The US older adult cohort (adults aged 65 years and older) is expected to double in number from 48.7 million in 2015 to more than 98 million in the year 2060, making it one of the fastest-growing populations.1 With increasing life expectancy for both male and female older adults, it is necessary to consider additional options to promote positive aging. Older adults prefer to remain in their homes and to age in place.2 However, as health issues and functional decline are encountered by this population, certain adaptations may be required for older adults to successfully manage their current living environment.

It has been demonstrated that technology is an important component in the aging in place equation.3,4 For many seniors, technology aids the management of chronic conditions and the ability to remain independent.3-5 Technology also improves healthcare cost-effectiveness and supports caregivers.3-5 Given the significant role of technology in promoting health, wellness, and independence among older adults, an examination of the perceptions of older adults and the factors that support technology acceptance preferences will enhance an understanding of what types of technology older adults find useful and what forms of technology are most likely to be adopted by them. This information can be used as a guide to inform research in this area.

Older Adults and Technology-Perceptions and Preferences

Overall, studies indicate that older adults have an overall positive perception of technology.6-10 However, this positive perception is linked to how easy the technology is to understand and perceived usefulness of the product.5 These two factors are instrumental in technology acceptance for older adults. There are reported barriers in the adoption of technology among this group. Negative perceptions arise from inconveniences often related to cost and consequences in using such technology.2 These consequences include aesthetics of the system, creating burdens for self or family members, loss of control, negative health effects, and fear of stigmatization.2 Ease of learning the technology also emerged as a concern, with older adults expressing a need for strong training and instruction.11

Privacy issues surfaced as a negative factor particularly in relation to older adults' perceptions of technology use in the home.12 One systematic review of technology acceptance and use among older adults revealed that a main concern for technology acceptance was a threat to privacy; of these studies, half mentioned privacy as a potential concern or barrier to use.2 However, some of the study participants indicated a willingness to compromise some privacy if the technology provided benefits and enhanced independence.2 Data from other studies suggest that for older adults threats to privacy are outweighed by the sense of safety that accompanies monitoring technologies. Participants made connections between the sensors in their homes and independence.8,13,14

This review of the benefits and barriers to older adults' acceptance of technology serves as a foundation to identify what factors are most important when developing technology for this population. The most important factor is older adults' perceived need of technology. A strong need or usefulness factor can serve as a catalyst for technology acceptance if it addresses a problem or concern of the senior, such as safety or security. Potential barriers and concerns can be mitigated if the technology has a strong usefulness potential.2,6,8-10,15

Technology Development-Factors to Consider

The approach for this study was based on research that reported on successful approaches in technology development with older adults. One study recommended older adults' participation in technology advancement.16 In particular, including older adults in the creation and modification of health-related technologies and interfaces increases their engagement in the use of the equipment and their healthcare. Other research found that participants wanted to be active in the decision-making process related to technology placement in the home and data distribution to others (eg, sharing information with healthcare professionals).17 Moreover, a need to focus on the perceptions of older adults related to the daily use of technology was pinpointed.3 This finding was significant since few other studies focused on older adults' tangible interaction with the technologies.3 The information summarized above informed the development of the questionnaire used in the interview process. This article reports on a content analysis from interviews obtained from older adults living with a sensor suite employed in 13 assisted-living facilities. The embedded sensor technology system will be discussed briefly.

METHODS

Sensor Suite Description

This qualitative study was part of a larger randomized prospective intervention study that examined the clinical and cost effectiveness of data from an environmentally embedded sensor system used for early illness recognition. The sensor system deployed in this intervention included a "standard" suite of environmentally embedded (nonwearable) sensors, which unobtrusively and automatically monitor the functional status of older adults, detect impending changes in health or functional status, and send early alerts to healthcare providers.18,19

This sensor suite included motion sensors to measure overall activity; an under-mattress hydraulic bed sensor to capture respiration, pulse, and restlessness during sleep; and a gait sensor. The gait sensor was a small depth sensor that employed nonidentifiable, silhouette images to calculate fall risk by continuously measuring gait speed, stride length, and stride time (Figures 1, 2, and 3).

FIGURE 1. Depth sensor fall alert image.

FIGURE 2. Kinect depth sensor mounted on top of PC housing; x10 motion sensor visible beneath painting.

FIGURE 3. Hydraulic bed sensor in place between mattress and box springs.

