Abstract
Objective
This study aimed to evaluate users’ acceptance of a teledentistry model utilizing a smartphone camera used for dental caries screening and to identify a number of areas for improvement of the system.
Methods
A store-and-forward telemedicine platform “Remote-I” was developed to assist in the screening of oral diseases using an image acquisition Android app operated by 17 teledental assistants. A total of 485 images (five images per case) were directly transmitted from the Android app to the server. A panel of five dental practitioners (graders) assessed the images and reported their diagnosis. A user acceptance survey was sent to the graders and smartphone users following completion of the screening program.
Results
Of the 22 surveys sent out, 20 (91%) were completed. Generally, users showed optimism towards the use of the teledentistry system, and strongly positively assessed items on content and service quality. The majority of graders took less than 15 min to read the images while phone users took 5–10 min to complete the dental photography using the Android app. This study identified a number of factors that are essential for improving the current system, such as optimization of smartphone camera features, the format of the server, and the orientation of images and using oral retractors during photography.
Conclusions
Users appear to be generally satisfied with the proposed teledentistry model. However, they have specific concerns to address, many of which could be resolved through more effective training, coordination between sites and upgrading the current system.
Introduction
Teledentistry, as a subspecialty of telemedicine, can be defined as “the provision of real-time and off-line dental care such as diagnosis, treatment planning, consulting and follow up via electronic transmission from different sites.” 1 (p. 399) For several decades, telemedicine has played a role in bridging gaps and overcoming barriers through spreading healthcare to previously unreachable populations. 2 Most teledentistry studies were at a small scale and limited to short-term outcomes, 3 utilizing now superseded technologies or expensive peer-to-peer system approaches. Evidence indicates that teledentistry examinations are comparable to clinical examinations in screening for caries.4–6 However, none has demonstrated superior clinical results (i.e. validity, reliability, and effectiveness) compared to traditional settings. It is postulated that teledentistry offers a cost-effective means to help reduce some obstacles to optimal oral health, particularly for underserved populations.7,8
There are many methods for screening for oral diseases. However, the most common method is the face-to-face examination. The rapid advances in digital imaging and other technologies have provided practitioners with alternatives to traditional settings. 9 Assessment of intra-oral photographs can maintain a good level of sensitivity and specificity of visual detection of caries. 10 Dental photography can also be less stressful and intimidating for young children than a conventional dental examination. 9 With smartphone camera technology improving significantly and widespread availability of the cellular networks, utilization of smartphone cameras in dental imaging has grown.11–13
Derived from the theory of planned behaviour (TPB), the technology acceptance model (TAM) aims to explain user acceptance and to predict the adoption of technologies. 14 TAM has been widely adopted in research, due to its parsimonious nature and wealth of recent empirical support of its role in predicting the acceptance behaviour of a technology.15,16 TAM posits that behavioral intention is determined jointly by attitude and perceived usefulness, the latter also affects attitude directly. Meanwhile, the perceived ease of use directly influences both attitude and perceived usefulness. 17 TAM works as an appropriate framework for our survey as it gives attitude a key role in predicting the potential user’s behavioural intention to use a technology; a role that has been shown to be fundamental in the acceptance of telemedicine.15,18
In response to the increasing demand for oral care services, particularly in remote or rural regions, we have developed a store-and-forward telemedicine platform called “Remote-I” and a novel image acquisition Android app that can be used for screening purposes. 19 The present study builds on an initial validation (proof-of-concept trial) study that tested the validity and reliability of the teledentistry approach in the screening for dental caries. 19 The findings show that the proposed teledentistry model for dental screening using a smartphone camera offers a valid and reliable alternative to visual dental examination. 19 Since perceived usefulness and ease of use are critical factors in the acceptance of a technology, we sought to evaluate users’ acceptance of the teledentistry model and to identify the factors that contribute to the improvement of the current system.
Methods
Teledentistry system
A telemedicine system, “Remote-I,” based on a store-and-forward method, was developed by the Australian E-Health Research Centre (AEHRC) to work as a platform for data storage and management.
