Video-based documentation is a growing practice in health care, but both clinicians and patients have concerns about associated ethico-legal issues [1,2]. Video recordings capturing specific aspects of a patient’s condition may offer advantages that could support patient assessment and monitoring, support clinical decision-making, and lead to better clinical outcomes. However, little is known about the application, acceptability, and impact of video recordings for direct care purposes.

Video recording has demonstrated value across research, education, and performance assessment, as well as in clinical audit and quality improvement [1,2]. Researchers value the richness of video data that include sound, environmental context, and body language, which facilitate the objective and accurate documentation of behavior [3,4]. Video-based feedback assists clinicians in recognizing visual and auditory clues during clinical consultations that cannot be derived from text-based learning, and video-based curricula have led to an accelerated learning curve for surgical trainees [2]. Furthermore, video-based surgical case reviews have informed quality improvement initiatives through the provision of increased detail and nuance beyond what exists in operative notes alone [2,5-7].

Certain aspects of a patient’s condition (such as functional ability, cognition, and behavioral symptoms) can be challenging to convey to others involved in a person’s care, and recognizing changes in a patient’s condition over time can be difficult, particularly when a patient receives care from multiple different clinical staff [8]. Moreover, information captured about a patient away from the clinical setting can be useful to inform diagnosis, care, and treatment (eg, symptoms occurring in the home). As ineffective communication of pertinent patient information is known to negatively affect care continuity and is linked to medical errors and adverse events [9,10], new consideration should be given to how best to capture and communicate these data [11].

One of the proposed solutions is video recording. Video recordings offer a detailed and objective record that can be reviewed repeatedly as a longitudinal record by multiple clinical staff members [8]. Video recordings provide a record that has not been condensed into words and thus preserves the original situation without the added lens of a patient, family member, or intermediary health professional [7]. However, privacy concerns and issues with storage in the electronic patient record have limited the use of video recording for direct care purposes [12,13]. Early studies suggested that patients can feel censored or self-conscious in front of a camera, and video recording can alter the dynamic between the patient and clinician in negative ways [13,14]. However, over the last decade, the proliferation of smartphones has resulted in photographs and videos being captured in everyday life and being used for agile communication. Technological advances are supporting lawful collection, handling, and cloud-based storage of visual data to address storage limitations and protect people’s privacy in line with data protection legislation [15-18].

At a time when video technology is ubiquitous in everyday life, little is known about how video recordings can support the safety and quality of individual patient health care. The aim of this review was to synthesize the international empirical literature on video recording patients for direct care purposes to evaluate acceptability, effectiveness, and ethico-legal considerations and to summarize the relevant guidance of professional and regulatory bodies in the United Kingdom.

The review protocol was specified in advance and registered in the PROSPERO database (CRD42022331825).

Two separate search strategies were used: (1) empirical studies (international) and (2) professional and regulatory guidance (United Kingdom only).

Owing to the inclusion of various types of evidence encompassing a variety of study designs and qualitative, quantitative, and mixed methods approaches, quality appraisal was undertaken using the Quality Assessment with Diverse Studies (QuADS) appraisal tool [23]. The use of a single tool permits the evaluation of methodological quality, evidence quality, and quality of reporting across a body of diverse evidence, and the QuADS tool considers the extent to which there is transparency and congruency in the research and its reporting and the implications for evidence quality [23].

We used the Joanna Briggs Institute Critical Appraisal Checklist for Case Reports [24]. For conference abstracts, we devised a bespoke appraisal tool based on the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) checklist of items to be included when reporting observational studies in conference abstracts [25]. Study quality was assessed independently by 2 reviewers.

The study’s findings were synthesized using a narrative approach. To evaluate acceptability, recruitment and retention rates were extracted or calculated (reported data permitting), and the reported views and experiences of patients and clinical staff were analyzed thematically using a framework-based synthesis [20]. The Theoretical Framework of Acceptability (TFA) of health care interventions (Textbox 1) was applied as a preliminary coding framework, and an inductive approach was used to derive new themes from the data [21]. The framework approach was also applied to synthesize reported ethico-legal considerations. To evaluate effectiveness, we selected studies reporting any clinically relevant, objective measure of impact of the video recording intervention, considering the results of observational studies and randomized controlled trials (RCTs) separately, because although RCTs are usually considered to be the gold standard design for decisions about the effect of interventions, they are arguably less applicable to real-world clinical practice than observational studies, and this review was concerned with direct care delivery [22]. Key recommendations for professional and regulatory bodies are summarized and tabulated.

We followed the reporting recommendations outlined in the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement [26] (Multimedia Appendix 2).

