and 12). The absence of significant effects in the excess zero component suggests that the odds of receiving the service were similar between groups.\u003c/p\u003e\n\u003cp\u003eIn models 13 and 14, we present results from the ZIP models for the number of unique short-term institutional stays, showing no differences between groups. Similarly, the ZINB models assessing hours of practical assistance (models 15-18) and the number of safety alarm activations (models 19-20) show no significant differences between groups.\u003c/p\u003e\n\u003cp\u003e[Table 4 here]\u003c/p\u003e\n\u003cp\u003eWe conducted two post hoc analyses (not shown) to examine treatment effects among the treated: an \u003cem\u003eas-treated\u003c/em\u003e analysis, comparing participants who received a Komp with controls, and a \u003cem\u003eper-protocol\u003c/em\u003e analysis, comparing those with registered Komp use with controls (Ahn \u0026amp; Kang, 2023). For the primary outcome, the as-treated analysis was largely consistent with the main analysis but indicated a significantly lower risk of moving to long-term institutional care and dying at home in the analyses treating deaths as an event (HR=0.692 [0.532, 0.901]). In contrast, the per-protocol analysis consistently indicated that Komp users had a significantly lower risk of moving to long-term institutional care (HR = 0.255 [0.113-0.575]). For the secondary outcomes, both post hoc analyses showed higher use of formal care services among the treated. Significant effects were observed for home nursing care (as-treated IRR = 1.396 [1.132, 1.720]), number of short-term institutional care stays (as-treated IRR = 1.244 [1.025, 1.509]; per-protocol IRR = 1.447 [1.068, 1.960]), practical assistance (as-treated IRR = 1.400 [1.181, 1.659]; per-protocol IRR = 1.682 [1.300, 2.176]), and safety alarm activations (as-treated IRR = 1.741 [1.101, 2.752]). No significant effects were observed for the number of days in short-term institutional stays.\u0026nbsp;\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis trial investigated the effects of Komp on the time spent living at home and the use of formal care services among older adults receiving long-term care at home. The findings did not support our hypothesis that access to Komp would prolong home-dwelling. On the contrary, the intervention group had a 10 to 16 percent higher risk of moving to a long-term care facility These findings were, however, not statistically significant. Given the reduced statistical power, these estimates cannot be taken as evidence of a true effect.\u003c/p\u003e \u003cp\u003eThe lack of significant effects in the primary outcome may be caused by the lack of any intervention effect, but are as likely to be explained by the limitations in statistical power and implementation. Due to delays in approval processes, the trial commenced six months later than planned, by which time several participants had become unavailable. As a result, only half of the intended devices were distributed, reducing the contrast between the groups. This reduced the statistical power of the trial, increasing the risk of actual effects going undetected.\u003c/p\u003e \u003cp\u003eIn addition, the intervention was implemented in a narrower form than initially intended. The trial design intended a dual use of Komp, by both family members and municipal care staff. As the municipal care services never adopted the use of Komp, we could measure only the effect of family use, which likely limits the effectiveness of the intervention. On the other hand, the absence of use from the health care services allowed us to isolate the effect of family use alone, and thus avoid conflating effects.\u003c/p\u003e \u003cp\u003eFurther, the results show that the intervention group experienced reduced usage of some formal care services, as we recorded a significant reduction in the duration of short-term institutional stays. On average, the intervention group spent approximately five fewer days in short-term care compared to controls. Notably, there were no differences between groups in hours of home nursing care, number of short-term institutional stays, hours of practical assistance, or safety alarm activations. This pattern suggests that the reduction in duration of short-term institutional stays was in fact a true reduction and not offset by increased use of other formal home-based services. However, as the intervention group showed a slight increase in long-term institutional stays, we cannot rule out that some substitution occurred between short-term and long-term care. As Komp users could bring their device to short-term care facilities, this might have facilitated contact with relatives during stays and thus supported participants to return home sooner. Meanwhile, Komp user also had the assurance of close family contact after leaving the short-term facilities, which may have provided additional support and made it easier to return home. On the other hand, the facilitated contact during the short-term stay might have allowed family members to monitor the users\u0026rsquo; condition more closely. If so, they may have been more likely to advocate for long-term institutional care after admission to short-term care. Thus, the findings related to short-term care must be interpreted with caution. Given the lack of significant effects on other outcomes, there is also a risk that the observed reduction in short-term institutional stays may be significant at random.