20
Remote-I is capable of image acquisition, data entry, storage and retrieval of data. An image acquisition app was built and installed on a Motorola MotoG smartphone (USA) to facilitate entering patient details and capturing dental photos, and then uploading data corresponding to each patient to Remote-I, using Wi-Fi hotspots or cellular data networks. After obtaining their informed consent, participants were enrolled in a trial to obtain oral images using smartphone cameras. Each participant received an in-person oral examination by a dentist to record caries and existing restorations, before trained teledental assistants took photographs from each participant’s oral cavity, using a Motorola smartphone camera. Records were then directly transmitted as encrypted data from the smartphone to the Remote-I for evaluation by a dentist at a distance. The images captured using the smartphone were deleted from the phone immediately after transmission of the images to the Remote-I system to avoid any privacy issues (Figure 1). The components of the teledentistry system are summarized in Table 1.
Architecture of the teledentistry system. The components of the teledentistry system.
Participants (users)
Graders (dental practitioners)
Reviewing of dental images was carried out by five independent dental practitioners using a web-based data and image-viewing app built upon the Remote-I system. Although graders did not receive any training on how to use the system, a simple user manual and cover letter were sent to graders explaining the purpose of the study and how to use the system. They were able to access the database using individual user identities (IDs) and passwords. The system enabled graders to review images and insert comments on the predefined oral assessment form and submit reports or recommendations into the Remote-I server (Figure 2).
Snapshot of Remote-I server illustrating the dental chart.
Teledental assistants (smartphone users)
Patient recruitment and dental photography were performed by 17 trained teledental assistants (dental students, dental assistants, dental practitioners) using a smartphone camera. The teledental assistants received hands-on training on how to capture good quality images using a smartphone camera. Recruitment of patients and dental photography were completed in a series of locations; a small practice, hospital, and large dental facility. Over six months, up to 100 records were uploaded to the Remote-I server; these comprised 485 images (approximately five images per case) and anonymous patient demographic data.
Questionnaire instrument
The survey questions are based on a standard, validated, and reliable instrument for end-user satisfaction modified for telemedicine.21,22 Our study has good content validity, as the survey items were vetted and revised by three dental practitioners, and the questions reflected their concerns and areas of satisfaction. The survey comprised four sections. The first section included questions about the users’ demographic characteristics. The second section comprised of 11, five-point, Likert-type questionnaire (never/almost never = 1; seldom = 2; about half the time = 3; most of the time = 4; always/almost always = 5) used to assess the quality of the system which comprised five dimensions; content, format, information quality (accuracy), ease of use, service quality, and support. An additional two items were included to assess the usefulness and overall satisfaction with applications. The third section examined the average time that was spent to create a record and complete photography using the Android app and grade a record on the Remote-I server. The final section solicited free comments about whether users have suggestions to improve existing systems. Following completion of the trial, the survey was distributed by e-mail with a cover letter stating the purpose of the survey. A reminder was sent to all users who did not respond to the initial correspondence. The protocol for this study was approved by the Human Research Ethics Committee of The University of Western Australia.
Results
A total of 22 requests for completion of the survey were sent to all users of either the smartphone (teledentistry assistants) or Remote-I (graders) users. A total of 20 completed surveys were received (91% response rate); only two smartphone users did not respond to the survey. The mean age of respondents was 34 years and the majority of respondents (80%) were women. The respondents’ professions were dentists (n = 11; 55%), dental therapists (n = 3; 15%), dental students (n = 3; 15%) and nurses (n = 3; 15%). All respondents confirmed that they used computers both at home and work, and reported good typing proficiency.
User acceptance of Remote-I system and smartphone app
The overall level of satisfaction with both the Remote-I server and Android app was positive. Generally, smartphone users had strong positive assessments on items assessing content, format and service quality (system’s stability). The answers to “How often is the content presented in the app sufficient and appropriate?,” “How often does the content meet your needs?,” “How often do you think the output is presented in a useful format?,” and “How often is the information clear?” were given as “most of the time” or “always” by more than 87% of app users. The answer to “How often is the system subject to technical problems or crashes?” was answered with “seldom” or “never” by 80% the users.
