Older people’s use of technology is increasingly coming in focus with the demographic changes. Gerontechnology (technology for the aging population) is a growing field in transdiciplinary research as well as in the development of new products [1,2]. Previous research into older people’s technology use has identified that the design [3,4] and technology adoption [5,6] perspectives are important to ensure appropriate functionality and remove various barriers for use. Personal factors, such as self-efficacy and proficiency [5,7,8], as well as subjective technology adaptivity [9,10] have also been identified as significant predictors of technology use in old age.

One area that has attracted interest is the use of technology by older adults in various health settings, both in formal health care [11,12] and from salutogenic perspectives [13,14] targeting social isolation and participation. However, there is a strong need to find evidence of the effectiveness and efficiency of health technology interventions, which is becoming increasingly important as the number of available health technology solutions grows [15].

Several instruments for evaluating interactions with health technology exist today. Common theoretical concepts addressed in those instruments are effectiveness, efficiency, hardware/software, perceived ease of use, and satisfaction, although validated psychometric instruments measuring personality traits are sparse [16].

One of the most widely used instruments is the technology acceptance model (TAM) [17] and its subsequent developments [18] for evaluating attitudes predicting intentions to use and how users come to accept and use a particular information technology. In this model, two specific factors determine the user’s acceptance: perceived usefulness and perceived ease of use. Although TAM was not developed specifically for health information technology (HIT), it has found its way into this area as a measurement of end users’ reactions to HIT [19]. Expansions to adapt TAM to a more specific HIT context have been made: the health information technology acceptance model (HITAM) [20]. Application and problematization of TAM toward older persons have also been made: the senior technology acceptance model [21] and HITAM of older persons [22]. The importance of contextual factors [23,24], as well as the usability and acceptability for older adults with mild cognitive impairment and dementia [25] have recently been in focus.

Another widely used model is the Delone and McLean information systems (IS) success model, which considers constructs of intention to use, user satisfaction, and net benefits as the outcomes of three sets of indicators that are information quality, service quality, and system quality [26]. The Delone and McLean IS success model and its extended variants, such as human, organization, and technology-fit [27], have been widely used in health technology [28]. How traits affected by or closely related to technophobia, such as technostress [29], can impact the satisfaction construct in the Delone and McLean IS model has been discussed [30].

Usability is another essential aspect of HIT evaluation. Both the design and evaluation of artifact, efficiency, effectiveness, and satisfaction are important. The Health Information Technology Usability Evaluation Scale [31] and the System Usability Scale [32] are instruments for usability evaluation that have been used in a HIT context [33,34]. In what is an extension of the TAM perspective, Kamin and Lang [10] explored the motivational resources for older persons’ technology use by the concept of subjective personal adaptivity. They found that positive beliefs about the benefits of technology, the time and effort invested to learn how to use technology, and a sense of trustworthiness and safety while using technology is connected to both perceived technology competence and technology usage.

Usability, acceptance, motivation, and adoption of a specific technology or service are important factors in evaluating the impact of a HIT intervention. However, basic attitudes and sentiments toward technology (eg, technophilia) could be argued to influence both the level of satisfaction with the technology itself as well as the perception of the health intervention outcome. This would constitute a personality trait, an underlying factor that would create a preintervention entry level of acceptance and interest, positive or negative. Edison and Geissler [35] argue that this “affinity” toward technology is a more general attitude that precedes the more specific attitudes resulting from the rational (reasoned or planned) process that is measured in TAM. Plociennik et al [36] found that technophilia has a direct influence on perceived usefulness in TAM. Ronit [37] sees technophilia as the enthusiasm toward technology with its rewarded and knowledgeable adoption correlated with both perceived usefulness and perceived ease of use. Kamin and Lang [9,10] showed a correlation between usability and utility and a subjectively perceived interest and competence.

Technophilia has no universally established definition but generally refers to a strong enthusiasm and love for modern technology. Seebauer et al [38] define it as “an attitude toward ICT [internet and communication technology], representing a subaspect of technology-related values, just as ICT are a subcategory of modern technology.” Osiceanu [39] defines technophilia as an “attraction, enthusiasm of the human individual determined by the activities which involve the use of advanced technologies. It is expressed by easy adaptation to the social changes brought by technological innovations.” Martínez-Córcoles et al [40] suggest that merely enthusiasm and desire is not enough for technophilia, but also an acquired need for (dependency) and joy of having and displaying the latest products/versions (technoreputation). In a working paper, Li and Fuller [41] suggest a definition of technophilia as “positive affective states that arise momentarily in response to an individual’s ICT context that is appraised to exceed his or her expectations and goals.”

