This was a prospective pilot clinical trial that investigated the utility of the PainNavigator mobile app. This study assessed pain scores and functional outcome data from patients with chronic LBP before and after they used the platform. As this study was exploratory, all participants received the intervention, and there was no control group.

Study participants were provided with a digital informed consent form through the mobile app, and they provided a signature on acceptance of the terms of consent. Participant data were deidentified for privacy and confidentiality protection. Compensation in the form of cash gift cards was provided for participation. This research was approved by the Ascension-Illinois Institutional Review Board (RIL20210036) and was carried out in accordance with the Helsinki Declaration's ethical standards.

A total of 75 participants were recruited via health care providers within the Ascension-Illinois system. Participants either were given PainNavigator enrollment information, as deemed appropriate by their primary care provider during a clinic visit, or obtained the Ascension-Illinois flyer through Ascension-Illinois Marketing outlets. Participants were incrementally compensated up to US $80 in the form of cash gift cards based on the completion of modules. The inclusion criteria were men and women aged 18 years or older and those experiencing LBP for greater than 4 weeks that was nonsurgical and nonmalignant in etiology. Given the limited access among many people to multidisciplinary pain management clinics, this study aimed to target patients before they considered pain management clinics or those without access to such resources. The exclusion criteria were as follows: an age of under 18 years; the inability to complete the pain questionnaire in written English; the inability to utilize the software app; recent back surgery in the past 6 months; back pain due to malignancy; a diagnosis requiring surgery; uncontrolled depression, anxiety, or a severe mental disorder; severe medical conditions (including heart disease, lung disease, a history of strokes, and neurological disorders such as paralysis or uncontrolled seizures); patients who are advised against physical exercise or mental health self-therapy by a health care provider; patients undergoing interventional pain management techniques; patients undergoing any form of outpatient psychotherapy; pregnant patients; and adults who are unable to consent. Participant age and baseline pain duration demographics are shown in Table 1.

PainNavigator is a web-based mobile app that users can access for pain self-management (Figure 1). At baseline, the app gave participants pain and function questionnaires (Multimedia Appendix 1). These included demographics (ie, age, gender, and activity level); symptomatology (ie, the duration of LBP and the effect of LBP on behavior); treatments (ie, the names of opioids, the doses being taken [if applicable], and the likelihood of pursuing opioid or surgical intervention); the Pain, Enjoyment, and General Activity (PEG-3) scale; and the 4-item anxiety and depression questionnaire that was based on the Patient Health Questionnaire-4 (PHQ-4). This foundation promoted the development of a personal functional goal for users to achieve during the program. The app then guided users through prerecorded medical education and wellness strategy content, including evidence-based cognitive behavioral therapy techniques, yoga, mindfulness, and exercise therapy. The prerecorded content was led by a pain management physician, a health psychologist, a certified yoga instructor, and a doctor of physical therapy. The educational videos taught common causes of and solutions for pain and wellness strategies that users could learn for managing pain better. The movement modules taught exercises and stretches for back pain and mindfulness to emphasize a mind-body connection. Users set goals for activities that pain was holding them back from and, via weekly phone calls, worked one-on-one with a live wellness coach who supported goal achievement and program consistency. To solidify their understanding of the content, users completed postvideo actions, including the use of pain, mood, and food journals. Throughout the program, users leveraged these journals to identify triggers and drive behavior change. Upon finishing the program, users answered the completion survey, which included a pain and function questionnaire (Multimedia Appendix 2).

The data analysis included a comparison of deidentified, participant-inputted responses to questionnaires. Responses that were given before, during, and upon the completion of PainNavigator were analyzed for statistically significant changes in pain scores, as defined by standard statistical methods. Analyses were completed only for participants who answered each question, and composite scores, including the PHQ-4 and PEG-3 total scores, were only calculated for participants who answered each item that contributed to the scores. Data were analyzed in SPSS version 26 (IBM Corporation). The significance level was set to an α of .05. Data were first assessed for skewness. Paired sample t tests were used to compare mean differences in scores from before and after participants received the intervention. The Cohen d was calculated from the raw means that yielded effect sizes. Power was also calculated with G*Power version 3.1.9.7 (Heinrich-Heine-Universität Düsseldorf) [19] post hoc, using the sample size, α, and the effect size.

