Abstract
Introduction:
We developed the Technology in Caring Questionnaire (TCQ) to assesses use of technology-based strategies by dementia caregivers.
Methods:
100 caregivers completed a survey that included TCQ items along with measures of technology proficiency and patient- and caregiver-centered outcomes.
Results:
The final 34-item TCQ scale had adequate to excellent internal consistency (raw Cronbach’s alpha = .75; standardized Cronbach’s alpha = .95; Guttman’s Lambda-6 = .97). TCQ scores demonstrated modest convergent associations with scores from measures of smartphone (r = .265, p < .01) and computer proficiency (r = .230, p < .05) but a strong association with overall technology experience scores (r = .578, p < .001). Elevated TCQ scores were associated with reduced informant-reported cognitive symptoms (B = −0.003, p < .05), increased ability of caregivers to find support and information (B = 0.03, p < .001), and increased direct care strain (B = .03, p < .05), after controlling for dementia severity and demographics.
Discussion:
The TCQ has good psychometric properties for assessment of technology-based care strategies among dementia caregivers. Findings imply that use of technologies may aid in symptom management and finding support and information but may also increase caregiver strain.
Keywords: Caregiver, dementia, technology, cognition, burden, psychometrics
The Technology in Caring Questionnaire (TCQ)
Caregivers to persons with dementia (PWD) are increasingly using technology-based strategies to manage symptoms and improve quality of life.1,2 Commonly owned technologies, like smartphones, smart speakers, and wearable devices, have an array of features that could potentially support dementia care, including automated reminders, location tracking, and activity monitoring.3,4 Evidence suggests these features may help limit injuries,5,6 reduce neurobehavioral symptoms,6-8 and improve well-being among PWD.9 However, studies of the benefits of technology use on independence of PWD and burden on caregivers have thus far been disappointing.10-13 Thus, there is a need to better understand how caregivers perceive, choose, and implement technologies.
There have been a few qualitative and survey studies on these topics. Hunter et al.14 interviewed 12 caregivers on their perceptions of mobility aids, finding that they viewed them as a tool to increase independence of PWD. A broader interview study with 21 caregivers found that caregivers used technologies for a variety of goals, including tracking rest and sleep, promoting leisure, facilitating social connectedness, managing instrumental activities of daily living, and accessing health care.2 Some potential disadvantages of technologies were also noted, including financial costs, usability issues, and privacy concerns. Survey studies also suggested moderate to high rates of technology use among dementia caregivers, ranging from 44% to 91%, depending on the modality.15-18 Similar to qualitative study participants, survey participants noted advantages and disadvantages of technology-based strategies.
These studies also brought to light the importance of taking a more structured and psychometrically robust approach to assessing technology-based strategies among dementia caregivers. Only one of the studies reviewed above developed a pool of questionnaire items to assess technology-based strategies,18 and none of the studies provided evidence of the reliability and validity of their measurement approach. We sought to fill this gap in the literature by developing the Technology in Caring Questionnaire (TCQ) and examining its psychometric properties.
The current work had four goals. First, we developed and refined the TCQ measure. Second, we evaluated the internal consistency of TCQ scores. Third, we assessed their construct validity by examining relationships between TCQ scores and scores from measures of technology use. We expected to observe significant positive correlations across the technology use measures with the caveat that there should be substantial non-overlapping variance specifically reflecting use of technology-based strategies for dementia care. Fourth, we assessed whether TCQ scores provided insight into modern hypotheses of technology use in aging and dementia. For example, the technological reserve hypothesis states that older adults who use technology will display greater resilience to cognitive and functional decline that occurs in aging and neurodegenerative diseases.19,20 A logical extension of this technological reserve hypothesis is that use of technology-based strategies may improve dementia care. Thus, we examined the extent to which TCQ scores were related to patient- and caregiver-centered outcomes. We expected higher use of technology-based care strategies to be associated with better outcomes.
Methods
Participants
Inclusion criteria included self-identification as someone who provides care on at least a weekly basis to a PWD of at least 50 years of age. Participation was restricted to unpaid caregivers (i.e., family members or friends providing care) residing in the United States. Demographics of the sample are described in Table 1. Caregiver relationships to PWD included being a spouse/significant other (37%), daughter (27%), other family member (22%), son (4%), daughter-in-law (3%), and friend (3%; the remaining 4% selected other). Participants reported seeing the PWD every day (59%), 4-6 days per week (17%), 2-3 days per week (12%), or once per week (11%), while 1% did not respond. Forty-six percent of the PWD lived independently, 40% at a family member’s home, 6% at an assisted living facility, 4% at an independent living facility, 2% at a skilled nursing care facility, 1% at a memory care unit, and 1% did not respond. Fifty-seven percent of the caregivers lived with the PWD, 38% lived separately, 4% said they lived with the PWD some of the time, and 1% did not respond. On average, caregivers reported spending 87.92 hours per week (SD = 64.03) with PWD and had known the PWD for 40.10 years (SD = 19.18).
Table 1.