\u003c/p\u003e \u003cp\u003eOur post hoc analyses complicate the interpretation of our results. We performed an \u003cem\u003eas-treated\u003c/em\u003e analysis comparing those who received a Komp with controls, and a \u003cem\u003eper-protocol\u003c/em\u003e analysis comparing participants with recorded Komp use to controls. In these analyses, we are no longer following a randomized design. Although they could represent a true effect of Komp, they could also be the result of selection bias. One possible explanation for the results is that frailer participants were more likely to accept the offer, as more active and technologically competent individuals have previously rejected Komp because they perceived it as too simplistic or even disempowering [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. This could explain the higher use of home-based formal care services that we observed in both post hoc analyses. On the other hand, these findings could come as a consequence of the prolonged home-dwelling observed in the per-protocol analyses. Yet, the observed prolonged home-dwelling itself could be a result of selection bias, as participants with the most active family ties would be more likely to have registered Komp activity and to receive more informal support, as previous studies have shown [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e, \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e]. Overall, these findings are best interpreted as results of selection bias rather than causal effects, and we therefore place greater weight in the ITT analyses. Still, we need more robust evidence to confidently conclude our findings.\u003c/p\u003e \u003cp\u003eAlthough Komp was designed to enhance social connectedness and thus reduce reliance on formal care services, we cannot rule out the possibility of unintended harms. For example, 97 Komp devices were not used during the project period. This means that 97 participants accepted the offer, but did not have anyone reaching out through the connected app. In some of these cases, participants might have paid little attention to it, and no harm may have occurred. However, for others, the presence of an unused device could have acted as a constant reminder of limited social contact. One could, in this scenario, imagine that the device reinforced feelings of social isolation, and thus inducing harm to participants in the intervention group. One potential way to prevent this would have been to select participants who were more closely aligned with Komp\u0026rsquo;s intended target group. For example, only including individuals with digital active family members, only those with family members living outside of the municipality, or participants identified by the municipal home care services as suitable candidates. Such an approach, however, has its downsides, as it could exclude individuals who might otherwise benefit from the intervention, and thus create a double burden for those already at risk of being left out.\u003c/p\u003e \u003cp\u003eTo the best of our knowledge, no previous trials have evaluated the effects of DCTs on similar outcomes. We therefore cannot directly compare our results to prior trials. However, as our program theory suggests that Komp may promote social connectedness, and thus prolong home-dwelling, we turn to studies on DCTs impact on subjective outcomes. As previously mentioned, only two full-scale RCT have evaluated DCT as standalone interventions. Czaja et al.\u0026rsquo;s [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e] computer system (PRISM) designed specifically for older adults were compared to Binder \u0026ndash; a notebook that contained paper content similar to the content in PRISM. They found that PRISM significantly reduced loneliness and increased social support, and showed tendencies to reduce social isolation after six months. These improvements, however, were not maintained at 12 months. Czaja et al.\u0026rsquo;s [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e] follow-up trial found significant reductions in loneliness, social isolation, and improvements in social support, quality of life, and health related quality of life. These improvements were only seen among participants in rural areas and senior housing locations \u0026ndash; not for those residing in assisted living communities. The follow-up trial also included a mediation analysis, indicating that the reduction in social isolation and loneliness were key pathways to improved quality of life. In sum, these trials indicate that DCTs may facilitate social connectedness in some contexts. Turning to interventions with clinical content, a meta-analysis of six RCTs on smart-phone based video calls and computer-based training aimed at reducing loneliness in older adults showed little to no effect [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThus, the evidence for significant large-scale effects of DCTs remains limited. From an economic perspective, this raises questions about the long-term sustainability of widespread DCT implementation in health and social care. However, when viewing the evidence of DCTs improving social connectedness and quality of life [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e], in light of our findings of potential reductions in formal care use and associated cost savings, a more tangible picture takes form. Specifically, the observed reduction in the duration of short-term institutional stays equals \u003cspan\u003e$\u003c/span\u003e1,500 per participant in a Norwegian context. Seen together, these findings suggest that while DCTs may hold promise as supportive tools in health and social care, the current knowledge base is still too weak to draw firm conclusions.