An ever-increasing number of patients access health information via the internet [1-3]. However, most of the health-related information available on the web is of low quality, while many patients are unable to adequately appraise the quality of such health information [4,5]. Tala-med was therefore envisaged and developed as an evidence-based, up-to-date, and easily understood web-based information resource for lower back pain (LBP) through the Gut informierte Kommunikation zwischen Arzt und Patient (GAP) project (well-informed communication between the general practitioner and patient) [6]. LBP in particular is a widespread health problem that causes substantial personal and financial burden [7,8]. In Germany, LBP has a 1-year prevalence of more than 60% [9,10], accounts for the most days of sick leave (6.1%) [11], and is one of the most common reasons why people visit their doctor [12].

Despite the existence of national and international clinical guidelines, approaches to treating LBP differ greatly among clinicians, institutions, and geographic regions [13,14]. However, the breadth of available web-based information often surpasses the variety of management approaches. Consequently, patients can often be confused and frustrated while searching for web-based information regarding LBP [15]. An overabundance of differing and contradictory information can make it difficult for patients to understand and assess the quality and trustworthiness of the information provided [15], often presenting a dilemma for patients searching web-based information for LBP.

Digital health interventions (DHIs) recommended by health care professionals (HCPs) may be a remedy for this dilemma, as they can provide patients with tailor-made, understandable, and high-quality information. Our web portal tala-med is one such DHI that could be recommended to patients by primary care physicians (PCPs). As an information resource based on national [16,17] and international clinical guidelines for LBP [18-20], tala-med can be classified as a DHI that can provide health content [21,22] to physicians and patients alike. Tala-med aims to improve the shared decision-making of PCPs, while enhancing patient-informed choices, participation, and self-management regarding LBP [6]. For interventions on shared decision-making to be most effective, Cochrane reviews have shown that these should be both physician and patient focused and include information that is indication specific [23,24]. Regarding DHIs on back pain, recent meta-analyses and systematic reviews have shown that such interventions, especially those that focus on self-management, can have clinically important effects in terms of relieving patient discomfort and improving their disability [25-27]. However, a key determinant of the effectiveness of DHIs on LBP [28] and DHIs in general [29] is adherence; that is, whether patients actually use the DHI to the intended extent [30].

Unfortunately, the extent to which patients use DHIs is often low [31-33]. In addition, promoting the use of DHIs by patients is complex, with only limited evidence available on successful strategies to do so [29]. Therefore, it is important to understand the web-based health information needs of patients [34] and other factors that may facilitate or hinder their use of a DHI. Regarding web-based information on LBP received by patients as an adjunct to their PCP consultation, Riis et al [15] found that readability, customization, design, credibility, and usability are important domains. However, these results were based on patient experiences using a variety of health-related websites and not on the use of a specific web portal provided by their PCP. Studies on patient perceptions of DHIs for LBP have also revealed that contextual factors, such as the support of HCPs, and individual factors, such as patient skill and preference, affect the acceptance of such DHIs [28,35]. A current systematic review of qualitative studies found only 4 studies that investigated the facilitators of and barriers to the use of DHIs by patients with LBP. Svendsen et al [28] thus state that “further primary research investigating the implementation of DHIs and user’s experiences is required.”

The aim of this study was to examine patient acceptance of our web portal tala-med as well as its usability and utility. We were eager to see how PCPs used the portal during the consultation, how they offered it to patients, and whether it was subsequently used. We also wanted to understand which parts of the portal were considered helpful and how patients perceived tala-med in terms of key characteristics such as comprehensibility and trustworthiness, enabling us to identify the portal’s strengths and weaknesses. This in turn provided us with insights into how specific features of the portal or its setting may have contributed to patient perception. Finally, we examined the utility of tala-med, in particular, the perception of any added value brought about by its use.

Tala-med is a German-language, evidence-based, comprehensible, and reliable internet information portal for LBP that is freely accessible and aimed at improving patient informedness and patient-doctor interaction. It was implemented within a prospective cluster-randomized controlled trial (RCT) with pre–, post– and 1 follow-up measurement [6]. Tala-med was designed to be used by PCPs and patients in the intervention group during and after LBP consultations. There are 2 versions that have been adapted to meet the demands and linguistic levels of PCPs and patients (Figure 1). To log in to the portal, both PCPs and patients were given an individual fictitious username consisting of a combination of 2 short animal, color, and fruit words (eg, PearOwl) and a password. This type of log-in was necessary because portal use data were also analyzed anonymously as part of the RCT [6].