Computer algorithms applied to the sensor data sent alerts to staff when changes in sensor data patterns were detected, possibly days or weeks before typical signs or symptoms were recognized by the study participant, family members, or providers.

Health alerts were sent to staff nurses through email. Each alert contained an electronic hyperlink that displayed the content of the health alert in the Web-based sensor data interface. Upon receiving the alert, staff nurses determined, based on their knowledge of the resident and his/her health conditions, whether additional assessment was required (Figure 4).

FIGURE 4. Sensor system data flow.

The sensor system served as a clinical decision support tool, augmenting the assessment of individual residents. Additionally, the depth sensor sent immediate alerts to staff when a fall occurred via an email to a cell phone or, in the case of these study sites, iPod Touch devices configured to receive these alerts. Each alert provided a short depth video clip of privacy-protecting, shadow-like, silhouette images of the alert trigger event. Staff can determine if an actual fall has occurred and respond accordingly.20-22 In the case of a false alarm, the depth video clip allowed staff to dismiss the alert without disturbing the study participant.

Content Analysis: Qualitative and Descriptive Study Methods on Participant and Family Perceptions

One objective of this study was to explore the perceptions of older adults on the usefulness of the sensor suite and their willingness to use it, as well as its impact on daily routine and privacy. With the consent of the participant, the perceptions of family members were also obtained to seek their opinions on the technology. Obtaining this information from participants and family provided an opportunity to address any concerns or problems that emerged through the sensor development phase and to learn more about how older adults adjust to and live with the technology.

Qualitative data were collected through individual face-to-face interviews with older adults who lived with the sensor suite. Additionally, face-to-face or telephone interviews were conducted with family members who were willing to participate in the study after the older adult gave permission to contact. After study introductions and the informed consent process, an initial set of interviews was conducted during the preinstallation phase of the project, using a structured interview guide. This interview guide was developed by experienced qualitative researchers and based on a quality-of-life technologies approach, which stipulates that understanding the needs and preferences of the end user, in this case, the older adult, is a first step in the technology development process.23 The Technology Acceptance Model was also used to inform the development of the interview guide.5 The structured interview guide included open-ended questions and prompts that encouraged the participant to expand on the answer. Questions covered content such as the following: Describe your experience living with the sensor suite. How does the sensor suite affect your daily living? Has the sensor suite helped in any way? Is the information you receive from the sensor suite helpful? How might you change any part of the sensor suite? These questions serve as examples of the type of information collected in the interviews. The interview questions were established by an expert panel of researchers to establish credibility.24 Institutional review board (IRB) approval was obtained prior to data collection.

We selected a content analysis approach method because it uses both quantitative and qualitative approaches.25 Using both types of analyses further establishes credibility and trustworthiness of the data.24,25

Sample

This study was conducted in 13 assisted-living facilities in Missouri. The sample was limited to participants in the larger study described previously who were living with the sensor suite. Participants met the criteria of no cognitive impairment or severe physical limitations imposed for the study. Screening tests for the study included the Mini-Mental State Exam and Patient Health Questionnaire. The sample consisted of 55 participants: 42 female, 13 male, 52 Caucasian, and three African American older adults. Beginning in July 2014, baseline interviews of intervention group participants were completed approximately 1 month after the sensor system was installed in participant apartments. Follow-up interviews were conducted quarterly throughout the study. Final interviews were collected in August 2016. Data are presented from 188 interviews from interview iterations 1 through 5 (Table 1).

Table 1 Participant Demographics by Interview Iteration

Interviews were also conducted with 13 participants' family members to obtain their perceptions of the technology (13 Caucasian, six daughters of study participant, seven sons of study participant, eight in-person interviews, five phone interviews). These interviews took place toward the end of the interview iteration process and were conducted at one point in time. Information on names, contact information, and permission to approach family members was obtained from the participant. A separate consent process was used for family members and approved by the IRB.

All interviews were audio-recorded with permission using a handheld digital recorder and transcribed by study staff. Using an open coding process, three researchers independently coded the transcripts by initially identifying meaning units and assigning codes to these meaning units.25 Using Dedoose (SocioCultural Research Consultants, Los Angeles, CA), study staff categorized the data through generated code presence matrices. These matrices were then used to identify themes. Descriptive statistics and frequencies were generated using IBM SPSS Statistics (IBM, Armonk, NY). A combined discussion of the qualitative and quantitative findings of this content analysis is reported in the next section.

