Bell palsy is an acute unilateral paralysis of the facial nerve, resulting in a patient partially, or completely, losing the ability to voluntarily move facial muscles on the affected side of the face [1]. It is the most common acute disorder affecting a single nerve, and its cause is unknown [2]. Each year, the condition affects 11 to 40 people per 100,000 in the population, most commonly in the age group of 30 to 45 years [3]. The annual incidence of Bell palsy in the United Kingdom is currently 37.7 per 100,000 population [4]. Bell palsy represents only approximately 60% of all facial nerve paralysis (FNP) cases [5]. The total number of FNP cases occurring annually in the United Kingdom is estimated to be at least 22,500, and 1 in 60 individuals will be affected over the course of their lifetime [6,7].

Epidemiological studies indicate that this neurological condition occurs more commonly in those with diabetes, obesity, hypertension, and upper respiratory conditions and people who are immunocompromised or pregnant [2,6,8] or following infection by a virus such as herpes simplex [9]. Data from the United States show a recent rise in incidence, possibly linked to increasing rates of herpes infections [10]. Without intervention, some patients will show an element of recovery within 2 to 3 weeks and complete recovery within 3 to 4 months [2,3]. However, although normal facial function is completely restored in approximately 70% of cases, 30% will have a poor recovery [11,12] with facial disfigurement and sometimes facial pain [3,13], and up to 16% of those affected will have residual involuntary movements known as synkinesis [3]. Research shows that people with these residual deficits experience a long-term reduction in quality of life, psychological distress, depression, and social alienation, often relinquishing a previous public-facing role [3,13-16]. As a result, patients with FNP continue to have relatively low public visibility, unless a high-profile international star reveals their own diagnosis [17].

Although various treatments are available, uncertainty exists regarding the effectiveness of many of these. Cochrane systematic reviews have confirmed the effectiveness and cost-effectiveness of corticosteroids (prednisolone) administered within 72 hours of onset of symptoms [18-20]. Beyond this initial treatment, for those with incomplete recovery, there are a number of medical options available. Various surgical procedures, together with botulinum toxin injections can attempt to normalize facial appearance [21-25]. However, a Cochrane review of surgical interventions has reported that there is insufficient evidence to decide whether such procedures are beneficial and has also concluded that further trials are unlikely [26].

Physical rehabilitation therapy can be used as an adjunct to medical treatments. The use of facial neuromuscular retraining (NMR) to strengthen muscle and improve nerve function has been evaluated more than other physical therapies [27-32]. A 2011 Cochrane review has concluded that there is some evidence that NMR can improve facial function (for moderate nerve paralysis and chronic cases) and reduce sequelae in acute cases, although it was recommended that both need to be confirmed in randomized controlled trials [33]. An update of this systematic review is currently underway [34]. A recent review of physical therapy combined with standard drug treatment (SDT) has reported evidence of positive effects on grade and time to recovery compared with SDT alone [35].

In the context of an incomplete evidence base, current international guidelines highlight the need to consider patients’ experiences and preferences [2,36]. Recent clinical practice guidelines from the United States, which conclude that physical therapy can provide potential functional and psychological benefit, add that there is a “large role for shared decision making” [2]. Although Canadian guidelines make no recommendation regarding the use of facial NMR in the acute phase, owing to a lack of good quality trials and risk of bias, its use is suggested for patients who do not have complete facial recovery [36]. In the United Kingdom, clinical guidelines produced by the National Institute for Health and Care Excellence (NICE) recommend rapid initial medication (prednisolone) and referral to a range of hospital-based medical specialists [37]. Guidelines for commissioning neurology services in the United Kingdom also include a recommendation to consider new transformational technologies [38].

The potential for the “digital patient” to transform the delivery of care has been demonstrated in many clinical areas [39], but there is a lack of evidence for FNP. The aim of this study, conducted in collaboration with patients and specialist clinicians, is to gather evidence about FNP treatment pathways in the United Kingdom, the referral process and timing of NMR, current outcomes used to monitor recovery, and the potential role of digital technology to assist in rehabilitation. The study was conducted as part of a research program (Facial Remote Activity Monitoring Eyewear [FRAME]) funded by the UK National Institute for Health Research (NIHR). The FRAME program aims to develop inconspicuous miniaturized sensor devices in spectacle frames to measure facial movement, providing biofeedback to patients and access for clinicians to outcome data, and to conduct early health technology assessment of the wearable FRAME digital technology shown in Figure 1 [40].

