Teledermatology in the digital age: Evaluating clinical efficacy, accuracy, and challenges – A literature review

Benefits of TD

The use of this technology by all medical professionals has been praiseworthy since it has made it possible to communicate and share information with thousands of patients, has improved diagnosis, prevention, and treatment, and has prompted ongoing training for all medical professionals. Due to the COVID-19 pandemic, this service has undoubtedly expanded in terms of the services it offers (such as emails, video calls, phone calls, tablet and smartphone applications, support groups, and short message services) as well as the remote support it offers patients so they are not left behind. Our analysis revealed that during the COVID-19 epidemic, using this service has been essential. The use of this technology by all medical professionals has been praiseworthy since it has made it possible to communicate and share information with thousands of patients, has improved diagnosis, prevention, and treatment, and has prompted ongoing training for all medical professionals^[5] [Figure 1].

Close

Figure 1:

Objectives for expansion of telemedicine.

Export to PPT

Telemedicine seems well-suited for the field of dermatology, which is one of the most visually dependent specialties. The majority of skin conditions are visible to the naked eye, allowing imaging tools to capture them precisely. Determining, storing, and transmitting clinical data is the primary objective of TD. Melanoma, the deadliest type of skin cancer, is caused by a malfunction in the melanocytes, which are cells that produce the pigment known as melanin. The past 30 years have seen a sharp increase in the morbidity of primary cutaneous melanoma. Only 1% of skin cancer cases in the US end up being aggressive melanoma, despite skin cancer being the most common diagnosis. Many newly discovered instances of most skin disorders are apparent to the unaided eye in 2018, making it possible for imaging devices to accurately record them. Tele dermatology’s main goal is to identify, store, and transmit clinical data. Melanocytes, the cells that make the pigment known as melanin, malfunction in melanoma, the most fatal form of skin cancer.^[6] In patients with lesions diagnosed by the TD clinic and then biopsied, we found a 62% diagnostic concordance between clinical and histological diagnosis. This number is comparable with earlier studies of diagnostic concordance utilizing the SAF approach for dermatology.^[7] In the current investigation, the two dermatologists had 97.7% diagnostic concordance. This rate is greater than in previous research, which can be due to the fact that the TD diagnosis in this study could not able to be verified by an in-person diagnosis. The physician quality rating scale (PQRS) score was higher in cases when both dermatologists made a diagnosis with complete certainty. Diagnoses of eczematous illnesses, vasculitides, connective tissue disorders, and immunobullous disorders were among the 2.3% of patients with diagnostic discordance. Due to their chronic nature, diverse morphologic patterns, and often huge surface area, many illnesses are challenging to represent in a few clinical pictures. To get a conclusive diagnosis, the majority of these conditions need tests like skin biopsies or immunofluorescence analyses. Illnesses such as this rate are greater than in previous research, which can be due to the fact that the TD diagnosis in this study could not able to be verified by in-person diagnosis. The PQRS score was higher in cases when both dermatologists made a diagnosis with complete certainty. Diagnoses of eczematous illnesses, vasculitis, connective tissue disorders, and immuno-bullous disorders were among the 2.3% of patients with diagnostic discordance^[8] [Figure 2].

Close

Figure 2:

Concordance versus disconcordance in teledermatology.

Export to PPT

Accuracy analysis of telemedicine in dermatology

Because dermatology places a strong focus on visual diagnosis, TD has become a more popular medical tool worldwide. Research on this specialism has been conducted worldwide.^[9] According to most TD research, accuracy rates are between 75 and 80%, which is similar to in-person care. Numerous articles addressed general dermatology or skin neoplasms, particularly skin cancer and pigmented lesions.^[10] Research done compared International Classification of Diseases-10 codes filled out by teledermatologists and in-person dermatologists, since patients had already been triaged by TD. Accuracy was measured by the percentage of complete, partial, and no agreement rates between the teledermatologist’s and the dermatologist’s diagnoses. For total agreement, we also computed Cohen’s kappa, a statistical indicator of inter-rater agreement. Variability by disease was shown by the mean complete agreement rate of 78% (31–100%) and kappa = 0.743 for all 20 dermatoses, followed by partial agreement at 8% and no agreement at 14%. The research demonstrated the great accuracy of TD for inflammatory dermatoses.^[11]