The searches of the MEDLINE, Embase, APA PsycINFO, CENTRAL, and HMIC databases yielded 14,218 study citations. Following the removal of duplicates, the titles with the addition or removal of abstracts of 8715 citations were screened. Of these, 5420 citations that clearly did not meet the inclusion criteria were discarded during the title screen, and a further 3205 were excluded on reviewing the abstracts. The full reports of the remaining 90 studies were retrieved and examined in detail. In total, 24 studies meeting the inclusion criteria were retained for review; 3 additional studies were identified through a hand search of reference lists (Figure 1). A total of 13 guidance documents were identified via the websites of public and regulatory bodies and Google search, of which 7 met the inclusion criteria and were retained for the review.

In total, 17 empirical studies were reported as full-text articles and were therefore suitable for methodological assessment using the QuADS appraisal tool [27,28,30-34,37-39,41-45,47,48]. Methodological and reporting quality was generally poor: only 4 studies achieved a maximum score of 3 in ≥10 of the 13 QuADS criteria [28,31,33,34] (Table S1 in Multimedia Appendix 4 [28-53]). Overall, QuADS scores tended to be higher in RCTs (mean total score 32.8, range 27-39; 4 RCTs) than in observational studies (mean total score 27.8, range 17-36; 13 observational studies). The quality of the 5 conference abstracts was mixed (Table S2 in Multimedia Appendix 4). The Joanna Briggs Institute Critical Appraisal Checklist indicated satisfactory quality of the 5 case reports (Table S3 in Multimedia Appendix 4).

Video recording patients for direct care purposes appears to be acceptable, and this is evident in the recruitment and retention rates in studies evaluating a video recording intervention and the predominantly positive views and experiences of patients and clinical staff members. Concerns about privacy and physical appearance were voiced by some patients, whereas clinical staff members were concerned about device security, technical issues, and operational and environmental considerations associated with implementing video recording in a clinical setting. Video recordings were perceived to be valuable in supporting patient assessment, care, or treatment; in promoting patient engagement; and in enhancing communication and recall of information. The results of observational studies suggested that video-based patient records are effective in supporting direct care delivery; however, RCTs demonstrated that the video recording intervention was not superior to the controls. Of note, methodological and reporting quality was generally poor considering the studies collectively and worse in observational studies compared with RCTs. This review summarizes and discusses relevant professional guidance and national legislation for UK practice, including recommendations for obtaining consent, protecting privacy and confidentiality, and securing data transmission and storage.

Although our review explored video recording in the context of direct care delivery, 2 previous literature reviews examined the evidence on video recording patients’ research and training purposes. A review of audio-visual recordings during general practitioner consultations for research purposes, published 15 years ago in 2008, found that patients who had not been videoed before were more likely to dislike the idea than patients who had previous experience with video recording during health care interactions [63]. In contrast, our findings are consistent with a later 2016 systematic review on the acceptability and design of video-based research, which reported that most patients and clinical staff regard video recording as acceptable and worthwhile for research and training [1]. It follows that the proliferation of smartphones over the last decade, their video recording functionality, and widespread adoption of videoconferencing during the COVID-19 pandemic have led to video recording becoming more acceptable in clinical practice for patients and clinical staff members [11].

Our review found that the factors influencing acceptability among clinical staff were technical issues and mobile device security. Similarly, previous research has reported issues with the installation or usability of video recording equipment, images not being transmitted to patient records, and mobile devices being stolen [12,64,65]. The inherent lack of anonymity in video-based records was reflected in patients’ concerns about their physical appearance or privacy; however, these concerns were voiced by only a small number of patients [31,34,42,48]. However, it is widely recognized that patients are apprehensive about how their personal data are handled and shared in the health care context [66,67]. A lack of transparency in how personal data are used undermines patients’ ability to trust that measures are in place to protect their privacy and thus reduces their willingness to engage with health information technologies [68]. As such, adherence to relevant legislation and professional guidance is not only essential from a medico-legal perspective, but it will also be important for the successful implementation of video-based records within health care settings.

Although video-based records were perceived to enhance several aspects of direct care delivery, the methodological and reporting quality of the body of evidence was generally poor, which limits the conclusions that can be drawn regarding effectiveness. Among the 5 RCTs, video recording interventions were not associated with improvements in clinically relevant outcome measures, although most trials were underpowered. Of note, our study was concerned with direct care delivery, and the applicability of RCTs to real-world clinical practice has been debated [22]. Although the research settings and target populations of the included RCTs were generally well described (Tables S1 and S3 in Multimedia Appendix 4), small sample sizes and lack of discussion around contextual factors that may have influenced the results make it difficult to judge how applicable these trials are to other settings. The results of the included observational studies (arguably more applicable to real-world settings if well designed) favored video recording, but the study designs lacked adjustment for confounders and were not suitable for estimating causal effects. A recent systematic review evaluating home videos for seizures demonstrated consistent improvements in diagnostic accuracy and clinical decision-making [69]. However, some studies in that review did not evaluate video recordings in a real-world direct care delivery context, and similar to our review, the findings were based on studies with diverse patient populations, interventions, and outcome measures that were not designed to test effectiveness, thus limiting internal and external validity.