RESULTS

Participant Interviews

Participants were interviewed in their home environment within the assisted-living facility. From five iterations of interviews conducted by research staff, the following themes emerged from the qualitative analysis of the interviews: (1) understanding and purpose, (2) daily life and benefits, (3) impact on privacy, and (4) sharing of information. These themes will be discussed, and descriptive statistics are included to provide additional support to the findings.

Understanding and Purpose: Descriptive Analysis

Residents were asked about their understanding of the type of data the sensors collected. Statistics from these interviews revealed that participants developed a better understanding of the sensor technology and its purpose over the course of the study. The percentage of participants who understood that the sensors collected data on in-room motion and falls increased remarkably from baseline to final interviews, while the percentage of participants who were unsure what data were collected fell from 41.8% to 0% by the end of the study. Similarly, the percentage of participants who reported that they thought the sensors collected data on motion in bed, respiration, pulse, and gait increased over the course of the study. Top responses for what type of data was collected included data on movement and falls.

Understanding of the sensors increased over the course of the study, perhaps associated with participants' increased knowledge of the role of the sensors in monitoring their activity and health data as shown in Table 2.

Table 2 Participant Responses

Understanding and Purpose: Qualitative Analysis

The qualitative analysis supported the descriptive analysis. Initially, there were varying levels of understanding of how the sensor suite actually worked. Data revealed three levels of understanding and behavior toward the sensor system: high, moderate, and low commitment levels.

Participants with a high level of commitment and understanding actively participated in the installation and placement of the sensor system. They wanted to know more about it and the progress toward development of this product. The following quote demonstrates an understanding of the technology and how it worked: "They're collecting data on my day-to-day activities and whether changes are taking place." This participant further stated, "They're checking that the same thing is happening that happened last week. That there's no change going on."

Participants who had a moderate commitment level understood what the sensor system did and the data it collected. They knew the technology would be useful to them and their healthcare and would help them in the future. These older adults knew the value of the sensor suite, but did not think about it much as articulated in the following quote: "[horizontal ellipsis]They want to see if I fall down on the floor or something, if I'm moving around."

Participants with a low commitment level had minimal understanding of the sensor suite. They were able to articulate that it was meant to help them with their healthcare; however, it was a passive acceptance. This level of understanding is reflected in the following quote: "Those things are sensoring me? Is that what they are doing? All I know is that if I fell on the floor, it would tell someone to come in."

Overall, participants' understanding of the sensor suite became stronger over time. After their orientation session and follow-up interviews, comments reflected a progressive change occurred from just knowing there were sensors and where they were placed to being able to describe what type of information each sensor collected.

Daily Life and Benefits: Descriptive Analysis

When asked how the sensors affected their normal routine, the descriptive analysis indicates that from baseline throughout the course of the study the majority of participants reported that they were not bothered by the sensors and that for the most part the sensors did not interfere with daily life. Participants reporting noninterference continued to increase from baseline to interview 5. Those who did express concern or awareness of the sensors reported that they were aware of a certain feature such as blinking lights, or they were bothered by the appearance or placement of the sensors. Throughout the course of the study, as participants expressed concern about aesthetic issues with the sensors, those issues were addressed. This proactive response had a positive impact in that the number of participants who expressed these concerns decreased from 10.9% to 0% throughout the course of the study. See Table 2 for a summary of the findings on reports of daily life experiences with the sensors.

When asked about the benefit of the sensors on everyday life, a small percentage of participants were able to respond. These benefits included sensors vaguely being of some help, glad to have the sensors, feeling safer with them, and residents who recalled staff responding to an actual alert. Over the course of the study, the number of older adults who were able to report benefits increased and they were more specific in describing how the sensor suite helped them. It is interesting to note that the number of participants who could recall staff responding to an alert increased from 5.5% at the beginning of the study to 14.3% at the end. Table 2 reports on the responses on the benefits of the system over time.

Daily Life and Benefits: Qualitative Analysis

While most participant comments revealed that the sensors did not interfere with their living environment, there was an adjustment period, as articulated in the interviews. Participants' comfort level with the technology grew over time. For example, in the initial interviews, participants expressed concerns about being "watched." Participants were aware of the sensor suite and regarded it as a new piece of equipment, which they acclimated to over time. As the participants lived with the sensors, the comments about "being watched" were replaced with comments about being monitored by a healthcare professional, which was regarded as beneficial. Benefits included feeling safer, feeling more secure, and being more careful and mindful of their movements.