The preliminary results from this pilot trial support PainNavigator's initial acceptability and can help inform a larger randomized controlled clinical trial. Using the app significantly impacted all PEG-3 scale and PHQ-4 measures, indicating that the platform addressed essential components for the self-management of LBP. The findings show a 36% reduction in the PEG-3 total score, a 40% reduction in pain intensity, and a 40% reduction in the PHQ-4 total score. The magnitude of change for each measure illustrated these effects, with a positive percent change indicating score improvement (Figure 2 and Figure 3). The sample size was limited but acceptable for an initial pilot study. Given this sample size, the large effect sizes play an important role in study power. The strong power in this study indicates that the observed significant changes are likely clinically relevant and warrant a more rigorous study.

Published literature on pain-focused mHealth apps shows efficacy data that can be compared with data from PainNavigator. A meta-analysis of pain eHealth applications found that program durations of ≤8 weeks (“less intensive duration”) had an advantageous effect on reducing pain intensity [16]. The 8-week program described in this study similarly found significant improvements in pain, as assessed by the PEG-3 scale item Average Pain. These findings may suggest that concise programming facilitates user engagement and retention. Furthermore, the meta-analysis effect sizes of immediate and short-term pain intensity were found to be −0.16 (95% CI −0.30 to −0.02) and −0.27 (95% CI −0.43 to −0.11), respectively [16]. The absolute effect size for average pain following the PainNavigator intervention was 1.118, which is greatly higher in comparison. This may indicate that intrinsic app factors, such as design and functionality, assist with pain reduction.

Beyond pain measures, psychology may also play an important role in LBP management. An investigation of musculoskeletal pain rehabilitation outcomes found psychological and pain indices to be significantly intercorrelated [21]. This supports the biopsychosocial approach, as it appears to incorporate multidimensional patient care. PainNavigator demonstrated significant improvements in the psychological measures assessed by the PHQ-4. These findings may indicate the importance of psychological components in pain management programs.

As an emerging field, mHealth apps also have challenges. These include a general lack of quality mHealth app assessments [22] and the number of chronic pain apps that only offer 1 self-management strategy [23]. Many health apps are also mostly used until initial milestones, with significantly decreased usage afterward [24]. PainNavigator’s effect may be the result of the mobile app addressing some of the aforementioned challenges. PainNavigator's interdisciplinary approach introduces users to multiple strategies (eg, home exercise, yoga, wellness, nutrition education, etc) instead of focusing on only 1 modality. As such, users can utilize the modalities that provide the most benefits. Additionally, PainNavigator emphasizes function as an equal or greater part of the self-management of chronic pain when compared to pain scores. The wellness coach and personal goals hold users accountable for completing the program and provide a framework for continuing usage once initial milestones are met.

Due to the single-arm design and the lack of a control group, our results should be interpreted cautiously. Nevertheless, our results were quite promising for informing future rigorous tests of the PainNavigator intervention. Another limitation may be the usability of mHealth technology in older populations. Despite this, over half of the participants who completed this study were aged over 50 years (19/30, 63%). Other back pain mobile app studies additionally reported disability data, which this study did not explicitly assess. There were also differences between the number of enrolled participants and the number of those who completed all surveys, indicating a drop-off in responses. Lastly, financial compensations were given to participants for their participation in this study. Although this is a well-accepted practice in research studies, there is some debate about how this impacts compliance.

This study’s findings will guide future research into pain-centered mHealth app efficacy and scalability. For example, the analysis found that the Felt Anxious, Felt Depressed, and Loss of Interest PHQ-4 submeasures had medium effect sizes, whereas large effect sizes were found in all other measures; a future focus may be directed on these targets. Since user engagement with mHealth apps is essential to their efficacy, many studies have prioritized users’ needs during quality improvement and described their methodology [25,26]. Clinician feedback also may help direct LBP mobile app quality improvement [27]. Accounting for both types of input is essential, given the need for dynamic communication between patients and physicians. This feasibility study demonstrates that our protocol works, and more extensive randomized controlled trials will be conducted. mHealth app development is a dynamic process centered around providing patients with the best care.

The PainNavigator mHealth app showed LBP management utility in this initial pilot trial. The significant improvements in all PEG-3 scale and PHQ-4 measures illustrate potential multidimensional, biopsychosocial management that is easily accessible to patients. This platform demonstrates clinical significance and can be easily utilized to improve patient care. Further randomized controlled trials are needed to expand upon these initial findings and explore the functional role of the PainNavigator platform in clinical settings.