\u003c/p\u003e \u003cp\u003eOur trial is the first RCT to evaluate the effect of a DCT on the use of formal care services among older adults. We have done so with a relatively large sample size, which proved necessary to be able to detect any differences between groups, in what turned out to be a trial marked by non-compliance. Moreover, the use of high-quality registry data available from Statistics Norway, allowed us to detect any skewness between groups on a broad range of variables with high accuracy. As our trial was conducted in a realistic municipal care setting within a conservative ITT approach, we have also enhanced the external validity and likely reduced the risk of unforeseen factors influencing the effects in a potential implementation at scale.\u003c/p\u003e \u003cp\u003eOur study has several limitations. First, the uptake of our intervention was low: of the 516 participants in the intervention group, 150 (29%) received a Komp device. This corresponds to 42 percent of those offered Komp (n\u0026thinsp;=\u0026thinsp;359). Among those who received a device, 53 participants were registered with use. This corresponds to 10 percent of the entire intervention group, 15 percent of those offered Komp, and a third of those who accepted one. Among those who were registered with Komp use, the usage differed substantially. As we follow an ITT-approach, the estimates should be interpreted as the effect of \u0026ldquo;belonging to the intervention group\u0026rdquo; rather than the isolated effect of the intervention [\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e]. While this reflects what could be expected in a large-scale implementation, it likely underestimates the potential positive or negative effects of the intervention itself.\u003c/p\u003e \u003cp\u003eSecond, the statistical power of the trial was limited. Calculations showed that we would not likely be able to detect smaller effects, which means that the lack of significant findings in both the primary and secondary outcomes should be interpreted with caution, as meaningful effects may have gone undetected. Third, there is a potential SUTVA violation, as two control group members resided with participants in the intervention group who received a Komp device. One control group member was also mistakenly offered a Komp. Given the low number of contaminated participants, it is unlikely to have influenced the results. Fourth, there is always a risk of unobserved variables having influenced the results. However, access to comprehensive background data reduces endogeneity concerns, as it let us adjust for differences in wealth and marital status. Lastly, the generalizability of our findings may be limited. Previous trials have demonstrated that contextual factors matters, as a DCT only proved effective among older adults living outside assisted living communities [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. Similarly, we must account for factors that restrict our findings to the context of our trial, e.g. that it was conducted in an urban area.\u003c/p\u003e \u003cp\u003eWe recommend that future studies continue to evaluate the effect of DCTs on formal care use in both similar and other contexts. These studies should also plan for investigating substitution effects between formal care services. Meanwhile, more trials assessing standalone DCT interventions, which are easier to implement at large scale, are needed. Additionally, it would be valuable for future research to explore more targeted approaches, as mentioned previously, which could both help prevent unintended harms and better reflect large scale implementations. Further, we recommend evaluations of how Komp, and other DCTs designed for individuals with less digital competence, affect users\u0026rsquo; social contact and well-being, to better understand the underlying mechanisms behind our findings. Finally, we recommend researchers to consider quasi-experimental approaches, such as propensity score matching, in trials where compliance is low. In sum, more robust evaluations of DCTs across a broader range of outcomes are necessary to strengthen the evidence base for the digitalization of welfare states.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn this trial, the results indicate that Komp, a digital communication technology, does not prolong home-dwelling among Norwegian older adults above 67 years. The results further indicate that Komp reduces reliance on some formal care services, with short-term institutional stays reduced with an average of five days in the intervention group. We recommend further research to investigate how Komp affects users\u0026rsquo; quality of life and well-being, to better understand the underlying mechanisms. We also recommend more robust evaluations of other types of digital communication technologies on various outcomes to strengthen the evidence base for the digitalization of welfare services.