In a comment about feeling more secure, one participant remarked, "I feel it's more becoming a security thing. Because if I am flat on the floor, I don't have the button around my neck, but if I were down on the floor, there is no way in god's world I could get to that button to turn it on for help, so it's a security blanket, I think, to have here." Another comment captured how the sensor suite encouraged mindfulness: "So I will take my time, be more careful, instead of bolting up and doing something[horizontal ellipsis] to be more mindful of what I am doing, rather than just charging around here and possibly fall or something." Participants at separate facilities reported that seeing the depth sensor in their apartment served as a reminder to be careful in motion to avoid falling. One participant remarked, "The sensors remind me to slow down and have deliberate moves. I think it reminds me[horizontal ellipsis] 'cause I know why it's there." A common concern expressed during the interviews was the prospect of falling and being unable to reach one's pendant or emergency pull-cord. One participant reported, "I do have the feeling that I'm safer by having the sensor in here, if I should fall and couldn't reach my emergency chain. I feel much safer with it."

For those who found the sensors helpful, common responses included comfort and safety. Over time, participants could recall more events in which staff responded to the sensor data. They felt more secure knowing staff were there to help them.

Over time, the sensor suite was regarded as a background utility that supported their independent lifestyle. It was regarded as a very practical feature and one that they did not need to take care of or worry about. Their confidence about the technology grew, and after a while, its impact diminished. One participant commented, "Well I just do what I do every day the same as usual, and it doesn't interfere with that. They're (the sensors) just there, you guys know what you are doing, and it doesn't affect me one way or the other that I have to do anything about them. They're just there, and they do what they want to do, and I just ignore them."

One concern related to the aesthetics of the system. The placement of the sensor suite introduced a change in the physical environment. Some participants had very particular concerns about the placement and the appearance of the sensors that needed to be addressed. One participant commented, "Well, I just thought it looked so massive for this little room. It looks more like an institution now than a regular room. If the box could have been smaller, and the wire could have been put down next to, in the corner, maybe. It's just too open."

In summary, after a settling-in period with the sensor suite, most participants adjusted to its presence. One participant summed up the typical experience, "I don't even notice they're there, until I happen to look up there and see them." For participants who found the sensors helpful, they felt more assured knowing that it would detect problems as they occurred. Commitment levels to the sensors were reflected in the way participants regarded the benefits of the sensor suite and their behavior toward it. Those who mentioned more active roles for the sensors, such as helping them to be more mindful, were more highly committed to the system. Participants who ignored the sensors reflected a lower commitment to the system.

Impact of Sensors on Privacy: Descriptive Analysis

Participants were asked if the presence of the sensors and the nature of data collected were a cause for privacy concerns. Across all interview iterations, the majority of participants reported that they had no such concerns, with 94.5% reporting no concerns at baseline and 92.9% reporting no such concerns at the end of the study. When asked about privacy issues, a small percentage of participants made comments about having nothing to hide due to their age or the way that they lived. This unique finding may represent continuity and longstanding opinions about limitations on privacy. The privacy concerns that were expressed were categorized as either issues with sensor images of movement or issues about where the data go and who had access to the data. These privacy concerns remained consistent over the course of the study. However, the one-on-one interviews and retraining on the sensor suite did appear to alleviate many of the concerns at the time they were expressed. It is important to note that participants chose to remain in the study even with these expressed concerns. Table 2 presents descriptive information on the impact of the sensors on privacy.

Impact of Sensors on Privacy: Qualitative Analysis

Data collected from the qualitative analysis support the descriptive findings. The majority of participants did not believe that the sensor system located in their living environment or the data collected violated privacy. The older adults expressed trust in the system, the type of data collected, and that data were collected to help them. They believed the information to be secure and that the sensor suite was not just taking random pictures. One participant said: "No, no I don't consider it an interruption to my privacy. They are here for a reason, and that reason is not harmful to me or will not get in the way of anything that I do or want to do or anything like that. It's more like research on activity, personal activity."

A change in expectations of privacy with age was cited as a reason for not being concerned. One participant remarked, "Well, that doesn't really bother me very much. It might somebody who is younger. By the time you reach 98, I think I've about seen it all as far as privacy is concerned." Another dismissed the notion of retaining much privacy in later life: "My life is a wide open book. I don't care."