To provide context for technology introduction, national surveys were conducted to explore patients’ and specialist therapists’ experiences of FNP care pathways in the United Kingdom and the potential role of digital technology. Separate questionnaires were co-designed for patients and therapists, piloted, and refined following feedback. Each questionnaire included a mix of closed and open-ended questions (Multimedia Appendix 1) and followed a similar format to enable cross-referencing of some responses [41]. Questionnaires collected demographic details, information on treatment pathways, and ratings of the importance of treatments personally experienced. Respondents were encouraged to add textual comments to expand on responses to closed questions. Further open-ended questions explored the potential value of the emerging FRAME technology. Both questionnaires were uploaded on a web-based platform.

For patients, an open recruitment strategy was adopted to achieve geographical spread. People with experience of FNP were recruited in collaboration with the national charity, Facial Palsy UK [42]; the patient survey was advertised widely, including via social media. Recruitment of specialist therapists was coordinated through a nation-wide professional group, Facial Therapy Specialists UK [43]; all members were emailed a personal invitation containing a link to the questionnaire, and the initial email was followed by 2 reminders over a period of 5 months.

A preliminary analysis of responses uncovered a range of methods used to report treatment outcomes. Therefore, a further survey was conducted to examine these in greater detail, especially any use of validated scales, and whether there is consensus on the definition of a clinically significant change for these measures. A convenience sample of 50 clinicians attending the Facial Therapy Specialists Annual Meeting held in London (October 1, 2018) were invited to complete this questionnaire.

The closed questions were analyzed using SPSS (version 25, IBM Corporation). Response patterns were summarized using mean and SD or median and IQR, depending on data distribution. Certain variables were grouped to explore changes over time (eg, time since diagnosis). Open-text comments expanding on responses to closed questions and text replying to open questions were analyzed using thematic content analysis [44]. Inductive coding was used, following a flexible analysis approach that helped account for any further themes emerging during the coding process [45]. Data were coded and analyzed for thematic patterns and meanings within the data until saturation was reached.

Ethical approval was granted by the Health and Life Sciences Research Ethics Committee, University of Coventry (ref: P48908).

An analysis of treatment pathways was conducted for patients with adult-acquired FNP. Respondents reported receiving an average of 3.27 (SD 1.60) different treatments following initial diagnosis, most commonly corticosteroids, advice on eye care, and facial NMR, as shown in Table 1.

Adult patients reported varying levels of adherence to their prescribed NMR program, with only 33% (32/97) recording very high levels, 41% (40/97) reporting medium-to-high adherence, 21% (20/97) recording poor-to-medium levels, and 5% (5/97) being uncertain about adherence. An analysis of patients’ open-text comments identified 3 main barriers, regardless of the level achieved: “difficulties in fitting facial exercises into daily life,” “using a mirror while completing exercises,” and “insufficient regular follow-up.” Patients who reported a very high level of adherence described 3 further facilitating factors: “observing improved outcomes in self,” “belief that the treatment will work,” and “observing positive outcomes in others.” These patients also highlighted a common fear that acted as a spur: “scared of not recovering.” An analysis of comments made by the medium-to-high adherence patient group uncovered 2 additional demotivating factors: an “inability to see any improvement” and “lack of funds to travel for check-ups.” In terms of travel, distances can be considerable. Although most adult patients (75/97, 81%) were referred to a center within their own region, 9% (8/93) reported traveling ≥115 miles to reach a specialist center. Among patients with poor-to-medium adherence, 2 further barriers were identifiable: “pain associated with facial exercises” and “exercises can be tiring”, possibly indicative of poor execution of a prescribed facial exercise regime.

When asked to describe their patients’ level of compliance, 36% (9/25) reported very high levels, 56% (14/25) reported medium-to-high adherence, 4% (1/25) recorded poor-to-medium levels, and 4% (1/25) were uncertain about adherence. Interestingly, therapists appear to have underestimated the level of poor-to-medium adherence, perhaps suggesting overoptimistic feedback by patients. The thematic content analysis shows that therapists identify the same 3 barriers limiting adherence (ie, “fitting exercises into daily life,” “lack of evidence of improvement,” and “use of a mirror”). Unlike patients, therapists did not identify patients’ travel costs. However, they did additionally think that the timing of other treatments influences adherence, for example, “poorer adherence post-surgery because this may be viewed as the primary treatment for managing their condition” and “Botox given too early may reduce compliance if patients rely on this to relieve feelings of tightness and synkinesis.”