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eOECD\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Organization for Economic Co-operation and Development\u003c/p\u003e\n\u003cp\u003eDCT\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Digital Communication Technology\u003c/p\u003e\n\u003cp\u003eRCT\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Randomized Controlled Trial\u003c/p\u003e\n\u003cp\u003ePRISM\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Personal Reminder Information and Social Management\u003c/p\u003e\n\u003cp\u003eITT\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Intention-to-Treat\u003c/p\u003e\n\u003cp\u003eCONSORT\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Consolidated Standards of Reporting Trials\u003c/p\u003e\n\u003cp\u003eADL\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Activities of Daily Living\u003c/p\u003e\n\u003cp\u003eSD\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Standard Deviation\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eHR\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Hazard Ratio\u003c/p\u003e\n\u003cp\u003eSHR\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Sub Hazard Ratio\u003c/p\u003e\n\u003cp\u003eZINB\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Zero-Inflated Negative Binomial\u003c/p\u003e\n\u003cp\u003eZIP\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Zero-Inflated Poisson\u003c/p\u003e\n\u003cp\u003eAIC\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Akaike Information Criterion\u003c/p\u003e\n\u003cp\u003eBIC\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Bayesian Information Criterion\u003c/p\u003e\n\u003cp\u003eIRR\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Incidence Rate Ratio\u003c/p\u003e\n\u003cp\u003eOR\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Odds Ratio\u003c/p\u003e\n\u003cp\u003eSUTVA \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Stable Unit Treatment Value Assumption\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch3\u003eEthics approval and consent to participate\u003c/h3\u003e\n\u003cp\u003eWe have institutional approval from the GDPR contact at the Faculty of Social Sciences, Oslo Metropolitan University. Ethical approval was obtained from the Regional Committees for Medical and Health Research Ethics (REK, 516796) on October 10, 2022, and from the Norwegian Agency for Shared Services in Education and Research (SIKT, 332463) on December 21, 2022. In accordance with the approval from REK, active informed consent was not required for the use of anonymized background data [55]. All participants were informed about the study and were given the opportunity to opt out of the use of their data. Informed consent to receive the intervention was obtained from all participants who were offered the intervention by healthcare professionals. All procedures were conducted in accordance with the ethical principles of the Declaration of Helsinki [56].\u003c/p\u003e\n\u003ch3\u003eConsent for publication\u003c/h3\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003ch3\u003eAvailability of data and materials\u003c/h3\u003e\n\u003cp\u003eThe data that support the findings of this study are not publicly available without further approval from Oslo Municipality and SIKT. The do-files can be accessed by contacting the authors. The trial is registered at ClinicalTrials.gov (NCT05919355). The trial protocol is publicly available [57].\u003c/p\u003e\n\u003ch3\u003eCompeting interests\u003c/h3\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003ch3\u003eFunding\u003c/h3\u003e\n\u003cp\u003eThis work was supported by the Research Council of Norway [grant number 331810].\u003c/p\u003e\n\u003ch3\u003eAuthors’\u0026nbsp;contributions\u003c/h3\u003e\n\u003cp\u003eEBR led the research project. EBR obtained funding with support from KAW and BSAT, and data with support from KAW, BSAT and KE. KE led the work with the present paper, analyzed and interpreted the data, and drafted the manuscript. KE, IMH, EBR, BSAT, AGT and KAW contributed to critically revising the paper and agreed to be accountable for all aspects of the work.\u003c/p\u003e\n\u003ch3\u003eAcknowledgements\u003c/h3\u003e\n\u003cp\u003eThe authors would like to thank Anna-Stina Slattum, Research \u0026amp; Impact manager in Abilia, who led the BoVel-project; Professor Anne Lund and Martin Vinther Bavngaard for their contribution to the development of, and participation in the project; the local project coordinators for making it possible to implement Komp; and Anett Günther Warnstrøm and Irene Oksdøl for their assistance related to the Gerica data, and initiating the idea to study safety alarm data. This work would not have been possible without your contributions.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003e\u003cstrong\u003eNasjonale befolkningsframskrivinger \u003c/strong\u003e[https://www.ssb.no/befolkning/befolkningsframskrivinger/statistikk/nasjonale-befolkningsframskrivinger/artikler/vi-blir-flere-men-ogsa-mye-eldre]\u003c/li\u003e\n\u003cli\u003eOECD: \u003cstrong\u003eHealth at a Glance: Europe 2024: State of Health in the EU Cycle\u003c/strong\u003e: OECD publishing; 2024.\u003c/li\u003e\n\u003cli\u003eHOD: \u003cstrong\u003eFellesskap og meistring: Bu trygt heime\u003c/strong\u003e. In\u003cem\u003e.\u003c/em\u003e Edited by Services MoHaC; 2022-2023.