One privacy issue that did emerge concerned the detail in the recorded images from the depth sensor. When shown clips of the actual image, most participant concerns were alleviated. "It doesn't show the body; it's not me, it's just an object." However, some participants wanted limits on when and where the images were recorded: "Maybe when I'm getting ready for bed, I wouldn't want them to watch me," said one participant. A few residents did express discomfort about being observed or watched: "I don't like the feeling that I'm being observed. I like my privacy." Participants who expressed privacy concerns were asked whether they wanted to discontinue, they agreed to continue due to the benefits of the technology. One resident stated, "I don't like it, but I wouldn't let it bother me."

The majority of participants were aware of the sensor suite and had confidence in technology operations and in how the data were being shared. One participant summed it up: "I don't think anything of it (sensors). It's there, but it is not spying on me or anything like that. It is just picking up data." Privacy concerns were alleviated through training and orientation to the technology or were not strong enough to cause attrition.

Sharing of Information: Descriptive Analysis

Participants were asked what should be done with collected health information and who should have access to it. Across all interview iterations, 53.8% of participants responded that they could not think of anyone they would want to exclude. Similarly, roughly 17% of participants responded that their data should be shared with whomever needs to see it, including family, clinicians, and researchers. Eight percent of participants stipulated that those in the position to share data should exercise discretion by not sharing data with everyone. Participants often had specific family members in mind like a son, daughter, spouse, or niece.

Approximately 45% wanted their children to see their data, followed by their physician (22%) and facility staff (10.2%). The desire for physicians, nurses, and administrators specifically to have access to the data was a frequent response.

Roughly 9% of participants expressed an interest in seeing their own data. One reason to show the participant depth sensor clips of falls was to illustrate why a fall occurred in order to prevent subsequent falls. Participants wanted to improve their activities and actions so that they could remain independent and in their current living environment.

Sharing of Information: Qualitative Analysis

Qualitative remarks about data sharing centered on the best people to see participant health data, including physicians, nurses, facility administrators, and others directly involved in their care. To this point, one older adult remarked, "Well if it interests anyone who takes care of me, I think they should know it." Participants were comfortable sharing information with children, nieces, nephews, spouse, and grandchildren as long as they had input into which relatives had access to the information. One participant said, "My son. He comes to see me nearly every day." Participants did not want their neighbors or anyone not directly involved with them or their care to have access. One quote describes this preference: "Well, I don't want anybody else to see it except the nurses here and my daughter, because it really doesn't concern anybody else except for me and my family [horizontal ellipsis]I just don't want everyone to see it." And another: "Well, I don't think the other residents would need to know." There was also a group of older adults who did not care who saw the data: "Anybody, as far as I'm concerned. It doesn't bother me. I mean they aren't going to do anything life threatening to me or anything."

The interviews again reflected the three levels of understanding and behavior that were previously identified: high, moderate, and low commitment levels. Participants with a high commitment level wanted to see information they could learn from to improve themselves, and how the information could be used to develop timely approaches for emerging health conditions. Others wanted to know if the data could foretell an illness. One participant wanted to review the clip of any fall that were to occur because "It could tell me what I shouldn't do in the future." Similarly, another stated, "I'd like to see me fall, and why I fell[horizontal ellipsis] I want to know what I'm doing wrong." Participants with a moderate commitment level had an interest in seeing the data for the sake of knowing. One resident remarked, "Well I'd like to know the results. I'm a nurse, so I would be interested." Finally, participants with a low commitment level did not have a desire to see their data or know what they would do with it, as stated: "Well there is nothing I would want to know. I'm glad the people upfront know what's going on in my room."

Family Interviews

Twelve of the 13 family members interviewed, who were children of participants, professed some understanding of the aims of the study and the sensor technology. One family member had only recently learned of his father's participation, since his father had not discussed the study with him before enrolling. Three themes emerged from these family interviews: (1) benefits of bed sensors, (2) family involvement and interaction with staff, and (3) protection of privacy versus sensor benefits.

Benefits of Sensors

Fall detection was family members' primary concern, and the majority (n = 11) reported that they were aware of the fall detection component of the study and regarded this feature as a main benefit. All family members reported that their parents had a history of falls, and two remarked that their parents had joined the study specifically because of this concern. Seven family members knew of at least one fall since the sensors had been installed, in the bathroom and closet. For this reason, these family members suggested that the range of the fall-detecting depth sensor be extended to include these locations.

The son of one participant recommended the depth sensor be repositioned in his mother's room in order to better capture falls: "I don't know if this is possible or not, but if you could install a ceiling sensor in the middle of the room that would cover a 360-degree area[horizontal ellipsis] with a wide-angle lens using the same system, it might be a little better."