Pertinent to video recording patients in health care, data protection legislation sets out how personal data must be collected, handled, and stored to protect people’s fundamental right to privacy. Although some countries, such as Canada, have complex data protection legislation at the federal and provincial levels, there is consistency in legislation across the globe pertaining to key issues, such as consent and data security, as laid out in the Personal Information Protection and Electronic Documents Act (2000; Canada) [70], the Data Protection Act (2018; the United Kingdom) [15], General Data Protection Regulations (the United Kingdom and European countries) [16,17], the Health Insurance Portability and Accountability Act (1996; the United States) [18], and the Privacy Act (1988; Australian Privacy Principles) [71]. As professional and regulatory guidelines are underpinned by data protection legislation, it follows that there is consistency in the recommendations for audio-visual recording patients. The professional and regulatory guidance collated for UK practice as part of this review lacks detailed information about how to operationalize the recommendations in clinical practice. Furthermore, there is tension between ensuring patient confidentiality and the accessibility of video recordings to inform clinical decision-making [51]. The Caldicott principles (Textbox 2) underline the need to strike a balance between protecting patient privacy and sharing information responsibly for individual care [56]. Technological advances permitting storage of electronic health records on secure cloud servers, integration of mobile device apps with electronic health records, and secure information exchange between providers will likely support health care organizations in achieving this balance. Although current professional and regulatory guidelines in the United Kingdom make recommendations for securely storing video recordings, they do not consider the importance of recordings being easily accessible to clinicians for direct care purposes.

Ensuring public trust and confidence in how personal data are used by the health service is a central concern for health care organizations and policy makers [72-74]. High levels of patient trust are paramount for the implementation of video-based patient records, particularly given the inherent lack of anonymity in video recordings. Clinical staff involved in delivering video recording interventions must be equipped to provide patients with information on how privacy and confidentiality will be protected [75]. As data controllers, health care organizations should have robust policies and procedures for operationalizing data protection legislation when implementing video-based patient records. This will include arrangements for secure storage and transmission of video recordings and facilities for remote content deletion where portable devices are used.

The methodological and reporting quality of the studies included in this review was generally poor, and further research is needed, particularly for the effectiveness of video-based patient records in direct care delivery. Although RCTs are considered the “gold standard” of causal impact evaluation, they should adopt a pragmatic attitude to evaluating video-based records for direct care purposes and be reported according in line with the CONSORT (Consolidated Standards of Reporting Trials) statement extension for pragmatic trials [76]. Theory-based mixed methods approaches are recommended to identify and explore the active ingredients of video recording practices, including the contextual and behavioral factors driving possible effects [77].

A strength of this review lies in the comprehensive synthesis of qualitative and quantitative data and professional guidance to address a range of important considerations for the implementation of video-based patient records. A robust systematic review process was followed, with 2 reviewers independently undertaking screening, data extraction, quality appraisal, and coding of qualitative findings. We ensured rigor and transparency in the qualitative synthesis by applying the framework method to derive insights from the included studies [20]. However, the number of studies on video recording for direct care purposes is small and scattered across traditional disease boundaries. Therefore, despite our efforts to carefully perform the search and selection processes, we may have missed some relevant studies. As it was anticipated that the body of literature would be small, we included studies with different designs and did not exclude low-quality studies. Heterogeneity relating to the study populations, video recording interventions, and outcome measures as well as the collectively poor methodological and reporting quality of the studies mean that firm conclusions cannot be drawn, particularly in relation to whether video-based patient records are effective in supporting direct care delivery. This review sought to collate professional and regulatory guidance for video recording patients in the British National Health Service. Although these recommendations may not directly translate to health care settings outside the United Kingdom, there is broad consistency in data protection legislation around the globe regarding several key issues, such as seeking consent and implementing data security measures.

Video technologies have been piloted in various health and care contexts to support diagnosis, care, and treatment. Video recording practices appear to be acceptable to patients and clinical staff despite concerns about privacy, technical considerations, and integration into clinical workflows. The methodological and reporting limitations of the studies included in this review prevented firm conclusions from being drawn. Pragmatic, mixed methods trials, particularly in the field of movement disorders, older adult care, and in patients’ homes, should evaluate how video recording impacts diagnosis and treatment monitoring, patient and care provider communication, and patient safety. Professional and regulatory documents should provide practical guidance for the secure, ethical implementation of video recording in routine practice.