\u003c/li\u003e\n\u003cli\u003eHaak M, F\u0026auml;nge A, Iwarsson S, Dahlin Ivanoff S: \u003cstrong\u003eHome as a signification of independence and autonomy: experiences among very old Swedish people\u003c/strong\u003e. \u003cem\u003eScandinavian journal of occupational therapy \u003c/em\u003e2007, \u003cstrong\u003e14\u003c/strong\u003e(1):16-24.\u003c/li\u003e\n\u003cli\u003eTate K, Bailey S, Deschenes S, Grabusic C, Cummings GG: \u003cstrong\u003eFactors influencing older persons\u0026rsquo; transitions to facility-based care settings: a scoping review\u003c/strong\u003e. \u003cem\u003eThe Gerontologist \u003c/em\u003e2023, \u003cstrong\u003e63\u003c/strong\u003e(7):1211-1227.\u003c/li\u003e\n\u003cli\u003eLuppa M, Luck T, Weyerer S, K\u0026ouml;nig H-H, Br\u0026auml;hler E, Riedel-Heller SG: \u003cstrong\u003ePrediction of institutionalization in the elderly. A systematic review\u003c/strong\u003e. \u003cem\u003eAge and ageing \u003c/em\u003e2010, \u003cstrong\u003e39\u003c/strong\u003e(1):31-38.\u003c/li\u003e\n\u003cli\u003eBlakely KK, Baker NR: \u003cstrong\u003eLoneliness in community-dwelling, older adults: an integrative review\u003c/strong\u003e. \u003cem\u003eThe Journal for Nurse Practitioners \u003c/em\u003e2023, \u003cstrong\u003e19\u003c/strong\u003e(1):104390.\u003c/li\u003e\n\u003cli\u003eSteiner JF, Ross C, Stiefel M, Mosen D, Banegas MP, Wall AE, Martin C, Kelly TS, Paolino AR, Zeng C: \u003cstrong\u003eAssociation between changes in loneliness identified through screening and changes in depression or anxiety in older adults\u003c/strong\u003e. \u003cem\u003eJournal of the American Geriatrics Society \u003c/em\u003e2022, \u003cstrong\u003e70\u003c/strong\u003e(12):3458-3468.\u003c/li\u003e\n\u003cli\u003eDonovan NJ, Wu Q, Rentz DM, Sperling RA, Marshall GA, Glymour MM: \u003cstrong\u003eLoneliness, depression and cognitive function in older US adults\u003c/strong\u003e. \u003cem\u003eInternational journal of geriatric psychiatry \u003c/em\u003e2017, \u003cstrong\u003e32\u003c/strong\u003e(5):564-573.\u003c/li\u003e\n\u003cli\u003eGriffin SC, Mezuk B, Williams AB, Perrin PB, Rybarczyk BD: \u003cstrong\u003eIsolation, not loneliness or cynical hostility, predicts cognitive decline in older Americans\u003c/strong\u003e. \u003cem\u003eJournal of aging and health \u003c/em\u003e2020, \u003cstrong\u003e32\u003c/strong\u003e(1-2):52-60.\u003c/li\u003e\n\u003cli\u003eLambotte D, De Donder L, Van Regenmortel S, Fret B, Dury S, Smetcoren A-S, Dierckx E, De Witte N, Vert\u0026eacute; D, Kardol MJM: \u003cstrong\u003eFrailty differences in older adults\u0026rsquo; use of informal and formal care\u003c/strong\u003e. \u003cem\u003eArchives of Gerontology and Geriatrics \u003c/em\u003e2018, \u003cstrong\u003e79\u003c/strong\u003e:69-77.\u003c/li\u003e\n\u003cli\u003eBigonnesse C, Chaudhury H: \u003cstrong\u003eThe Landscape of \u0026ldquo;Aging in Place\u0026rdquo; in Gerontology Literature: Emergence, Theoretical Perspectives, and Influencing Factors\u003c/strong\u003e. \u003cem\u003eJournal of Aging and Environment \u003c/em\u003e2020, \u003cstrong\u003e34\u003c/strong\u003e(3):233-251.\u003c/li\u003e\n\u003cli\u003eEmlet CA, Moceri JT: \u003cstrong\u003eThe Importance of Social Connectedness in Building Age-Friendly Communities\u003c/strong\u003e. \u003cem\u003eJournal of Aging Research \u003c/em\u003e2012, \u003cstrong\u003e2012\u003c/strong\u003e(1):173247.\u003c/li\u003e\n\u003cli\u003eSteels S: \u003cstrong\u003eKey characteristics of age-friendly cities and communities: A review\u003c/strong\u003e. \u003cem\u003eCities \u003c/em\u003e2015, \u003cstrong\u003e47\u003c/strong\u003e:45-52.\u003c/li\u003e\n\u003cli\u003eBavngaard MV, Lund A, Thordardottir B, Rasmussen EB: \u003cstrong\u003eThe Uses and Experiences of Synchronous Communication Technology for Home-Dwelling Older Adults in a Home Care Services Context: Qualitative Systematic Review\u003c/strong\u003e. \u003cem\u003eJournal of Medical Internet Research \u003c/em\u003e2024, \u003cstrong\u003e26\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003eDigital Technologies Can Help Older Persons Maintain Healthy, Productive Lives \u003c/strong\u003e[https://www.un.org/en/un-chronicle/digital-technologies-can-help-older-persons-maintain-healthy-productive-lives]\u003c/li\u003e\n\u003cli\u003eITU: \u003cstrong\u003eAgeing in a digital world - from vulnerable to valuable\u003c/strong\u003e: ITU Publications; 2021.\u003c/li\u003e\n\u003cli\u003eDesCh\u0026acirc;telets JR, Khowaja AR, Mechelse K, Koning H, Ventresca D: \u003cstrong\u003eExploring the access and use of social technologies by older adults in support of their mental health during the COVID-19 pandemic: A rapid review\u003c/strong\u003e. \u003cem\u003eCanadian Journal on Aging/La Revue canadienne du vieillissement \u003c/em\u003e2023, \u003cstrong\u003e42\u003c/strong\u003e(4):728-743.\u003c/li\u003e\n\u003cli\u003eAmbagtsheer R, Borg K, Townsin L, De Plaza MP, O\u0026apos;Brien L, Kunwar R, Lawless M: \u003cstrong\u003eThe effectiveness of technology interventions in reducing social isolation and loneliness among community-dwelling older people: A mixed methods systematic review\u003c/strong\u003e. \u003cem\u003eArchives of Gerontology and Geriatrics Plus \u003c/em\u003e2024, \u003cstrong\u003e1\u003c/strong\u003e(1).\u003c/li\u003e\n\u003cli\u003eYutong T, Yan Z, Qingyun C, Lixue M, Mengke G, Shanshan W: \u003cstrong\u003eInformation and Communication Technology Based Integrated Care for Older Adults: A Scoping Review\u003c/strong\u003e. \u003cem\u003eInt J Integr Care \u003c/em\u003e2023, \u003cstrong\u003e23\u003c/strong\u003e(2):2.