Demographics
| Caregiver | Person with Dementia | |
|---|---|---|
| Age M (SD) | 53.88 (17.98) | 73.85 (10.40) |
| Gender | ||
| Female | 73 (73%) | 60 (60%) |
| Male | 27 (27%) | 39 (39%) |
| Non-binary | 0 (0%) | 1 (1%) |
| Ethnicity | ||
| Non-Hispanic | 91 (91%) | 91 (91%) |
| Hispanic | 9 (9%) | 9 (9%) |
| Race | ||
| White | 83 (83%) | 80 (80%) |
| Black | 9 (9%) | 10 (10%) |
| Hispanic/Latinx | 3 (3%) | 5 (5%) |
| Asian | 3 (3%) | 3 (3%) |
| Native Hawaiian or Pacific Islander | 1 (1%) | 1 (1%) |
| Native American or Alaska Native | 0 (0%) | 1 (1%) |
| Multiracial | 1 (1%) | 0 (0%) |
| Education Years M (SD) | 16.13 (2.15) | 14.50 (3.13) |
| Income per year | ||
| <$30,000 | 13 (13%) | 26 (26%) |
| $30,000-$59,999 | 24 (24%) | 33 (33%) |
| $60,000-$100,000 | 30 (30%) | 13 (13%) |
| >$100,000 | 31 (31%) | 21 (21%) |
| No response | 2 (2%) | 7 (7%) |
Measures
Technology in Caring Questionnaire (TCQ)
An initial set of 19 items for the TCQ were generated by authors with clinical experience and/or clinical trial experience with technology use among PWD (AK, MS, JB). Four items were then added based on a review of the literature and survey research in a memory clinic.21 Another 13 items were added based on qualitative interviews with caregivers,2 resulting in a final pool of 36 items. Each item in the TCQ asks the caregiver to rate the frequency of use of a given technology-based strategy for helping someone who is experiencing cognitive impairment or dementia. Item responses are on a 5-point scale (0-never, 1-rarely, 2-sometimes, 3-often, 4-very often). The original TCQ items are available in the supplemental materials.
Measures of Technology Use and Proficiency
To evaluate the construct validity of the TCQ, we examined relationships of TCQ scores with measures of technology use and proficiency.
Computer Proficiency Questionnaire (CPQ).
The CPQ is a 33-item self-report measure of proficiency in computer skills.22 This measure yields one global scale made up of six subscales: (1) computer basics, (2) printer use, (3) communication, (4) internet use, (5) calendar use, and (6) entertainment. Item scores range from one to five with higher scores indicating greater ability to perform tasks with a computer. The CPQ is scored by computing the average of each subscale and then summing the averages to obtain the total CPQ score.
Mobile Device Proficiency Questionnaire (MDPQ).
The MDPQ is a 46-item, self-report measure of the respondent’s ability to perform operations on a mobile device.23 This measure yields eight subscales: (1) mobile device basics, (2) communication, (3) data and file storage, (4) internet use, (5) calendar use, (6) entertainment, (7) privacy, and (8) troubleshooting and software management. Item scores range from one to five with higher scores indicating greater ability to perform tasks with a mobile device. The MDPQ is scored by computing the average of each subscale and then summing the averages to find the total MDPQ score.
Technology Experience Profile (TEP).
The TEP is a 36-item self-report measure that evaluates respondent’s experience with a range of technologies.24 The TEP is scored using the frequency profile, which involves first recoding the scale into four points (0 = never used to 3 = frequently used) and then averaging across 36 items. Higher scores indicate greater breadth and frequency of technology use.
Patient-Centered Outcomes
Everyday Cognition (ECog).
The ECog is a 39-item, informant-reported measure of everyday cognitive function among older adults.25 Item scores range from one to four with higher scores indicating more cognitive difficulties. The total score consists of an average of answered items.
Functional Activities Questionnaire (FAQ).
The FAQ is a 10-item informant-based measure of functioning in instrumental activities of daily living.26 Item scores range from zero to three and are summed to create a total score wherein higher scores indicate greater functional impairment.
Caregiver-Centered Outcomes
Scales Measuring the Impact of Dementia on Carers (SIDECAR).
The SIDECAR is a 39 item, self-report measure of the impacts of dementia care on caregivers.27 This measure yields three subscales: (1) direct impact (strain associated with caring; e.g., “Caring prevents me from fulfilling my other responsibilities”), (2) indirect impact (empathic distress for the PWD; e.g., “It is distressing to see the person I care for changing”), and (3) support and information (ability to obtain support and information regarding caregiving; e.g., “I have enough guidance to know how to provide care”). Subscales are scored by summing items within each domain. Higher scores on the direct and indirect impact scales indicate greater strain and distress, respectively. Higher scores on the support and information scale indicate greater ability to obtain support and information.
Control Variables
Quick Dementia Rating System (QDRS).
We wanted to control for dementia severity in analyses because it is likely related to patient- and caregiver-centered outcomes. Thus, we administered the QDRS, a 10-item, informant-reported dementia staging tool.28 Respondents endorse items describing functioning in cognition, mood, and everyday activities. Item scores range from zero to three. Items are summed to yield a total score, with higher scores indicating a later, more severe stage of dementia. In addition, QDRS scores can be used to categorically stage severity of cognitive impairment as follows: 0-1 = normal, 2-5 = mild cognitive impairment, 6-12 = mild dementia, 13-20 = moderate dementia, ≥21 = severe dementia. In the current sample, the breakdown of severity was as follows: 9 mild cognitive impairment, 50 mild dementia, 33 moderate dementia, 7 severe dementia, and 1 no response.