Two family members mentioned that the system's fall detection could be especially helpful when the participant was unable to activate their fall alert pendant or pull the call light cord. One participant's son said, "In the case of my father, he's in an electric wheelchair. It's not always convenient to get to the push button. I think that if there was a sensor that allowed staff members to detect falls or spills that he couldn't notify them of, I could see that as being of very great value."

The benefits of the motion tracking feature also emerged from the data. Most (n = 8) mentioned that they were aware of the motion-sensing component of the sensor system. Three of these family members said that they knew the motion sensors were able to capture patterns of activity that were out of the ordinary. One son mentioned that the motion sensors could alert staff if his mother increased her frequency of getting out of bed at night, while the daughter of another participant hoped to learn whether her mother was keeping active. Two family members were interested in learning how the motion sensors captured data associated with urinary tract infections.

The daughter of a participant with visual, hearing, and cognitive impairment wondered if depth sensor footage could also be used to provide data on how her mother moved around her room that could be considered to adapt her environment to her needs.

Four family members discussed data collected by the hydraulic bed sensor and the benefits of such a system in tracking potential illnesses. One son was concerned about the possibility of his mother developing a cold or pneumonia and mentioned that the bed sensor could help to detect these conditions early. Another participant's daughter wondered what sort of bed motion was recorded by the bed sensor and whether a long period without movement would trigger an alert.

Family Involvement and Interaction With Staff

Five family members expressed an interest in receiving health alerts themselves, primarily in the form of a text message or email. One son reported that although his mother had not generated any actionable health alerts, he would like to receive updates about the state of the sensor system: "Even if it was a short letter or short email[horizontal ellipsis] It wouldn't have to be anything lengthy, but just to indicate that it is of value and that it's providing additional assistance over what she was receiving before the equipment was installed."

Six family members said that staff had communicated with them about the sensors in some capacity. Five family members said that they knew that a fall had been reported in their parent's apartment since the installation of the sensors, but were unsure whether the fall was caught by the depth sensor or by a staff member. Speaking to this ambiguity, one participant's daughter said, "I've always known about falls, and nobody's ever said, 'Oh, the sensor system told me.'"

Two family members had seen clips of falls, however; one recalled, "I can think of this one time she did fall, and they went back and had the video, to show her exactly what happened. She did fall, and they could play it back to see how she fell, what side, where she fell, and all." The second had seen multiple fall alert videos of his mother, who fell frequently over the course of the study.

Family members also wondered how the staff viewed the health data collected by the sensors and how they were trained to respond to alerts. One daughter specifically mentioned the issue of keeping staff trained in a high-turnover environment.

Privacy Versus Benefits

The majority (n = 7) of family members explicitly described their parent as unable to live independently, viewing the sensor system as a means of preserving their parent's health rather than promoting or regaining independence. Twelve family members expressed the belief that the sensor system could be helpful in maintaining health for their parent. Describing the feeling of security she gained, one participant's daughter said, "I think it's a little peace of mind knowing, the way I understand it, if there would be any movement that's unusual or a fall that it triggers some kind of sensor that notifies the staff immediately. So to me that's a lot of peace of mind, knowing that there there's an extra set of eyes watching her movement."

None of the family members expressed privacy concerns. In the words of one son, "I don't see it as an invasion of privacy, but as an extension of the care that's being provided her." Family members found the sensors to be an unobtrusive addition to their parent's apartment. One son remarked, "It's just been a very neutral part of her environment, as far as I'm concerned."

DISCUSSION

Participant Understanding

This study provided insight into the day-to-day experiences of older adults who were living with an embedded sensor suite in their home environments over an extended period of time. We were able to collect data over five points in time, which provides a unique opportunity to observe how perceptions and behavior toward the technology changed in this period. One discovery is an adjustment period in which participants learned about the purpose and functions of the sensors, and this was important in the technology acceptance process. Participants regarded the sensor system as a new feature that they acclimated to gradually. Our approach included an orientation process on the technology and also reinforcement of the purpose and how to operate the technology at face-to-face interviews or whenever the participant requested help. We found that one explanation of how the sensors functioned was not enough. After repeated explanations, participants had a better understanding of the sensors.