In their review of previous related research on technophilia and similar concepts, Seebauer et al [38] found three hierarchically nested perspectives on technophilia. At the top level, they found values connected to general technology-related values reflecting global beliefs in societal progress through technology. At the intermediate level were attitudes referring more specifically to ICT as a part of modern technologies. The lowest level was constituted by a keen interest in and use of a specific technology or service as a subcategory of ICT. Openness toward technology and innovation influences personal dedication to certain technological artifacts and services, whereas feelings of low enthusiasm may work in the opposite direction [38]. Donat et al [42] refer to this lower (negative) end of technology enthusiasm as technophobia, and their construct is made up from two opposite ends where technophilia is at the positive end. Nimrod [43] includes this positive end of the spectrum in her construct of a technophobia scale while investigating ICT use among older adults. Osiceanu [39] further views technophobia as the negative feelings about using technology, but also includes perceptions on the adverse effects that technology may have on society in the construct. This is a common way to conceptualize technophobia construct with both factors concerning personal feelings toward technology use together with overall perceptions of technology in society.

In this study, we choose to avoid the complexity of bringing the two concepts of technophilia and technophobia together and simply let technophilia refer to a person’s enthusiasm for and positive feelings toward their technology use and absence of the fears and doubts some older people could have about their ability to manage using new technology. It constitutes a personality trait, an underlying psychological construct [38] that would create a pretechnology and preintervention entry level of acceptance and interest. This could also be connected to other psychological characteristics (ie, the personality traits of openness and neuroticism) [44].

Steinerman et al [45] report that technophilia is a consistent predictor of openness to research participation for older adults and that research with older adults that incorporates technology should consider technophilia to be more successful in recruiting participants for the study.

Older people compared to younger persons are sometimes reported to score lower on technophilia scales [38]. Still, older people are not a homogeneous group, and there may be differences between “younger” old and “older” old individuals. Therefore, it is of importance to create an instrument with the ability to discriminate between high and low technophilia individuals within various groups of older people.

The purpose of this study is to develop a reduced and refined instrument for measuring older people’s attitudes and enthusiasm for technology based on relevant existing instruments for measuring technophilia (named TechPH, short for technophilia).

Technophilia is a general quality for any individual’s relationship to technology that could potentially influence a wide range of aspects of technology use, such as adoption, continuity, and perceived outcome.

In this text, we contextualize this general quality with a focus on older people for use in health technology intervention research as a complement to existing instruments. The new instrument requires that it should be short and simple to make it usable for older people.

Data were obtained from a sample of participants in the Swedish National Study of Aging and Care (SNAC). SNAC is a longitudinal cohort study of a representative sample of the aging Swedish population that began data collection in 2001. It is a comprehensive, interdisciplinary study that investigates the health and living conditions of the Swedish population aged 60 years and older. A detailed outline of the SNAC study is available by Lagergren et al [46]. Our study sample was based on participants from one of the four regions in the SNAC study, the SNAC Blekinge (SNAC-B) cohort with individuals living in the municipality of Karlskrona.

Data were collected through a questionnaire that was sent out in October 2017 to all participants in the SNAC-B study who were alive in January 2017 (N=878). Of these, 18 had deceased before answering the questionnaire. A total of 659 persons responded, corresponding to a response rate of 77% (659/860). Among nonresponders, 28% (57/201) were unable to respond due to their health conditions (eg, severe dementia or other diseases) and were considered nonusers of ICT. In this study, only individuals who responded that they were ICT users were included (N=374). Demographic data for the study population (ie, age, gender, and educational level) are presented in Table 1.

Development of the new short instrument, TechPH, was made in three steps. First, a search was made in Web of Science and Google Scholar for technophilia measurement instruments to use as a background for building the new instrument. To be included, the articles had to contain a psychometric instrument for use on the individual level. Eight relevant instruments were found to match the criteria (Multimedia Appendix 1), of which seven reported a full instrument and were included in this study. Instrument 8 [40] did not include the full instrument in their article and could not be used in the content analysis. Overarching themes (emotional, personal gain, openness/curiosity, competence, general attitudes) were identified through content analysis and items corresponding to these themes were constructed with a five-point Likert scale questionnaire, ranging from 1 (fully disagree) to 5 (fully agree). Following the instruments in the analysis, both questions with a positive and a negative direction were constructed to make sure both the lower and the higher end of the spectrum of technophilia were sufficiently covered. The instruments used in the analysis also gave reason to assume that more than one factor could be present. A total of six to eight questions was considered ideal both for the length of the instrument and for the case that more than one factor would emerge from the factor analysis.

The new instrument resulting from the content analysis was able to build on the previous instruments by including all themes in the same instrument. The analysis showed that none of the seven existing instruments investigated covered all the themes resulting from the analysis, making the new developed short instrument (TechPH) more comprehensive than its predecessors.