Procedures
Participants were recruited using several methods. First, we made in person or phone contact with individuals presenting to a neuropsychology clinic at an academic medical center. Second, the study was advertised via print flyers at local clinics, community organizations, and businesses. Third, we reached out via phone or email to a variety of caregiver support groups, Alzheimer’s Association chapters, nursing homes, adult day care centers, and churches. Fourth, we advertised the study online through a university announcement system, a university newsletter, the Alzheimer’s Association’s TrialMatch site, social media, and patient registries.
The survey was distributed online because data collection began during the COVID-19 pandemic. Potential participants were instructed to contact the study coordinator, who verified eligibility and provided a Qualtrics survey link if the individual was eligible. Upon accessing the survey, participants completed a captcha and pre-screening questions to confirm their eligibility. They then completed online consent procedures and responded to the survey. The survey included the instruments described in the measures section; initial items requested demographic information, with remaining measures presented in random order. Upon completion of the survey, participants were directed to a separate instrument to provide contact information to receive payment. The survey took participants roughly 60 minutes to complete, and participants received a $20 Amazon eGift Card as compensation.
Quality Assurance
Validity of survey responses was confirmed through a strict quality assurance process. Cases were eliminated if any of the following criteria were met: 1) >75% of the data were missing; 2) >1 of 4 attention check items were answered incorrectly (e.g., “If you are paying attention, please select “Questionable/Occasionally worse”); 3) the response was a duplicate (as identified by Qualtrics duplicate tag or use of the same IP address); 4) the survey was completed in less than 30 minutes; 5) the response was received from outside the U.S. (the only locale where recruitment was sought); or 6) the response was suspicious for other reasons (e.g., participant indicated they accessed the survey through a non-advertised route).
Analyses
TCQ Item Pool Refinement
We examined item endorsement rates and item-total correlations for the TCQ items. Items were deleted if 1) >75% of participants provided a response of “never”;29 or 2) the item-total correlation was <.20.30
Reliability
After finalizing the TCQ item pool, we examined several indicators of internal consistency reliability, including the raw Cronbach’s alpha, the standardized Cronbach’s alpha, and the Guttman’s Lambda-6 value.
TCQ and Demographics
Because technology use may vary as a function of demographic factors,31 we examined the associations of TCQ scores with caregiver age, sex, education, and income via Pearson (for continuous variables) or point-biserial (for dichotomous variables) correlation coefficients.
Construct Validity
To evaluate convergent validity of the TCQ, we examined Pearson correlations of TCQ scores with scores from measures of technology use and comfort, including the CPQ, the MDPQ, and the TEP.
Technology-Based Strategies and Patient- and Caregiver-Centered Outcomes
We examined the relationship of TCQ scores with scores on measures of informant-reported cognitive symptoms (ECog), informant-reported functioning in instrumental activities of daily living functioning (FAQ), and impacts on the caregiver (SIDECAR). For each outcome where the TCQ had a significant bivariate association, a hierarchical regression was conducted. We first entered control variables, including dementia severity (QDRS total score) and demographic factors. Control variables were selected using a confound-based, data-driven approach.32 Specifically, if the QDRS or a demographic variable had a significant correlation with the TCQ or with the outcome variable, it was included as a covariate in Step 1. Next, the TCQ was entered into the model, and we examined its regression weight and the change in R2 for significance.
Results
TCQ Item Pool Refinement
To refine the TCQ item pool, we examined the response frequencies and item-total correlations for each item (see Table 2). None of the items showed higher than 75% response rate for the “Never” anchor. The most commonly endorsed technology-based care strategies focused on use of online banking (item 13; 91%), autopay features (item 36; 91%), and electronic health portals or applications (item 4; 95%). However, item 13 (“I check online bank, credit card, and other financial statements to make sure nothing is wrong or unusual with the finances of the person I care for”) and item 36 (“I use auto-pay or automatic withdrawal methods to help the person I care for manage finances”) showed low item-total correlations and were removed from further analysis.
Table 2.