The portal was developed over a 6-month period by a multidisciplinary team consisting of 1 PCP, 2 physiotherapists, 2 researchers specializing in evidence synthesis, 2 web designers, a film crew concerned with exercise video production, a specialist for whiteboard videos, and a 3D specialist. The focus was on frequently occurring issues seen in LBP, and it contained descriptive material in short and long versions, such as an accordion style guideline, where users could click on headings and highlighted keywords to obtain further background information. The development of the portal, including information processing and design principles on which the presentation of the information was based, is described in the associated methodological guide, which is also publicly available [36]. The final version consisted of 6 infographics, with 20 illustrations, a 3D model animation of the lower back, an 11-minute explanatory whiteboard video containing 8 different topics, and 14 short exercise videos 2 to 4 minutes long, with an emphasis on teaching self-care. In addition, suggestions, infographics, and background information on well-informed shared decision-making and preventive lifestyle changes were also included. The PCPs were able to show or print material during their consultations. Study PCPs were instructed to show and explain the web portal to their patients and encourage its use. If this was not possible, for instance, if the consulting room had no computer access, PCPs were advised to promote its use, despite being unable to demonstrate the portal in real time. As a backup option, PCPs were given up to 4 hard-copy brochures, which could be handed out to patients who were unfamiliar with computer use. The restriction to 4 brochures was chosen to encourage PCPs to primarily use the web portal and only refer to a brochure when handing the portal to a patient seemed inappropriate. The brochures contained information from the web portal in printed form. Patients were able to use the patient-tailored version of the web portal at home.

We used a qualitative design in which patients with LBP, who had received access to tala-med via their PCP, were invited by post to individual telephone interviews. For administrative reasons and owing to the delay of patient postal responses, interviews were held 1 to 2 months after the consultation, in which patients received their log-in details to the web portal. Results were reported according to the consolidated criteria for reporting qualitative research (consolidated criteria for reporting qualitative research [37]).

All 190 patients with LBP from the intervention group were invited to participate in an interview after the last follow-up measurement of the RCT. Although patients received a book voucher for their participation in the RCT, no incentives were offered for their participation in the subsequent interview study. Then, 35 patients (18%) accepted the invitation and returned their informed consent and contact information. Three patients were not interviewed: 2 could not be reached and 1 was no longer interested in an interview, as she had not yet used the web portal. A total of 32 patients were interviewed, and 17% of all invitees were interviewed. The patient characteristics are shown in Table 1.

The study was approved by the ethics committee of the Albert-Ludwigs-University Freiburg (no 559-17).

All participants provided written informed consent before being interviewed. Apart from contact data, the interviewers had no prior information about the participants. At the beginning of the telephone interviews, patients were informed about data protection issues and the recording of the interviews (Multimedia Appendix 1).

Interviews were held between February 2019 and October 2020 by CS (31 interviews) and M Klimmek (1 interview). These were based on a semistructured interview guide developed by ACSL, CS, and NR, which contained questions about patient experiences of the consultation and web portal (Multimedia Appendix 1). The interview guide contained skip rules to ensure that interviewees were only asked questions that they could provide answers for. For example, patients who did not use the portal were not asked questions regarding the usability of the portal and its utility. The interviews lasted an average of 30 minutes, ranging from 15 to 45 minutes in length. Interviewers did not take field notes during the interviews. The interviews were digitally recorded by connecting a digital recorder to the interviewer’s landline telephone. Recordings were transcribed verbatim by a transcription service and analyzed using MAXQDA (version 20; VERBI Software GmbH). Participants did not receive the transcripts or feedback results.

The transcribed interviews were analyzed using framework analysis [38-40]. Textbox 1 outlines the 5 stages of the analysis as well as their implementation.

As described in Textbox 1 and in the previous section, CS, ACSL, M Klimmek, and NR were the researchers primarily involved in data collection and analysis. CS and ACSL were postdoctoral researchers in the field of health services research and rehabilitation research, who hold degrees in psychology. M Klimmek was a bachelor’s student in social work who worked as a student assistant in the same field. NR is a researcher in the field of medical psychology and medical sociology who holds a degree in health education. ACSL is experienced with qualitative studies. She introduced CS and NR to the framework analysis.