End-user acceptance survey of the Android smartphone app.
Never/almost never = 1; seldom = 2; about half the time = 3; most of the time = 4; always/almost always = 5. The percentage of respondents reporting “most of the time”/”great” and “always/almost always”/”very great” (Likert scales “4” and “5”) within a given scale were charted as percent positive, while the percentage reporting “never”/”not at all” and “seldom”/”very little” (Likert scales “1” and “2”).
End-user acceptance survey of the Remote-I server.
Never/almost never = 1; seldom = 2; about half the time = 3; most of the time = 4; always/almost always = 5. The percentage of respondents reporting “most of the time”/”great” and “always/almost always”/”very great” (Likert scales “4” and “5”) within a given scale were charted as percent positive, while the percentage reporting “never”/”not at all” and “seldom”/”very little” (Likert scales “1” and “2”).
Time taken for dental photography and grading
Duration of dental photography and grading process.
Users’ suggestions of ways to improve the system
Users were enthusiastic about the use of the teledentistry system for screening for caries based on photographic assessments. The majority of smartphone users (67%) suggested a number of areas for improvement; in particular in the zoom and autofocus features of the smartphone camera. Similar opinions were expressed by other users (27%) who felt that the quality of some images was not good due to lighting issues related to the inbuilt camera flash (Figure 3). More than half of smartphone users (53%) suggested using disposable retractors during photography to help in obtaining clear images. Few users (20%) suggested developing an app that can support other operating systems such as iphone operating system (iOS) and Windows, and also suggested making the app available in the Google or Apple stores.
Example of a smartphone camera shots. (a) An intraoral occlusal view of upper teeth with poor image quality (blurred) due to being out of focus. It is not possible to observe teeth or soft tissue in this image. (b) Shows intraoral occlusal view of upper teeth with excellent image quality. It is possible to see decayed first molars and decayed left second molar.
Categories and frequencies of suggestion by respondents.
Discussion
Overall, our survey showed that users perceived the store-and-forward based teledentistry system positively as useful for screening purposes. This reflects previous research which has shown excellent levels of acceptance of real-time based teledentistry system by both patients and professionals.23,24 Although, most teledentistry systems utilize a real-time modality, 3 the practices of asynchronous or store-and-forward telemedicine have proven to be more cost-effective and efficient compared to real-time and in-person models in some clinical disciplines.8,25,26 Historically, mobile devices suffered from a low storage space and low quality of imaging, and this could be a reason for underusing smartphones in dental photography. With recent significant improvements in smartphone camera technology, utilization of smartphones in dental photography and for screening purposes has grown.11–13 Even when a telemedicine system proves to be useful, users also note room for improvement or offer suggestions. Collecting feedback on areas of satisfaction, barriers, and suggestions from users about a telemedicine program can help foster better-designed programs that can be more successfully implemented. 27
Despite the majority of users believing that the Remote-I system is accurate, stable, easy to use, and presented in a useful format, our findings showed that perceived accuracy, format, and ease of use could be improved. A few concerns persist among respondents, including difficulty in reviewing posterior teeth because the oral cavity was not retracted enough, system architecture such as dental chart layout and design, and orientation of images and labelling. Another matter affecting the graders’ ability to review images and grade teeth is the quality of the images that the teledental assistants send. The quality of images and the capability to grade images accurately are very important factors when evaluating the feasibility of telediagnosis of diseases. 28 Many of the problems encountered could be overcome following further modification and optimization of the server.