The Item-Total Correlation and Frequencies of Responses for Each Original Item on the Technology in Caring Questionnaire
| Item | Item-Total Correlation |
Very Often | Often | Sometimes | Rarely | Never |
|---|---|---|---|---|---|---|
| 1 | .419*** | 46 (46%) | 18 (18%) | 8 (8%) | 7 (7%) | 20 (20%) |
| 2 | .711*** | 19 (19%) | 10 (10%) | 20 (20%) | 7 (7%) | 43 (43%) |
| 3 | .750*** | 16 (16%) | 13 (13%) | 16 (16%) | 7 (7%) | 47 (47%) |
| 4 | .245* | 46 (46%) | 26 (26%) | 16 (16%) | 6 (6%) | 5 (5%) |
| 5 | .663*** | 17 (17%) | 14 (14%) | 18 (18%) | 3 (3%) | 47 (47%) |
| 6 | .645*** | 14 (14%) | 18 (18%) | 20 (20%) | 4 (4%) | 43 (43%) |
| 7 | .278** | 34 (34%) | 32 (32%) | 14 (14%) | 3 (3%) | 16 (16%) |
| 8 | .631*** | 14 (14%) | 9 (9%) | 14 (14%) | 11 (11%) | 51 (52%) |
| 9 | .606*** | 17 (17%) | 15 (15%) | 12 (12%) | 4 (4%) | 51 (52%) |
| 10 | .518*** | 16 (16%) | 12 (12%) | 8 (8%) | 9 (9%) | 54 (55%) |
| 11 | .707*** | 13 (13%) | 13 (13%) | 9 (9%) | 6 (6%) | 58 (59%) |
| 12 | .668*** | 13 (13%) | 11 (11%) | 11 (11%) | 3 (3%) | 61 (62%) |
| 13 | .107 | 50 (51%) | 20 (20%) | 16 (16%) | 4 (4%) | 9 (9%) |
| 14 | .476*** | 20 (20%) | 15 (15%) | 17 (17%) | 11 (11%) | 36 (36%) |
| 15 | .596*** | 9 (9%) | 13 (13%) | 11 (11%) | 2 (2%) | 64 (65%) |
| 16 | .734*** | 18 (18%) | 13 (13%) | 16 (16%) | 7 (7%) | 45 (45%) |
| 17 | .731*** | 8 (8%) | 21 (21%) | 24 (24%) | 14 (14%) | 32 (32%) |
| 18 | .718*** | 7 (7%) | 15 (15%) | 14 (14%) | 7 (7%) | 56 (57%) |
| 19 | .594*** | 24 (24%) | 20 (20%) | 24 (24%) | 11 (11%) | 20 (20%) |
| 20 | .546*** | 27 (27%) | 22 (22%) | 24 (24%) | 11 (11%) | 15 (15%) |
| 21 | .643*** | 15 (15%) | 17 (17%) | 19 (19%) | 10 (10%) | 38 (38%) |
| 22 | .526*** | 22 (22%) | 16 (16%) | 11 (11%) | 9 (9%) | 41 (41%) |
| 23 | .486*** | 23 (23%) | 20 (20%) | 25 (25%) | 16 (16%) | 15 (15%) |
| 24 | .761*** | 10 (10%) | 12 (12%) | 21 (21%) | 14 (14%) | 42 (42%) |
| 25 | .483*** | 32 (32%) | 26 (26%) | 20 (20%) | 5 (5%) | 16 (16%) |
| 26 | .722*** | 21 (21%) | 15 (15%) | 12 (12%) | 10 (10%) | 41 (16%) |
| 27 | .652*** | 16 (16%) | 21 (21%) | 19 (19%) | 9 (9%) | 34 (34%) |
| 28 | .688*** | 12 (12%) | 10 (10%) | 18 (18%) | 4 (4%) | 55 (56%) |
| 29 | .688*** | 6 (6%) | 6 (6%) | 12 (12%) | 8 (8%) | 67 (68%) |
| 30 | .720*** | 3 (3%) | 10 (10%) | 22 (22%) | 9 (9%) | 55 (56%) |
| 31 | .648*** | 13 (13%) | 12 (12%) | 25 (25%) | 7 (7%) | 42 (42%) |
| 32 | .379*** | 38 (38%) | 12 (12%) | 16 (16%) | 6 (6%) | 27 (27%) |
| 33 | .620*** | 12 (12%) | 16 (16%) | 28 (28%) | 14 (14%) | 29 (29%) |
| 34 | .759*** | 7 (7%) | 6 (6%) | 11 (11%) | 6 (6%) | 69 (70%) |
| 35 | .661*** | 23 (23%) | 25 (25%) | 10 (10%) | 5 (5%) | 36 (36%) |
| 36 | .062 | 51 (52%) | 12 (12%) | 20 (20%) | 7 (7%) | 9 (9%) |
Note. N = 99 for all items. * = p < .05, ** = p < .01, *** = p < .001.
Reliability
The internal consistency of the final 34-item TCQ total scale was adequate to excellent, depending on the index examined (raw Cronbach’s alpha = .75; standardized Cronbach’s alpha = .95; Guttman’s Lambda-6 = .97).
TCQ and Demographics
Relationships of TCQ scores with demographic factors are presented in Table 3. TCQ scores were strongly negatively associated with age (r = −.567, p < .001), indicating that younger caregivers more frequently employed technology in caregiving. TCQ scores were not significantly associated with sex, education, or income.
Table 3.