Two themes were related to patient acceptance of the portal: PCP use of the portal during consultation (theme 1) and patient use of the portal (theme 2).

Responses regarding the usability of the portal (theme 3) relied on the answers of 26 patients. However, 10 of these patients mentioned that they had difficulty remembering the portal in detail. Textbox 3 shows how patients rated the portal and their respective reasons. In general, most patients rated the portal as positive. They mentioned that the portal was interesting and informative, provided good information, was easy to use and easy to implement, and contained useful exercises and videos. Neutral and negative overall ratings were obtained from patients who were disappointed that the portal did not recommend their preferred treatment (injection) or lacked information about a particular type of back pain (upper back pain).

Reasons for positive design ratings were that the portal was perceived as visually well-structured and uncluttered. Patients also liked the brevity and simplicity of the exercise videos:

Patients with neutral or negative perceptions of the portal design reported that they preferred dealing with a real person, needed more interactive elements and animations, and found that the portal was poorly designed for smartphone use. The portal was optimized for notepad-sized screens or larger screens.

Most patients perceived the portal as clear and did not have any problems interacting with it. They found it simple and well structured, well described, and thus easy to navigate. For some patients, the portal structure was unclear; 1 patient stated that he could not find everything he needed from the outset, reporting that “it would be easier for a younger person who sits in front of the computer all the time” (patient 21, male, age category: 50-59 years). The other patients had problems dealing with the portal, needing help, or finding it too convoluted and deeply structured, requiring too many clicks to find what they were searching for.

Almost all patients perceived the portal information as easy to understand, even for back pain novices, because the information was rated as simple and well described, containing only a few specific terms with no extensive texts. The portal information was unanimously perceived as trustworthy. Patients mentioned 4 main reasons for this (Textbox 4). Many patients perceived the portal as trustworthy, as it was developed or recommended by a source with a high level of expertise:

Furthermore, underpinnings of trustworthiness were the well-founded and scientific information of the portal and its serious presentation, putting an emphasis on the clear presentation of information, while avoiding distractions or advertisements. Patients also perceived the portal as trustworthy, as it contained noncontradictory information with aspects the patients knew to be true. The use of a fictitious username with great attention to data protection also supported the impression of trustworthiness for some patients.

A total of 14 patients had suggestions for improvement. Regarding additional content, patients suggested adding the addresses of recommended specialist centers, doctors, or therapists in their region, with freely available consultation slots when a second opinion was needed at short notice, as well as an advice hotline on the contents of the portal. Including more exercises for the upper back, alternative therapies such as acupuncture, and offering support for other conditions in addition to LBP were also suggested. Design improvements were also envisaged through the use of greater customization, with separate access for patients with little experience of LBP, as well as with more experience of LBP. Regarding the exercise videos, shorter sequences for use on the go, as opposed to the current 2- to 4-minute videos on offer, would have been appreciated, as well as the ability to loop videos, adjusting the number of repetitions available for exercising in tandem with the videos. One patient would have appreciated instructions to be given throughout the program, while another patient felt the need for a diagnosis-orientated search facility, with diagnosis-specific preselection of information and exercises. Regarding accessibility, many patients felt that the portal should be made freely available to everyone, which was unfortunately not the case at the time of the study. One patient suggested that the portal be made widely available in waiting rooms and pharmacies. Some hoped that it would be made available as a smartphone app, while others were eager for a hardcopy printout to be made available to those without a computer or access to the internet. Almost all patients affirmed that they would recommend the portal to others and some had already done so.

The utility of the portal was assessed by its added value (theme 4) and its effects (theme 5).

Patients accepted the portal well, appreciating its use both in the consultation and at home. PCPs mainly used the portal as an additional information resource for their patients and recommended the exercises as described. Although patients appreciated this use, it only partially exploited the information potential of the portal. The parts of tala-med, which aimed to improve the consultation and in particular shared decision-making, seem to have had little impact. With respect to these aspects, the findings suggest that tala-med or its implementation could be improved. Many patients perceived the exercises, presented as guided videos, as the most helpful part of the portal, this being the sole area of interest for some. This finding supports the suggestion of Wollmann et al [34] that videos and tutorials about health information would be well received by patients.