Table 2 shows that 73% (70/96) of patients received regular feedback on their recovery during the course of treatment. This is provided in a number of ways with no consistent pattern nationally. Table 3 indicates that the regularity of feedback to referring physicians varies; 52% (12/25) of therapists provide feedback throughout treatment, but 40% (10/25) only report a final outcome. Final discharge summaries also vary, some include validated measures such as the Sunnybrook Scale and others provide photographs showing the patient’s progress.

In response to the follow-up survey, clinicians reported the experience of using various methods for recording treatment outcomes ranging from photographic evidence to recognized disease-specific scales. When asked to name scales they had used, 17 different instruments were identified (Table 4). The most frequently cited was the Sunnybrook Scale (26/28, 93% of respondents) [46], followed by the Facial Disability Index (FDI) identified by 54% (15/28) of respondents [29], and the House-Brackmann (HB) Scale identified by 50% (14/28) of clinicians [47]. Relatively few respondents were able to define a clinically significant change for these measures: 27% (7/26) for Sunnybrook, 7% (1/15) for FDI, and 21% (3/14) for HB. Where answers were provided, there was no consensus.

A thematic analysis of open-text comments on the potential value of digital technology identified 4 superordinate themes, as shown in Textbox 1. The first theme, System, was predominantly voiced by specialist therapists who highlighted the potential for this technology to help reduce pressure on their time, improve their ability to monitor a patient’s progress, increase coverage, and reduce travel for outpatient reviews. An extra subtheme, expressed by patients within the System theme, was that new technology could help raise medical awareness, especially in primary care. A second theme, Self-management, was identified by both groups. This focused on improving treatment adherence and emphasized factors such as enabling people to fit facial exercises into their daily routines and providing motivational feedback through regular review of daily performance.

The third theme, Identity, emerged mainly from patients’ responses. This highlighted psychological benefits such as improved confidence in people who are frightened to draw attention to themselves and hope in individuals that their condition might improve. The final theme, Innovation, encompassed views, such as, that innovation is natural and that it can bring positive societal benefits, together with an awareness that funding constraints limit implementation of innovations. Patients demonstrated a great willingness to support future introduction of digital technology, but they also emphasized the need to raise awareness among GPs alongside implementation. Therapists expressed a similarly positive view of telerehabilitation, with a clear consensus that digital technology could not replace the initial face-to-face consultation because patients need to be carefully trained to be able to complete their exercises properly.

A number of limitations should be borne in mind when considering this research. First, patient respondents may not be representative of the wider population because participants were recruited via a specialist support group. Second, there may be recall inaccuracy when participants are asked to provide information sometime after the event. Third, the therapist RR (51% national survey and 56% follow-up survey) means that data may not fully reflect the national picture. Finally, although the greatest care was taken in the questionnaire design, as with all surveys that record individuals’ views, the validity and reliability of the data could not be tested independently.

To date, little research has explored the potential value of digital technology in assisting facial NMR therapy. This study provides a baseline overview of FNP treatment pathways in the United Kingdom, the factors limiting access to NMR and influencing therapy adherence, the main methods used to record treatment outcome, and the potential role of digital technology. The study indicates that harmonization of outcome measures is required to both strengthen the evidence on treatment effectiveness and to better support MDT management. The main factors limiting NMR adherence could all be addressed through the use of real time digital technology. However, for the type of wearable technology being considered, product design will be an additional factor likely to influence adherence [64], especially in the younger 30- to 45-year age group affected by this condition [3]. However, although the study clearly demonstrates positive attitudes toward the introduction of digital technology, economic barriers may prove to be a challenge. Previous research has identified funding as a barrier for access to surgical treatments for FNP in England [65] and more recently in Wales, Scotland, and Northern Ireland [66]. This study has similarly identified funding barriers to NMR referral. Finally, organizational and cultural factors are acknowledged to act as important barriers to implementation for all digital health innovations [67], often reinforced by policy priorities [68]. One legacy of the COVID-19 pandemic is that health systems worldwide are rapidly adopting digital options in many clinical areas [69]. Thus, barriers to the introduction of digital technology to assist facial NMR therapy may now be lower, especially if cost-effectiveness can be demonstrated.