Bivariate Correlation Between Study Variables
| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1. TCQ | -- | ||||||||||||
| 2. CPQ | .230* | -- | |||||||||||
| 3. MDPQ | .265** | .769*** | -- | ||||||||||
| 4. TEP | .578*** | .507*** | .567*** | -- | |||||||||
| 5. Age | −.567*** | −.102 | −.122 | −.336*** | -- | ||||||||
| 6. Sex | −.173 | .084 | .134 | .105 | .010 | -- | |||||||
| 7. Education | .140 | .314** | .337** | .289** | .113 | −.058 | -- | ||||||
| 8. Income | .185 | .255* | .207* | .352*** | −.027 | −.068 | .452*** | -- | |||||
| 9. QDRS | −.254* | −.059 | −.080 | −.026 | .206 | .157 | −.025 | −.049 | -- | ||||
| 10. ECog | −.445*** | −.090 | −.105 | −.162 | .392*** | .238* | −.022 | −.065 | .695*** | -- | |||
| 11 FAQ | −.253* | .056 | .016 | −.049 | .361*** | .099 | .137 | .199 | .542*** | .623*** | -- | ||
| 12. SIDECAR-d | .272** | −.074 | −.042 | −.006 | −.259* | −.020 | −.029 | −.127 | .283** | .171 | .085 | -- | |
| 13. SIDECAR-i | .147 | −.020 | −.027 | −.108 | −.086 | −.100 | .068 | −.047 | .115 | .037 | −.132 | .613*** | -- |
| 14.SIDECAR-s | .391*** | .009 | .120 | .407*** | −.122 | .026 | .135** | .313** | .026 | −.045 | .074 | −.178 | −.273** |
| N | 99 | 96 | 99 | 98 | 100 | 100 | 100 | 98 | 99 | 99 | 99 | 97 | 99 |
Note. Demographic variables reflect caregiver characteristics. Sex coded as male = 1, female = 2. TCQ = Technology in Caring Questionnaire; CPQ = Computer Proficiency Questionnaire; MDPQ = Mobile Device Proficiency Questionnaire; TEP = Technology Experience Profile; QDRS = Quick Dementia Rating System; Ecog = Everyday Cognition Scale; FAQ = Functional Activities Questionnaire; SIDECAR-d = Scales Measuring the Impact of Dementia on Carers - direct impact; SIDECAR-i = indirect impact; SIDECAR-s = support and information; N = sample size. *** p < .001, ** p <.01, * p <.05.
Construct Validity
Relationships of TCQ scores with other measures of technology use and comfort are presented in Table 3. TCQ scores had small-to-moderate positive correlations with scores from measures of proficiency with computers (r = .230, p = .025) and smartphones (r = .265, p = .008). Further, TCQ scores were strongly positively associated with scores from a measure of overall breadth and frequency of technology use (r = .578, p < .001).
TCQ and Patient-Centered Outcomes
Relationships of TCQ scores with scores from patient-centered outcomes are presented in Table 3. Higher TCQ scores were associated with fewer cognitive symptoms (r = −.445, p < .001), less functional impairment (r = −.253, p = .012), and reduced dementia severity (r = −.254, p = .011). Table 4 displays the results of the hierarchical linear regressions. The relationship between TCQ scores and functional impairment was no longer significant with covariates in the model. However, the TCQ explained modest, but significant incremental variance in cognitive symptoms (ΔR2 = 1.9%), after controlling for demographic factors and dementia severity.
Table 4.
Hierarchical Regression Results
| Models and Variables |
ECog | FAQ | SIDECAR-d | SIDECAR-s |
|---|---|---|---|---|
| Step 1 | ||||
| age | 0.01 (0.003)*** | 0.10 (0.03)** | −0.07 (0.02)*** | −0.02 (0.01) |
| sex | 0.21 (0.10)* | -- | -- | -- |
| education | -- | -- | -- | 0.01 (0.15) |
| income | -- | -- | -- | 0.95 (0.32)** |
| QDRS | 0.08 (0.01)*** | 0.65 (0.11)*** | 0.25 (0.07)*** | 0.05 (0.05) |
| F | 41.45 | 27.14 | 10.46 | 3.52 |
| df1, df2 | 3, 94 | 2, 95 | 2, 93 | 4, 91 |
| Total R 2 | 0.569 | 0.363 | 0.183 | 0.096 |
| Step 2 | ||||
| age | 0.01 (0.003)* | 0.11 (0.04) | −0.04 (0.02) | 0.01 (0.02) |
| sex | 0.17 (0.10) | -- | -- | -- |
| education | -- | -- | -- | −0.07 (0.15) |
| income | -- | -- | -- | 0.79 (0.30)** |
| QDRS | 0.07 (0.01)*** | 0.64 (0.11)*** | 0.27 (0.07)*** | 0.07 (0.05) |
| TCQ | −0.003 (0.002)* | .003 (0.02) | 0.03 (0.01)* | 0.04 (0.01)*** |
| F | 33.23 | 17.66 | 9.18 | 2.66 |
| df1, df2 | 4, 93 | 3, 93 | 3, 92 | 5, 90 |
| Total R 2 | 0.588 | 0.363 | 0.230 | 0.251 |
| R2 Change | 0.019* | −0.001 | 0.047* | 0.117*** |
Note. QDRS = Quick Dementia Rating System; ECog = Everyday Cognition Scale; FAQ = Functional Activities Questionnaire; SIDECAR-d = Scales Measuring the Impact of Dementia on Carers - direct impact; SIDECAR-s = support and information; df = degree of freedom. *** p < .001, ** p <.01, * p <.05.
TCQ Caregiver-Centered Outcomes
Relationships of TCQ scores with scores from caregiver-centered outcomes are presented in Table 3. TCQ scores were positively associated the caregivers’ perceived ability to obtain support and information (r = .391, p < .001). TCQ scores were also positively associated with greater care strain (r = .272, p < .001). No significant association was observed between TCQ scores and indirect impacts on the caregiver. As shown in Table 4, hierarchical linear regressions indicated that, after controlling for covariates, TCQ scores explained 11.7% incremental variance in the support and information domain and 4.7% incremental variance in the direct impact domain.