Despite attention to many issues (over bright or dark images) related to smartphone camera use, and adjustment of photography protocols prior to starting the trial, concerns with the information quality and ease of use of smartphone camera were observed (Figure 3). Although, the majority of the smartphone users believed that the Android app was accurate and easy to use, the survey results showed that a substantial number of Android users (20%) experienced difficulty with the use of the smartphone camera and a majority of the smartphone users suggested that the quality of images and the smartphone camera need further improvement. Users’ concerns with the smartphone camera mainly pertained to the difficulty of obtaining good images due to issues relating to the optimization of phone camera features, the absence of oral retractors, or lack of training.
The survey shows that the average time spent on creating a record, uploading and grading was 20 min. Although this time is longer than an in-person oral assessment (10–15 min), a duration of 20 min is still considered acceptable as the only alternative is to spend hours traveling to the nearest practice or sending a practitioner to a remote site. Although training of teledental assistants to provide good quality oral images was provided and does not consume much time, adherence to the photography protocol provided was hard to achieve. Many Android users’ concerns could be resolved through more effective training and hands-on experience.
The results of this study partly support the TAM model of telemedicine acceptance. Similar to the assertion of TAM, perceived ease of use and usefulness was relatively high among users and was positively associated with users’ attitude toward using the system, and these results are consistent with previous reports.14,29 However, our data indicates that the perception of the system’s stability, content, and format have a strong influence on the attitude towards teledentistry system. The intention to use the system is not only predicted by its perceived usefulness but also determined by the system’s stability, format, and content which were rated consistently high in the survey. Perceptions of information quality and support were not found to influence the intention to use the system significantly and the latter could be attributed to the absence of coordination between sites and lack of prior personal experience with teledentistry. Another important factor that may affect a user’s intention to use telemedicine is the need for collaboration and coordination between remote and hub sites when implementing a telemedicine system. 30 Because the supervisory dental team at the hub site cannot do a hands-on examination, they have to rely on the assessment performed by the local practitioner at the remote site. Repeated practice, establishing confidence, and good working relationships between team members at both sites can establish a reliable and smooth teledentistry process. 30
This study was not designed to examine in depth the users’ perception of teledentistry, but it gives insights on its acceptance. Further investigations on how different types of factors (external or internal) operate to support or hinder the adoption of a teledentistry model need to be addressed. Despite the rapid growth of teledentistry, it has not yet become an integral part of mainstream oral health care. Future research is needed to determine the obstacles that delay the implementation of teledentistry as an adjunct for a comprehensive health care system. Identifying and addressing the barriers to adoption of teledentistry could motivate dental providers to adopt the use of telemedicine services in daily practice.
Conclusion
Generally, users considered the teledentistry model as useful, despite their concerns with specific aspects of the system. This study provides developers with key observations about user needs when building a teledentistry system and can form the basis for the development of a more user-friendly system. Utilization of teledentistry approaches for caries screening has steadily gained acceptance in recent years. More recently mobile teledentistry, often seen as a sub-set of telemedicine, has also emerged as a possible means of screening for oral diseases. This area is particularly attractive due to the fact that many smartphones are now equipped with digital cameras, and there is widespread penetration of smartphones and cellular network reception globally even in underserved regions. We believe that these technologies (teledentistry and smartphones) can be combined to create an inexpensive and powerful tool for screening purposes. This approach could offer a practical and potential cost-saving means to screen for oral diseases among a population with high levels of need who have no access to care.
The development of a cloud-based server and an image-acquisition Android app for screening purposes is the first of its kind in Australia and has the potential to serve rural providers to reduce inappropriate referrals and waiting lists for consultation as well as facilitating timely information to the local practitioner for better decision-making. Evaluation of users’ acceptance provides valuable insight into the factors which impact professionals’ intentions to adopt teledentistry services. Building on these experiences, the next step will be the implementation of multi-site, community-based, and large-scale projects that can incorporate remote dental screening and oral health promotion, and involve different members of the dental team such as dental therapists, dental hygienists, and dental nurses.
Footnotes
Acknowledgement
We thank Professor Boyen Huang from the Charles Sturt University for his contribution. We also thank all dental students contributed in this study.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. The authors report no potential conflicts of interest.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