\u003c/li\u003e\n\u003cli\u003eNoone C, McSharry J, Smalle M, Burns A, Dwan K, Devane D, Morrissey E: \u003cstrong\u003eVideo calls for reducing social isolation and loneliness in older people: a rapid review\u003c/strong\u003e. \u003cem\u003eCochrane Database of Systematic Reviews \u003c/em\u003e2020, \u003cstrong\u003e5\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003ePetersen B, Khalili-Mahani N, Murphy C, Sawchuk K, Phillips N, Li KZ, Hebblethwaite S: \u003cstrong\u003eThe association between information and communication technologies, loneliness and social connectedness: A scoping review\u003c/strong\u003e. \u003cem\u003eFrontiers in Psychology \u003c/em\u003e2023, \u003cstrong\u003e14\u003c/strong\u003e:1063146.\u003c/li\u003e\n\u003cli\u003eChoi HK, Lee SH: \u003cstrong\u003eTrends and Effectiveness of ICT Interventions for the Elderly to Reduce Loneliness: A Systematic Review\u003c/strong\u003e. \u003cem\u003eHealthcare \u003c/em\u003e2021, \u003cstrong\u003e9\u003c/strong\u003e(3):293.\u003c/li\u003e\n\u003cli\u003eForsman AK, Nordmyr J, Matosevic T, Park A-L, Wahlbeck K, McDaid D: \u003cstrong\u003ePromoting mental wellbeing among older people: technology-based interventions\u003c/strong\u003e. \u003cem\u003eHealth promotion international \u003c/em\u003e2018, \u003cstrong\u003e33\u003c/strong\u003e(6):1042-1054.\u003c/li\u003e\n\u003cli\u003eHansen T, Nes RB, Hynek K, Nilsen TS, Reneflot A, Stene-Larsen K, Tornes RA, Bidonde J: \u003cstrong\u003eTackling social disconnection: an umbrella review of RCT-based interventions targeting social isolation and loneliness\u003c/strong\u003e. \u003cem\u003eBMC Public Health \u003c/em\u003e2024, \u003cstrong\u003e24\u003c/strong\u003e(1):1917.\u003c/li\u003e\n\u003cli\u003eHeins P, Boots LM, Koh WQ, Neven A, Verhey FR, de Vugt ME: \u003cstrong\u003eThe effects of technological interventions on social participation of community-dwelling older adults with and without dementia: a systematic review\u003c/strong\u003e. \u003cem\u003eJournal of clinical medicine \u003c/em\u003e2021, \u003cstrong\u003e10\u003c/strong\u003e(11):2308.\u003c/li\u003e\n\u003cli\u003eJin W, Liu Y, Yuan S, Bai R, Li X, Bai Z: \u003cstrong\u003eThe effectiveness of technology-based interventions for reducing loneliness in older adults: A systematic review and meta-analysis of randomized controlled trials\u003c/strong\u003e. \u003cem\u003eFrontiers in Psychology \u003c/em\u003e2021, \u003cstrong\u003e12\u003c/strong\u003e:711030.\u003c/li\u003e\n\u003cli\u003eTeo AR, Benton MC, Hooker ER, Zaccari B, Hidalgo NJ, Newell S, Tuepker A, Greaves DW, Nicosia FM: \u003cstrong\u003eEffect of telehealth yoga on loneliness and social isolation among rural older adults: a randomized controlled trial\u003c/strong\u003e. \u003cem\u003eAging \u0026amp; Mental Health \u003c/em\u003e2025:1-9.\u003c/li\u003e\n\u003cli\u003eJarvis MA, Padmanabhanunni A, Chipps J: \u003cstrong\u003eAn evaluation of a low-intensity cognitive behavioral therapy mHealth-supported intervention to reduce loneliness in older people\u003c/strong\u003e. \u003cem\u003eInternational journal of environmental research and public health \u003c/em\u003e2019, \u003cstrong\u003e16\u003c/strong\u003e(7):1305.\u003c/li\u003e\n\u003cli\u003eCzaja SJ, Boot WR, Charness N, Rogers WA, Sharit J: \u003cstrong\u003eImproving Social Support for Older Adults Through Technology: Findings From the PRISM Randomized Controlled Trial\u003c/strong\u003e. \u003cem\u003eThe Gerontologist \u003c/em\u003e2018, \u003cstrong\u003e58\u003c/strong\u003e(3):467-477.\u003c/li\u003e\n\u003cli\u003eCzaja SJ, Charness N, Rogers WA, Sharit J, Moxley JH, Boot WR: \u003cstrong\u003eThe benefits of technology for engaging aging adults: Findings from the PRISM 2.0 trial\u003c/strong\u003e. \u003cem\u003eInnovation in aging \u003c/em\u003e2024, \u003cstrong\u003e8\u003c/strong\u003e(6).\u003c/li\u003e\n\u003cli\u003eRasmussen EB, Askheim C, Oppedal B, Haldar M: \u003cstrong\u003eVarm teknologi mot ensomhet blant eldre? En skriptanalyse av kommunikasjonsl\u0026oslash;sningen KOMP\u003c/strong\u003e. 2021.\u003c/li\u003e\n\u003cli\u003eAkhtar SA: \u003cstrong\u003eSocially connected while apart: the use of technology to increase social connection between nursing home residents and their relatives\u003c/strong\u003e. \u003cem\u003eFrontiers in Public Health \u003c/em\u003e2024, \u003cstrong\u003e12\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eRasmussen EB, Akhtar S: \u003cstrong\u003eEnkel og trygg p\u0026aring;r\u0026oslash;rendekontakt: Forskningsrapport fra et pandemiprosjekt om bruk av sosial teknologi for eldre p\u0026aring; langtidshjem\u003c/strong\u003e. \u003cem\u003eSkriftserien \u003c/em\u003e2022:71-71.\u003c/li\u003e\n\u003cli\u003eBerg A, Myren GES: \u003cstrong\u003eDigital teknologi for sosial kontakt: En deskriptiv studie av de eldste eldres erfaringer med KOMP\u003c/strong\u003e. \u003cem\u003eNordisk tidsskrift for helseforskning \u003c/em\u003e2021, \u003cstrong\u003e17\u003c/strong\u003e(2).\u003c/li\u003e\n\u003cli\u003eFunnell SC, Rogers PJ: \u003cstrong\u003ePurposeful program theory: Effective use of theories of change and logic models\u003c/strong\u003e: John Wiley \u0026amp; Sons; 2011.