Discussion
The current study provided evidence for a psychometrically robust measure of technology-based care strategy use among dementia caregivers, the Technology in Caring Questionnaire (TCQ). The goals of the study were to 1) develop and refined the TCQ measure; 2) assess the internal consistency of TCQ scores; 3) examine the construct validity of the TCQ; and 4) evaluate the TCQ in the context of the technological reserve hypothesis.
TCQ Item Endorsement, Discrimination, and Reliability
The most commonly endorsed technology-based care strategies focused on use of online banking, autopay features, and electronic health portals or applications. Digital methods for managing finances were also commonly endorsed by older adults in prior research.33 However, digital strategies for financial management did not discriminate between caregivers high and low in technology-based strategy use (as reflected by very small, non-significant item-total correlations), probably because they were so ubiquitously endorsed. As a result, these two items were removed from the TCQ. The final 34-item TCQ scale had adequate to excellent internal consistency and is available in the supplemental materials.
TCQ and Demographics
There was a strong, negative association between TCQ scores and age, suggesting that younger caregivers are likely to use technology-based care strategies. This finding is consistent with prior research on the age-related digital divide in technology use.31 Surprisingly, TCQ scores were not significantly associated with education and income, as has been reported in prior literature.31 This lack of association may reflect that we used an online survey because data collection began during the COVID-19 pandemic. As a result, participants had to be educated and wealthy enough to have the technology and skills to access the survey in the first place. However, it should be noted that there was no obvious restriction in range in education or income (see Table 1). An alternative interpretation then is that there truly is no significant relationship of education and income with technology-based strategy use among dementia caregivers. To confirm this possibility, findings could be replicated using a paper and pencil approach.
Construct Validity
We expected TCQ scores to demonstrate significant, positive associations with scores from measures of technology use. However, we also anticipated that the TCQ would have non-overlapping variance with scores from these measures. Consistently, TCQ scores demonstrated modest positive associations with scores from existing, widely used measures of mobile device and computer proficiency but a strong positive association with scores from a broader measure of technology use. Three key conclusions can be drawn from this pattern of associations. First, the TCQ does appear to capture technology use, given positive associations with all technology use measures. Second, while TCQ scores likely reflect familiarity with specific devices to a degree, they are better indicators of the breadth and frequency of overall technology use, given the larger correlation with the TEP, as compared to the CPQ and MDPQ. Finally, TCQ scores also capture something unique beyond simple technology proficiency, as >65% of the variance was non-overlapping. We believe this unique variance in part reflects use of technology-based strategies for providing care to PWD, and the TCQ is the first rigorously developed and tested measure of this construct.
The TCQ and Patient-Centered Outcomes
In line with the technological reserve hypothesis,19,20 we examined relationships of TCQ scores with patient- and care-partner centered outcomes. Greater use of technology-based care strategies was associated with fewer informant-reported cognitive symptoms and less functional impairment in instrumental activities of daily living. The relationship with cognitive symptoms held when controlling for demographic factors and dementia severity, but the relationship with functional impairment did not. This finding is perhaps unsurprising. Caregivers might be able to use technology-based strategies to help the PWD to mask day-to-day cognitive difficulties. For example, by using a digital calendar to help the PWD track their schedule, the PWD may less frequently ask for reminders about upcoming events. As a result, the caregiver might perceive the PWD to have fewer memory problems; however, it is unlikely that the caregiver would describe the PWD as more independent, since the caregiver is the one implementing the strategy. Of course, the direction of effect between technology-based strategy use and symptom severity could be reversed. That is, it may be that caregivers to individuals with more severe symptoms are more likely to turn to technology-based strategies. We attempted to control for this possibility by including the QDRS total score as a covariate in the regression models, and the relationship between TCQ scores and subjective cognitive concerns held even when controlling for dementia severity. Nonetheless, longitudinal research is necessary to confirm that technology-based strategy use is associated with improved symptom management over time.
The TCQ and Caregiver-Centered Outcomes
In addition to patient-centered outcomes, we examined how TCQ scores related to impacts on the caregiver. Greater use of technology-based strategies was most strongly related to the ability of caregivers to obtain information and support, explaining almost 12% incremental variance beyond demographics and dementia severity. This finding likely reflects that a number of TCQ items cover use of technologies to obtain support (e.g., “To cope with stress that comes up from providing care, I reach out to my support network via calls, texts, or emails”; “I look for caregiving help and advice on websites”).
Interestingly, increasing use of technology-based care strategies was associated with greater direct care strain. One possible explanation of this finding is that while technology-based strategies may improve management of dementia symptoms, they could also make life more difficult for caregivers. Qualitative research has found that caregivers can view technologies as frustratingly complex and too costly.2 Thus, there may be direct emotional and financial burdens associated with implementing technology-based care strategies. Results imply that we may need to update measures of care burden or strain to address the influence of technologies. An alternative interpretation of our results may be that caregivers experiencing greater strain are more likely to turn to technologies as a solution. Again, we attempted to control for this possibility by including QDRS total score as a covariate in the regression model, and the relationship between TCQ scores and direct impacts of care held even when controlling for dementia severity.