Participant Attitudes Toward Technology

This study indicates that, participant disposition toward the sensors improved over time. Once the adjustment period passed, most participants reported that the sensors did not negatively affect their daily living routine; rather, they reported on the usefulness of the sensor suite in maintaining good health and independence. Participants liked the fall detection feature of the system and found that it was helpful and reduced their concerns about falling. These finding are similar to previous work in which an adjustment period to living with the sensors was identified, followed by use of sensor data to improve health and well-being.8 In the initial adjustment period, it is essential to address any aesthetic or functional concerns expressed by the participant regarding the sensor equipment. A willingness to listen and respond to concerns are critical to a successful adjustment, as noted in the research of Le et al.16 Thanks to the longitudinal nature of this study, over time three different levels of commitment toward the technology became evident-high, medium, and low. Those with a high commitment level were more accepting and able to verbalize ways in which the technology would be a help to them. In this group, the sensor system served as an educational tool that helped them be more mindful and proactive about their health. Those at the moderate level were able to articulate that they knew the technology would be helpful, but they were less clear about how it might help them in the future-they had a general idea that the sensor system would be useful. Those with a low commitment seemed to passively accept the system and looked to the health professionals for direction.

Participants and Privacy

Throughout the course of the study, participants expressed few privacy concerns about both living with the sensors and the type of data the sensors were collecting, consistent with findings from previous research.6,8,13 The primary concerns about privacy were about the type of images collected by the depth sensor and the level of detail of these images, which might reflect very private tasks such as undressing and going to the bathroom. To address these concerns, depth images were purposely undefined and vague, and participants' concerns were alleviated when images were viewed. While we found that the population of older adults involved in this study was willing to compromise privacy for independence, not all may be so willing to compromise privacy as noted by Berridge,12 and privacy preferences should be acknowledged early in the technology adoption period.

Participants and Health Information

In this study, there was a range of willingness to share information; some were particular about who should have, others placed a few restrictions on access, and some did not care who had access to the data. However, most participants were willing to share health data collected by the sensors with healthcare personnel and designated family members. Given this range in preferences, it is important that older adults be given the opportunity to designate how and with whom to share their health information. It is also important to note that preferences did not change over time, and the consistency may be a result of long-standing family relationships and trust in the care being provided by their healthcare professionals. Participants had confidence that the healthcare team would know what to do with the data.

When participants were asked whether they wanted to see their own data, there was a similar range of responses. Some participants wanted to actively participate in improvement of their behavior and health, others wanted to see the data with no plan to use it, and some did not care to see their data at all. We categorized responses as indicative of high, moderate, and low commitment to the system. Previously, a fall was a reactive event, but with the sensor suite, participants could be more proactive and learn from their mistakes. The notion of introducing mindfulness into daily practices came across in some interviews. Participants with moderate commitment were interested in gauging how they were doing, but did not have a plan or thought in mind about how to use this information for improvement. Participants with low commitment were passively involved in their healthcare and quite willing for someone else to handle the care. It could be that low commitment may reflect disease status, and those further along in the progression of their illness may not have the energy or attention span to be more involved in their healthcare.

Family Perceptions

Family interviews revealed that family members had an overall positive opinion of the sensor system. While falls and fall detection were of the greatest concern, most family members also had some understanding of the role of motion and vital sign detection via the bed sensor.

Family members professed a limited understanding of the mechanisms through which staff received and responded to health alerts and were generally interested both in learning more about how these mechanisms work and in receiving health alerts themselves. Family members reported limited interaction with staff regarding the sensor system and alerts. It is always a delicate balance to determine the extent of family involvement with independent older adults. However, it is often the family members who are contacted by care staff when an older adult's health declines or when there is a crisis. Future work should seek to involve family members to the extent that HIPAA requirements permit. Family members should be those who have access to residents' medical information and are selected by residents.

Family members reported no privacy concerns and viewed the sensors as a helpful, unobtrusive addition to their parent's care. This finding is consistent with findings from previous studies in which family members articulated the importance of the sensors' capabilities to help maintain their loved ones' independence without compromising privacy.8

Lessons Learned-Practice Implications

The integration of technology into the living environment of older adults who had varying levels of familiarity with technology had its challenges. Over the course of the interview and data collection period, research staff learned from observations and comments of participants and family members best approaches when conducting this type of work. The content that emerged was organized into seven practice recommendations:

1. Explain and reinforce the purpose of the sensors multiple times to enhance participants' understanding of the technology and its ability to capture health information and falls. We found that there were more questions about the sensor technology during the first 6 months to 1 year from the installation of the system. Clear explanations and reinforcement of the purpose of the sensors cleared up any confusion that participants may have experienced during the course of the study.