\u003c/li\u003e\n\u003cli\u003eHopewell S, Chan A-W, Collins GS, Hr\u0026oacute;bjartsson A, Moher D, Schulz KF, Tunn R, Aggarwal R, Berkwits M, Berlin JA: \u003cstrong\u003eCONSORT 2025 statement: updated guideline for reporting randomised trials\u003c/strong\u003e. \u003cem\u003eThe Lancet \u003c/em\u003e2025, \u003cstrong\u003e405\u003c/strong\u003e(10489):1633-1640.\u003c/li\u003e\n\u003cli\u003eChristensen K, W\u0026aelig;rness K: \u003cstrong\u003eLong-term care services in Norway: a historical sociological perspective\u003c/strong\u003e. In: \u003cem\u003eThe Routledge handbook of social care work around the world.\u003c/em\u003e Routledge; 2017: 15-28.\u003c/li\u003e\n\u003cli\u003eColombo F, Llena-Nozal A, Mercier J, Tjadens F: \u003cstrong\u003eHelp wanted? Providing and paying for long-term care.\u003c/strong\u003e Paris: OECD; 2011.\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003eAntall personer p\u0026aring; venteliste for langtidsplass p\u0026aring; sykehjem \u003c/strong\u003e[https://www.helsedirektoratet.no/statistikk/venteliste-for-langtidsplass-pa-sykehjem]\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003eSykehjem og hjemmetjenesten i Norge \u003c/strong\u003e[https://www.ssb.no/helse/helsetjenester/artikler/sykehjem-og-hjemmetjenesten-i-norge]\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003eGjennomsnittlig botid p\u0026aring; langtidsopphold i institusjon \u003c/strong\u003e[https://www.helsedirektoratet.no/rapporter/kommunale-helse-og-omsorgstjenester-i-2023/utvalgte-temaer/gjennomsnittlig-botid-pa-langtidsopphold-i-institusjon]\u003c/li\u003e\n\u003cli\u003eBroglio K: \u003cstrong\u003eRandomization in clinical trials: permuted blocks and stratification\u003c/strong\u003e. \u003cem\u003eJama \u003c/em\u003e2018, \u003cstrong\u003e319\u003c/strong\u003e(21):2223-2224.\u003c/li\u003e\n\u003cli\u003eStataCorp: \u003cstrong\u003eStata Statistical Software: Release 18\u003c/strong\u003e. In\u003cem\u003e.\u003c/em\u003e: StataCorp LLC.; 2023.\u003c/li\u003e\n\u003cli\u003eCox DR: \u003cstrong\u003eRegression models and life\u003c/strong\u003e\u003cstrong\u003e‐tables\u003c/strong\u003e. \u003cem\u003eJournal of the Royal Statistical Society: Series B (Methodological) \u003c/em\u003e1972, \u003cstrong\u003e34\u003c/strong\u003e(2):187-202.\u003c/li\u003e\n\u003cli\u003eKleinbaum DG, Klein M: \u003cstrong\u003eSurvival analysis a self-learning text\u003c/strong\u003e: Springer; 1996.\u003c/li\u003e\n\u003cli\u003eRanganathan P, Pramesh C: \u003cstrong\u003eCensoring in survival analysis: potential for bias\u003c/strong\u003e. \u003cem\u003ePerspectives in clinical research \u003c/em\u003e2012, \u003cstrong\u003e3\u003c/strong\u003e(1):40.\u003c/li\u003e\n\u003cli\u003eFine JP, Gray RJ: \u003cstrong\u003eA proportional hazards model for the subdistribution of a competing risk\u003c/strong\u003e. \u003cem\u003eJournal of the American statistical association \u003c/em\u003e1999, \u003cstrong\u003e94\u003c/strong\u003e(446):496-509.\u003c/li\u003e\n\u003cli\u003eHilbe JM: \u003cstrong\u003eModeling count data\u003c/strong\u003e. In: \u003cem\u003eInternational encyclopedia of statistical science.\u003c/em\u003e Springer; 2011: 836-839.\u003c/li\u003e\n\u003cli\u003eHarris T, Hilbe JM, Hardin JW: \u003cstrong\u003eModeling count data with generalized distributions\u003c/strong\u003e. \u003cem\u003eThe Stata Journal \u003c/em\u003e2014, \u003cstrong\u003e14\u003c/strong\u003e(3):562-579.\u003c/li\u003e\n\u003cli\u003eHardin JW, Hilbe JM: \u003cstrong\u003eRegression models for count data based on the negative binomial (p) distribution\u003c/strong\u003e. \u003cem\u003eThe Stata Journal \u003c/em\u003e2014, \u003cstrong\u003e14\u003c/strong\u003e(2):280-291.\u003c/li\u003e\n\u003cli\u003eBurnham KP, Anderson DR: \u003cstrong\u003eMultimodel inference: understanding AIC and BIC in model selection\u003c/strong\u003e. \u003cem\u003eSociological methods \u0026amp; research \u003c/em\u003e2004, \u003cstrong\u003e33\u003c/strong\u003e(2):261-304.\u003c/li\u003e\n\u003cli\u003eOppedal B, Askheim C, Haldar M: \u003cstrong\u003eErfaringer med KOMP som velferdsteknologi\u003c/strong\u003e. \u003cem\u003eSkriftserien \u003c/em\u003e2019:56-56.\u003c/li\u003e\n\u003cli\u003eFeinman RD: \u003cstrong\u003eIntention-to-treat. What is the question?\u003c/strong\u003e \u003cem\u003eNutrition \u0026amp; Metabolism \u003c/em\u003e2009, \u003cstrong\u003e6\u003c/strong\u003e(1):1.\u003c/li\u003e\n\u003cli\u003eNESH: \u003cstrong\u003eGuidelines for Research Ethics in the Social Sciences and the Humanities\u003c/strong\u003e. In\u003cem\u003e.\u003c/em\u003e, 5th edn: National Committee for Research Ethics in the Social Sciences and the Humanities (NESH); 2022.\u003c/li\u003e\n\u003cli\u003eWMA: \u003cstrong\u003eWorld Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects\u003c/strong\u003e. \u003cem\u003eJama \u003c/em\u003e2013, \u003cstrong\u003e310\u003c/strong\u003e(20):2191-2194.\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003eResearch protocol for the BoVEL project \u003c/strong\u003e[https://static1.squarespace.com/static/6315e99c979ac470513caea9/t/6405b982ee0eef5b7d402eb6/1678096770598/Forskningsprotokoll+BoVEL.pdf]\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTable 1 to 5 are available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-geriatrics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bgtc","sideBox":"Learn more about [BMC Geriatrics](http://bmcgeriatr.