Limitations and Future Directions
These findings should be interpreted in the context of some additional limitations. First, measures used in the current study were all subjective self-reports. Examining the TCQ in relation to objective measures of technology use would provide further evidence of validity. Second, though our sample included a good cross-section of income groups, it was skewed toward non-Hispanic White women, such that the TCQ should be evaluated in a more diverse sample. Third, it should be noted that technology advances rapidly. Though we tried to design TCQ items to be general enough to apply across times and situations, the measure will need to be updated as new technology-based care strategies emerge.
Conclusions
These limitations notwithstanding, the current study provides evidence for the reliability and validity of scores from the TCQ in assessing technology-based strategies among dementia caregivers. This measure could be used a clinical outcome in trials designed to enhance technology use by dementia care partners and also holds promise as a clinical tool for care support and planning. Findings indicated that higher scores on the TCQ (i.e., greater technology-based care strategy use) are associated with improved management of dementia symptoms and enhanced caregiver ability to find support and information. However, using technologies may also increase the strain of care to a degree, potentially due to costs and complexity. Further longitudinal research is needed to examine test-retest reliability of TCQ and clarify directions of effects.
Supplementary Material
Conflicts of Interest:
The authors have no conflicts of interest to declare. This work was funded under a Career Development Award by the National Institute on Aging (NIA) of the National Institutes of Health under Award Number U54AG063546, which funds NIA Imbedded Pragmatic Alzheimer’s and AD-Related Dementias Clinical Trials Collaboratory (NIA IMPACT Collaboratory). Additional support was provided by NIH AG082783. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
References
- 1.Madara Marasinghe K. Assistive technologies in reducing caregiver burden among informal caregivers of older adults: a systematic review. Disability and Rehabilitation: Assistive Technology. 2016;11(5):353–360. [DOI] [PubMed] [Google Scholar]
- 2.Mikula CM, Boone A, Perry C, Benge JF, Scullin MK. Dementia caregiver insights on use of assistive technologies. Work, Aging, and Retirement. 2022;e-publicatio ahead of print [Google Scholar]
- 3.Fabricatore C, Radovic D, Lopez X, Grasso-Cladera A, Salas CE. When technology cares for people with dementia: A critical review using neuropsychological rehabilitation as a conceptual framework. Neuropsychological Rehabilitation. Sep 13 2020;30(8):1558–1597. doi: 10.1080/09602011.2019.1589532 [DOI] [PubMed] [Google Scholar]
- 4.Brims L, Oliver K. Effectiveness of assistive technology in improving the safety of people with dementia: a systematic review and meta-analysis. Aging & Mental Health. Aug 3 2019;23(8):942–951. doi: 10.1080/13607863.2018.1455805 [DOI] [PubMed] [Google Scholar]
- 5.Neubauer NA, Lapierre N, Rios-Rincon A, Miguel-Cruz A, Rousseau J, Liu L. What do we know about technologies for dementia-related wandering? A scoping review: Examen de la portee : Que savons-nous a propos des technologies de gestion de l'errance liee a la demence? Canadian Journal of Occupational Therapy-Revue Canadienne D Ergotherapie. Jun 2018;85(3):196–208. doi: 10.1177/0008417418777530 [DOI] [PubMed] [Google Scholar]
- 6.Chan DKY, Chan LKM, Kuang YM, Le MNV, Celler B. Digital care technologies in people with dementia living in long-term care facilities to prevent falls and manage behavioural and psychological symptoms of dementia: a systematic review. European Journal of Ageing. 2021;doi: 10.1007/s10433-021-00627-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Saragih ID, Tonapa SI, Sun T-L, Chia-Ju L, Lee B-O. Effects of robotic care interventions for dementia care: A systematic review and meta-analysis randomised controlled trials. Journal of Clinical Nursing. Nov 2021;30(21-22):3139–3152. doi: 10.1111/jocn.15856 [DOI] [PubMed] [Google Scholar]
- 8.Anderson MS, Homdee N, Bankole A, et al. Behavioral Interventions for Alzheimer's Management Using Technology: Home-Based Monitoring. Current Geriatrics Reports. Jun 2020;9(2):90–100. doi: 10.1007/s13670-020-00312-y [DOI] [Google Scholar]
- 9.Sweeney L, Clarke C, Wolverson E. The use of everyday technologies to enhance well-being and enjoyment for people living with dementia: A systematic literature review and narrative synthesis. Dementia-International Journal of Social Research and Practice. May 2021;20(4):1470–1495. 1471301220929534. doi: 10.1177/1471301220929534 [DOI] [PubMed] [Google Scholar]