Expert knowledge in gerontology from both the medical side (gerontologist) and the technical side (design expert) was used to phrase and formulate the questions for the target group.

Secondly, the questionnaire was pretested and cognitive interviews [47] were made with the target group (eight individuals of both sexes varying in age between 60 and 82 years). The interview persons were given the 8-item questionnaire and were encouraged to think-aloud when they read them. The interviewer would also follow-up with verbal probing (ie, questions about how the interview persons understood the question) based on item wording, terminology, and if the structure was clear and easy to understand. Specifically, the questions “Can you repeat the question I just asked in your own words? Was there anything confusing about this question? What does the word <term> mean to you as it is used in the question? Tell me what you were thinking when I asked about <topic of question>.”

The item questions were then revised according to the feedback from the interviews with respect to the verbal probing. Especially important was to make sure that the questionnaire was using terminology relevant to older people using technology to ensure face validity. The resulting questionnaire items (Table 2) was also pretested on four persons from the target group.

Finally, factor analyses were made. First exploratory factor analysis to see the factor structure and decide on a one or multiple factor solution. Before the exploratory factor analysis, Bartlett test of sphericity was used to ensure significant correlation and Kaiser-Meyer-Olkin test of sampling adequacy for sufficient variance among the items. Maximum likelihood factoring and Promax with Kaiser normalization were used, and only factors with eigenvalues greater than 1 were included.

The reliability of the questionnaire was calculated with the Cronbach alpha coefficient to ensure sufficient internal consistency.

Confirmatory factor analysis was then used to verify the factor structure. Measures of fit are reported, such as chi-square statistic and its significance, the adjusted goodness-of-fit Index (AGFI), the comparative fit index (CFI), the root mean square error of approximation (RMSEA), and the standardized root mean square residual (SRMR).

Finally, an index based on sum scores was calculated [48], and sociodemographic attributes and self-assessed technical competence were assessed for group comparisons and as indicators of criterion validity.

The Kaiser-Meyer-Olkin test of sampling adequacy was in the adequate range of 0.76 [49] and the Bartlett test of sphericity (χ²₂₈=554.1 was significant (P<.001), indicating that the items were appropriate for a factor analysis [50].

A two-factor solution for technophilia emerged with an eigenvalue greater than 1 and examination of the scree plot. A one-factor solution was also tested for but showed low internal consistency and was discarded. The two factors were distinctly different with respective clear loadings. The factor with questions regarding positive feelings toward technology (ie, items 1-3) contained items with various aspects of enthusiasm toward technology was accordingly named techEnthusiasm. The factor with more negative feelings (items 4-6), contained items with different aspects of anxiety toward technology with respect to use and competency and was named techAnxiety.

The questions regarding general attitudes (items 7 and 8) gave low loadings and cross-loaded on both factors above the recommended maximum threshold of 0.32 [51]; therefore, they were removed.

The final six-item solution (Table 3) gave satisfactory loadings of above 0.5 [50], and a total variance explained of 63.5% for the two factors together.

Convergent validity with average variance extracted above 0.5 [50] and discriminant validity shown with only small cross-loadings, together with a factor correlation of −.48. The reliability of the questionnaire, in terms of internal consistency, was calculated by Cronbach alpha (Table 3) with satisfactory results for the small item number (techEnthusiasm Cronbach alpha=.72 and techAnxiety Cronbach alpha=.68).

The final confirmatory factor analysis conducted showed relatively good [52] fit indexes for the two-factor model (χ²₈=21.2, χ²/df=2.65, CFI=0.97, AGFI=0.95, RMSEA=0.067, SRMR=0.036). The model showed satisfactory (>0.5) standardized factor loadings given the sample size [50] confirming construct validity. Table 4 shows the standardized parameter estimates.

A composite score [50] (see Table 5) was created from the six items in the two factors, techEnthusiasm and techAnxiety (the latter reversely coded due to the negative correlation). Each item was weighted with its loading before sum scores were created and averaged [48] and standardized back to a 1 to 5 scale so that the TechPH index could be interpreted on a five-point response scale, ranging from 1 (fully disagree) to 5 (fully agree), where the higher the index indicates a higher level of technophilia.

Distribution of all individuals’ scores in the two factors, techEnthusiasm and techAnxiety, is presented in a scatterplot (see Figure 1). A low negative correlation between the two factors was observed (R²=.12).

The TechPH index (Table 5) was used for group comparison for gender, age group, level of education, self-assessed technical skills, and internet use frequency. This comparison yielded significant (P<.05) results. Cronbach alpha was .71, which signifies moderately good reliability of the index and internal consistency.

The TechPH index scores reflect the same finding or assumptions in technology acceptance by demographic variables; the index decreases in the older old (≥75 years) group and is slightly higher for men, which confirms other findings with regard to age groups and gender [53].