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bgtc/default.aspx","title":"BMC Geriatrics","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Age in place, Survival analysis, Home and community based care and services, Long-term Care, Intervention study design/analysis","lastPublishedDoi":"10.21203/rs.3.rs-8378683/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8378683/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eNorway\u0026rsquo;s population is aging, increasing the pressure on already strained health and care services. In response, Norwegian policymakers have promoted a strategy of aging in place, i.e. that older adults are encouraged to live longer in their own homes. Low social contact is one of many risk factors for transitions to long-term care facilities. Digital technologies that facilitate social contact have been promoted as promising tools for increasing social contact among older adults and may also contribute to reduce strain on health and care services. In this RCT, we evaluate whether a digital communication technology, Komp, may prolong home-dwelling and reduce the need for formal care services among frail older adults. Given Komp\u0026rsquo;s documented potential to promote social connectedness and facilitate informal care, we hypothesize that Komp prolongs home-dwelling and reduces reliance on formal care services.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eWe compared an intervention group offered to try Komp (n\u0026thinsp;=\u0026thinsp;516) with a control group receiving services as usual (n\u0026thinsp;=\u0026thinsp;595). 150 Komp units were delivered. Outcomes were assessed 28 months after randomization. We obtained administrative data from Statistics Norway and service use data from the boroughs\u0026rsquo; journal records. We analyzed 1,099 participants (mean age: 84.6 years, range: 67\u0026ndash;98; 64.7% [\u003cem\u003en\u0026thinsp;=\u0026thinsp;711\u003c/em\u003e] female) using cox proportional hazards regression (primary outcomes) and ZINB and ZIP regression models (secondary outcomes).\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eThe intervention group had a higher, though non-significant, risk of moving to long-term care institutions compared to the control group (HR\u0026thinsp;=\u0026thinsp;1.162 [.944, 1.431]). ZINB regression models indicated that the intervention group experienced a reduction of five days of short-term institutional stays (IRR\u0026thinsp;=\u0026thinsp;.833 [.728, .952]). There were no differences between groups in the amount of other formal care services received.\u003c/p\u003e\u003ch2\u003eDiscussion and Implications\u003c/h2\u003e \u003cp\u003eThe implementation of Komp did not prolong home-dwelling among frail older adults. However, the results showed a reduction in formal care services, indicating that Komp may reduce some strain on health and care services. Future studies should evaluate the effect of other types of digital communication technologies on objective outcomes.\u003c/p\u003e\u003ch2\u003eTrial registration\u003c/h2\u003e \u003cp\u003eClinicalTrials.gov: NCT05919355. Date of registration: 16.06.2023.\u003c/p\u003e","manuscriptTitle":"The effect of digital communication technology on older adults’ formal care use in Norway: a randomized controlled trial","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-16 00:05:53","doi":"10.21203/rs.3.rs-8378683/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-02-20T18:22:21+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-02T00:29:53+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-26T19:37:51+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"201080004110777604934826657731280950986","date":"2026-01-20T15:40:13+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"276259643940586947223198327870848966373","date":"2026-01-20T04:07:15+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"92663682283041727119923284418322190654","date":"2026-01-19T12:46:41+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-01-08T04:47:06+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-08T04:37:33+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-12-26T05:06:21+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-12-23T17:30:37+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Geriatrics","date":"2025-12-23T17:24:32+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-geriatrics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bgtc","sideBox":"Learn more about [BMC Geriatrics](http://bmcgeriatr.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bgtc/default.aspx","title":"BMC Geriatrics","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"f10b8ea8-7d38-40bd-9d5e-f347a1ce9d12","owner":[],"postedDate":"January 16th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-04-13T16:14:04+00:00","versionOfRecord":{"articleIdentity":"rs-8378683","link":"https://doi.org/10.1186/s12877-026-07322-z","journal":{"identity":"bmc-geriatrics","isVorOnly":false,"title":"BMC Geriatrics"},"publishedOn":"2026-04-11 15:59:30","publishedOnDateReadable":"April 11th, 2026"},"versionCreatedAt":"2026-01-16 00:05:53","video":"","vorDoi":"10.1186/s12877-026-07322-z","vorDoiUrl":"https://doi.org/10.1186/s12877-026-07322-z","workflowStages":[]},"version":"v1","identity":"rs-8378683","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8378683","identity":"rs-8378683","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}