Patient use of the portal was facilitated through PCP behavior during the consultation, such as introducing the portal and recommending it to patients. This finding is consistent with previous studies showing that HCP recommendations and support for the DHI were facilitators of use [28]. The current evidence underlines the important role of HCPs in promoting patient use of DHIs. Our findings also showed that portal use was facilitated through further questions from patients and through new onset and moderate LBP, which also served as a reminder to patients to take action. Other levels of LBP, in particular no LBP or excessive LBP, hindered patient use of the portal. According to a current systematic review, DHIs for the self-management of LBP should be tailored to pain severity [28]; otherwise, patients would not use them [15,28,42]. Our finding that absent or severe LBP was a barrier to portal use may reflect the degree to which tala-med’s contents align with pain severity, suiting patients with mild to moderate pain but less so those with absent or severe pain. This result is also consistent with a recent study by Geraghty et al [35] on the use of a DHI in primary care, which aimed to support patients in self-managing their LBP, who found mild and severe pain as barriers to DHI use. Thus, in primary care, DHIs for the self-management of LBP seem to be used primarily by patients with sufficient but not severe pain. In line with previous evidence [28], our findings showed that the use of the portal was facilitated by the high self-motivation of patients.

Barriers to the use of the portal included other more intensive back pain therapies and other acute conditions. Similar to our findings, Geraghty et al [35] described concurrent health conditions and comorbidities as barriers, although these specific comorbidities differed. Mental health conditions, especially depression, which is often associated with LBP [43], were not mentioned in previous studies as a barrier. However, the hindering effect of depression due to reduced drive and low energy has been observed with respect to the use of face-to-face pain self-management programs [44,45]. If depression hinders patient engagement in face-to-face programs, this hindering effect should be even stronger with DHIs such as tala-med, which do not involve direct contact with an HCP and thus provide less-direct guidance and encouragement for use. Furthermore, we also found that comorbidities could facilitate patient use of a DHI if this had a positive effect on both LBP and the comorbidity, for instance, in the case of hypertension and LBP. Similar to previous studies [28,29,42,46,47], we also found that technical problems with the portal or little technical skill of the patient were an important barrier to portal use. The technical requirements and required technical skills of a DHI are particularly important because they can exacerbate inequalities in access to quality health information [48,49]. For example, older people and those with lower education and lower income could be disadvantaged, as a larger proportion of these groups do not have access to or cannot use the internet [50,51]. Therefore, when developing and using a DHI, care should be taken to ensure that the information contained could be made available to patients in a nondigital format. The web portal tala-med considered this by offering a hard-copy brochure that covered all topics of the portal, as well as printable information graphics, fact sheets, checklists, and exercise sheets that contained summaries of single topics.

Overall, patients perceived tala-med as usable, rated it positively, and considered it a clear, comprehensible, trustworthy, and practical resource that they would recommend to others. Our findings are consistent with previous studies showing that approval by an HCP supported patient trust in the quality of the contents of a DHI [29,52]. In addition, they provided insights into further characteristics of the portal content or its presentation that also contributed to the portal’s trustworthiness from the patients’ perspective. The usability of tala-med could be further enhanced by offering it as a smartphone app, in addition to greater customization and the inclusion of information to help patients find or contact back pain specialists. The latter 2 suggestions corroborate previous studies in which patients with LBP felt that the DHI could be more customized to their needs and provide an opportunity to contact an HCP [15,53]. Patients perceived tala-med as offering added value to other sources of information. In particular, patients felt that they could trust the portal; its contents; and its provision of in-depth, comprehensive, and appropriate information. Trustworthiness, in particular, reflects an important feature of web-based health information [34] that is deficient in many websites [15]. Tala-med may remedy the information dilemma faced by many patients when searching for health information on the internet by adding value to self-guided web-based searches.