2. Actively listen to the older adults' concerns and make adjustments in a timely manner. During the initial interview, some residents expressed concern over the placement of a sensor, the aesthetics of the system, or the way the sensor fit the decor of the environment. We found that addressing these concerns soon after we became aware of them helped with acceptance of the technology and gave the participant an active voice in the integration of the sensors.

3. Provide the opportunity for the older adult to have input into the location of the sensors. For instance, some older adults expressed concern about placement of the sensors in the bathroom. Other participants suggested the placement of sensors in areas the research team did not anticipate. The resident and family members are aware of where problems are likely to occur in the living environment, and the strategic placement of these sensors in key areas will assist with the success and relevance of the data collected.

4. Prior to the installation of sensor monitoring equipment and periodically afterward, check with the older adult about any privacy concerns that may be present or that emerge over time. While very few participants expressed privacy concerns, and those who did received additional training on the system (ie, a discussion of exactly what data are collected and who sees the data, a review of the fall image data), practitioners should be prepared to withdraw the technology from the living environment if privacy concerns cannot be addressed with training or a repositioning of the sensors.

5. Comply with HIPAA regulations when using health data generated from sensors. It is important to remember that health data are HIPAA-protected data and that the necessary permissions and consents need to be obtained before any health information can be shared. Involve designated family members and friends to the extent that HIPAA requirements permit. Once an understanding is reached about information sharing, the necessary permissions and releases should be documented. A periodic review of these sharing preferences should occur and any changes noted formally.

6. Provide the opportunity for older adults to determine how and with whom the health information generated from the sensors is shared. Our analysis revealed various levels of comfort with who sees the participants' health data, from not caring who sees it to the designation of a select few who should see it. Initial and periodic clarification as to who has access to the data will help to ensure that information is not shared inappropriately.

7. Provide initial and periodic education to family members and friends designated by the older adult to receive healthcare information. This education may include a discussion of how the sensors work (including how healthcare staff receive and act on data) and what data they collect. Several family members expressed a need for this type of training during the interview process. For family members designated by the older adult to receive sensor information, clinicians should be encouraged to tell whether a sensor-derived alert was of assistance. Incorporation of an ongoing family member education component to describe how sensors work, what data they collect, and how alerts are received and acted upon might be a mechanism to minimally involve family members without violating an older adult's preference for information sharing.

Working with older adults with fluctuating physical and cognitive functioning can be challenging when both introducing and living with technology. Following these recommendations can assist with technology acceptance on the part of older adult users and their families.

Limitations

Because this study was a content analysis of interviews conducted with 55 participants, who are mostly older adult Caucasian females representing residents living in 13 facilities, the results of this study are not generalizable. Another limitation is that given the age of the population group studied and that participants were living in assisted-living facilities, medical and/or cognitive functioning may have prevented long and lively interviews. Shorter responses to the questions may have occurred due to physical and cognitive limitations. This factor may explain why some participants took only a passive interest in the technology.26 However, the results provide important insights in older adults' adjustments to living with technology and privacy and information-sharing concerns and preferences. Additional research with a more heterogeneous population and across geographical areas and care settings is needed and may result in different findings.

CONCLUSIONS

Through a series of interviews over five points in time, this study explored the perceptions of older adults on the usefulness of a sensor suite environmentally embedded in an independent/assisted-living home environment, their willingness to use the system, and its impact on daily routine and privacy. Family members' input was also obtained in an attempt to seek out their opinions of the sensor system. For this sample population, the sensor suite was regarded as helpful in maintaining independence, good health, and physical functioning. Input from older adults and families suggests that the willingness to adopt the sensor suite was motivated by both a decline in functional status and a desire to remain independent and avoid the next level of care-a nursing home setting. Three levels of commitment toward the sensors emerged from the data: high or active commitment; moderate, which included clear understanding of the technology and articulation that it is helpful; and low commitment, where there was minimal understanding and a passive acceptance of the system. The ability of the study participants to maintain their independence was viewed as more important than any compromises in privacy that might occur when using the technology. Participants were willing to share their health data with healthcare providers and select family members. That being said, many older adults in this study wanted input into who sees their data. Older adults should be involved in all stages of sensor installation, and their input should be taken into account by researchers and technicians alike. Recommendations for future practice were provided based on our data analysis.

References