- 10.Leroi I, Woolham J, Gathercole R, et al. Does telecare prolong community living in dementia? A study protocol for a pragmatic, randomised controlled trial. Trials. Oct 23 2013;14349. doi: 10.1186/1745-6215-14-349 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Howard R, Gathercole R, Bradley R, et al. The effectiveness and cost-effectiveness of assistive technology and telecare for independent living in dementia: a randomised controlled trial. Age and ageing. 2021;50(3):882–890. doi: 10.1093/ageing/afaa284 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Davies A, Brini S, Hirani S, et al. The impact of assistive technology on burden and psychological well-being in informal caregivers of people with dementia (ATTILA Study). Alzheimer's & dementia (New York, N Y). 2020. 2020;6(1):e12064–e12064. doi: 10.1002/trc2.12064 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Konig T, Pigliautile M, Aguila O, et al. User experience and acceptance of a device assisting persons with dementia in daily life: a multicenter field study. Aging Clinical and Experimental Research. Apr 2022;34(4):869–879. doi: 10.1007/s40520-021-02013-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Hunter SW, Meyer C, Divine A, et al. The experiences of people with Alzheimer's dementia and their caregivers in acquiring and using a mobility aid: a qualitative study. Disability and Rehabilitation. Nov 2021;43(23):3331–3338. doi: 10.1080/09638288.2020.1741700 [DOI] [PubMed] [Google Scholar]
- 15.Efthymiou A, Papastavrou E, Middleton N, Markatou A, Sakka P. How Caregivers of People With Dementia Search for Dementia-Specific Information on the Internet: Survey Study. Jmir Aging. Jan-Jun 2020;3(1)e15480. doi: 10.2196/15480 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Wojcik D, Szczechowiak K, Konopka P, et al. Informal Dementia Caregivers: Current Technology Use and Acceptance of Technology in Care. International Journal of Environmental Research and Public Health. Mar 2021;18(6)3167. doi: 10.3390/ijerph18063167 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Sriram V, Jenkinson C, Peters M. Carers' experiences of assistive technology use in dementia care: a cross sectional survey. Bmc Geriatrics. Aug 2021;21(1)471. doi: 10.1186/s12877-021-02417-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Sriram V, Jenkinson C, Peters M. Using rapid cycle tests of change to develop the Carers Assistive Technology Experience Questionnaire: a cognitive interview study in the UK. BMJ open. 2021;11(3):e042361. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Benge JF, Scullin MK. Implications for technological reserve development in advancing age, cognitive impairment, and dementia. Behavioral and Brain Sciences. 2020. 2020;43e157. doi: 10.1017/s0140525x20000126 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Wolff JL, Benge JF, Cassel CK, Monin JK, Reuben DB. Emerging topics in dementia care and services. J Am Geriatr Soc. 2021;69(7):1763–1773. [DOI] [PubMed] [Google Scholar]
- 21.Benge JF, Dinh KL, Logue E, Phenis R, Dasse MN, Scullin MK. The smartphone in the memory clinic: A study of patient and care partner’s utilisation habits. Neuropsychological rehabilitation. 2018; [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Boot WR, Charness N, Czaja SJ, et al. Computer proficiency questionnaire: assessing low and high computer proficient seniors. Gerontologist. Jun 2015;55(3):404–11. doi: 10.1093/geront/gnt117 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Roque NA, Boot WR. A New Tool for Assessing Mobile Device Proficiency in Older Adults: The Mobile Device Proficiency Questionnaire. J Appl Gerontol. Feb 2018;37(2):131–156. doi: 10.1177/0733464816642582 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Barg-Walkow L, Mitzner T, Rogers W. Technology Experience Profile (TEP): Assessment and Scoring Guide. 2014. [Google Scholar]
- 25.Farias ST, Mungas D, Reed BR, et al. The measurement of everyday cognition (ECog): scale development and psychometric properties. Neuropsychology. 2008;22(4):531. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Pfeffer RI, Kurosaki T, Harrah C Jr, Chance J, Filos S. Measurement of functional activities in older adults in the community. Journal of gerontology. 1982;37(3):323–329. [DOI] [PubMed] [Google Scholar]
- 27.Horton MC, Oyebode J, Clare L, et al. Measuring Quality of Life in Carers of People With Dementia: Development and Psychometric Evaluation of Scales measuring the Impact of DEmentia on CARers (SIDECAR). Gerontologist. Apr 3 2021;61(3):e1–e11. doi: 10.1093/geront/gnz136 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Galvin JE. The Quick Dementia Rating System (QDRS): a rapid dementia staging tool. Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring. 2015;1(2):249–259. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Farias ST, Gravano J, Weakley A, et al. The Everyday Compensation (EComp) Questionnaire: Construct Validity and Associations with Diagnosis and Longitudinal Change in Cognition and Everyday Function in Older Adults. Journal of the International Neuropsychological Society. Mar 2020;26(3):303–313. Pii s135561771900119x. doi: 10.1017/s135561771900119x [DOI] [PubMed] [Google Scholar]
- 30.Kline P. A handbook of test construction (psychology revivals): introduction to psychometric design. Routledge; 2015. [Google Scholar]
- 31.Yoon H, Jang Y, Vaughan PW, Garcia M. Older adults’ internet use for health information: digital divide by race/ethnicity and socioeconomic status. Journal of Applied Gerontology. 2020;39(1):105–110. [DOI] [PubMed] [Google Scholar]
- 32.Field-Fote E. Mediators and moderators, confounders and covariates: Exploring the variables that illuminate or obscure the “active ingredients” in neurorehabilitation. Journal of Neurologic Physical Therapy. 2019;43(2):83–84. [DOI] [PubMed] [Google Scholar]
- 33.Benge JF, Aguirre A, Scullin MK, et al. Digital Methods for Performing Daily Tasks Among Older Adults: An Initial Report of Frequency of Use and Perceived Utility. Experimental Aging Research. 2023:1–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.