The practical relevance of tala-med and its reported positive effects underline the fact that patients were able to understand and make good use of the portal’s information, while also adding value to the PCP consultation. The ability to use tala-med independently and flexibly is an advantage typical of DHIs [28,42]. As a consequence of these features and the incorporated exercise videos, the portal motivates and empowers patients to manage their back pain whenever there is a need or free time to do so. By providing informational, motivational, and practical support, the portal contributed to a decrease in back pain and reduced the need for patients to visit their PCP repeatedly because of LBP. Receiving the portal from their PCP added value to patients. It enriched preceding consultations by allowing patients to repeat and reflect on the contents of the consultation, thus increasing patient informedness while also facilitating and encouraging patient participation in future consultations through increased prior knowledge. These findings are in line with a recent systematic review, which suggests that by using DHIs, “users improved understanding of LBP and enhanced communication with their HCP during subsequent consultations” [28]. Overall, patient perceptions support the utility of the portal, especially when used in combination with PCP consultations.

PCPs could make good use of well-designed web portals for LBP as a supplement to their consultations [35]. Patients welcomed this additional web-based resource if they were familiar with digital technology and the internet. To support patient acceptance of a DHI and its positive effects, it should be integrated into consultations or patient treatment plans. Even if patients can easily use the DHI on their own, PCPs play a crucial role in deciding whether their state of health, in particular pain intensity, comorbidities, and further treatments are compatible with using a given DHI, and which one may offer the most helpful content to patients in terms of prescribing exercises or further information. PCPs and patients benefit from using DHIs such as tala-med as a supplement to their consultations [47], as it prepares patients for upcoming consultations and increases their participation. Because the aforementioned implications and their underlying findings are not unique to patients with LBP or tala-med, they may be generalized to other DHIs providing health content [21,22] used during and after PCP consultations.

With 32 half-hour interviews, our qualitative study had a comprehensive information base, including both patients who used tala-med and those who did not. Nonetheless, our study relied on a self-selected sample of only 17% of all invited patients. As a result, negative and rare experiences with our portal may have been missed because the patients who experienced them did not participate in our study.

Compared with previous studies, a strength of our study was that patients were able to report their actual experiences with the consultation and the web portal, rather than their anticipated preferences [15]. However, for several patients, these experiences were not particularly vivid at the time of the interview, as the interviews took place at least 4 weeks after the consultation. These patient perceptions of their PCP use of the portal during the consultation may reflect their PCP’s actual behavior unreliably, as subtleties of the consultation, such as brief uses or remarks about the portal during the consultation, may have been forgotten. Recall difficulties may also have led patients to perceive distinguishable aspects of usability less distinctively owing to halo effects [54] and may have led to more socially desirable responses. When investigating details of the portal, such as its usability, it would have been helpful to have the portal open in front of patients, as they were interviewed.

Researchers from different departments developed the portal (AM and SVR) and conducted interviews (CS and M Klimmek). Nevertheless, patients may have mistakenly assumed that they were speaking with someone who had also developed the portal, as the interviewers worked at the same medical center as the portal developers. Thus, patients may have felt inhibited in expressing any criticisms of the portal. However, this limitation only related to patient experiences of the portal. The fact that the interviewers were employed at a university medical center independent of and in a different federal state to that of the PCPs appeared to offer an advantage, in that patients did not have to be concerned when speaking freely about their perceptions of the PCP consultation.

Future studies that quantitatively investigate how patients perceive and evaluate tala-med or similar informational DHIs and their effects may mitigate the aforementioned limitations. This could further expand our knowledge of the acceptance, usability, and utility of web portals for LBP.

Most patients accepted our web portal well. Patient perception also affirmed its usability and utility. Tala-med may thus mitigate the information dilemma of patients and seems well suited as a supplement to PCP consultations. The facilitators of and barriers to use in our study are consistent with previous findings and indicate that PCPs should consider pain severity, comorbidities, other therapies, and IT equipment and skills of patients to support their acceptance of the portal. The setting itself, that is, the distribution of the portal by PCPs, seems appropriate as it supports the perceived acceptance and trustworthiness of the portal by patients, as well as bringing added value to current and future consultations. Patient perceptions thus highlight the appropriateness of the portal and the setting. However, they also indicate that many PCPs do not make full use of the portal, rarely integrating it into their consultations. Beyond feedback on tala-med and its implementation, the insights of our study on the acceptance, usability, and utility of tala-med offer valuable suggestions for the development of DHIs and their successful use as supplements to PCP consultation.