Challenges

Numerous obstacles prevent TD from becoming widely used and accurate. Assessments involving non-specialists frequently have lower agreement rates between TD and in-person consultations, underscoring the necessity of appropriate image acquisition training and the use of top-notch equipment.^[12] The accuracy and dependability of diagnosis are decreased by subpar image quality, irregular methods, and a lack of standardized processes. Concerns regarding TD’s function in conclusive diagnosis are further raised by the very low concordance between it and histopathology, particularly across lesion types. Although teledermoscopy has the potential to increase accuracy, study designs frequently obscure its actual advantages by combining data from studies with and without the use of dermatoscopes or by excluding low-quality images, which fails to capture the complexity of actual practice. Furthermore, heterogeneity, a lack of randomized controlled trials, recollection bias, and insufficient stratification by study design are common problems in research that lead to inconsistent evidence and inconsistent results.^[13]

These diagnostic difficulties are accompanied by more general systemic and implementation-related obstacles. The lack of standardized frameworks for training, reporting, and data collection is reflected in the significant variations in agreement rates between TD and in-person consultations among researchers. Although live video consultations have demonstrated potential in enhancing diagnostic performance, there isn’t enough research to draw firm conclusions. It is challenging to evaluate the efficacy of TD in various healthcare systems due to the under-representation of studies from low- and middle-income nations and the exclusion of grey literature, both of which further limit generalizability.^[12] Data security and privacy are still major issues, particularly when integrating patient photos and videos into electronic medical records. Lastly, even though artificial intelligence has a lot of potential to improve TD, especially in picture recognition and diagnosis, its function has not yet been well investigated, raising considerable concerns regarding potential future uses.^[14]

Despite the widespread use of TD, veterans frequently have to make two separate trips to Veterans Affairs Medical Centers (VAMCs): One for treatment and one for biopsy. The Atlanta VAMC undertook a cross-sectional study to see whether TD may expedite care by triaging patients for Mohs surgery. One of the four attending dermatologists evaluated the standardized image sets that the dermatology residents had acquired, which included dermoscopic, closeup with ruler, and distant localization photographs. 321 lesions were evaluated in person and included in the analysis out of 376 consultations where non-melanoma skin cancer (NMSC) was mentioned as a potential diagnosis. Lesions were classified as high-suspicion when NMSC was the primary diagnosis and as low-suspicion when NMSC was a secondary consideration.^[10] True positives, true negatives, false positives, and false negatives were defined using biopsy data, and diagnostic accuracy metrics were computed by classifying lesion locations based on Mohs Appropriate Use Criteria.

Comparison between remote and in-person

The concept of TD was first introduced in 1995 to offer remote dermatology services. More recently, the arrival of the COVID-19 pandemic changed the scheme of in-person consultation within the healthcare system, except for those cases deemed emergencies. Although TD was already in use worldwide, the arrival of COVID-19 and the subsequent confinement period consolidated the application of the tool.^[4]

Research contrasting in-person dermatologists with TD reveals significant differences in diagnosis agreement. To compare diagnoses made in-person versus remotely, TD research should include two separate in-person consultations. High levels of diagnostic concordance were found in some studies but not in others. According to a meta-analysis, the concordance rates reported in clinical (in-person) consultations and TD consultations differed considerably (odds ratio = 0.55 [Mantel-Haenszel, fixed effect model, 95% confidence interval = 0.42–0.72], χ² = 11.87, P < 0.05, I2 = 58%). The overall findings demonstrated that primary diagnoses made in person are substantially more concordant than those made remotely. The findings also imply that TD and in-person diagnoses differed somewhat but significantly from remote diagnosis.^[15] The consultant dermatologists then review the computerized medical data, look at the pictures, and suggest a course of action or therapy at the referral hospital. PC doctors subsequently assess these suggestions and get in touch with the patient to share the findings. TD care encompasses a number of domains, including synchronous, asynchronous, and hybrid approaches. Since the dermatologist sees the patient’s photograph in real time (live), synchronous TD provides an instant diagnosis. The asynchronous approach, also known as storage TD, involves sending the patient’s photos to a dermatologist who makes a diagnosis after they have been put onto a platform via a PC. The asynchronous model is the most commonly used in clinical practice. There are fewer waiting lists because patients who need an in-person appointment or have an unclear diagnosis are contacted sooner. Compared to conventional dermatology consultations, TD has certain advantages. It lowers waiting times and expenses, enhances accessibility in remote locations and/or for patients who have trouble making in-person visits, and lowers absenteeism and appointment cancellations. In addition, TD favors early referral in emergency situations and is a valuable diagnostic and follow-up tool.^[4]

In a study, two dermatologists who saw each other in person had an 83.3% agreement rate for the primary diagnosis, while two dermatologists who reviewed histories and photographs had an 81.0% agreement rate. The range of agreement between dermatologists who saw photographs and dermatologists who saw them in person was 78.2% to 83.9%.^[16] In the second study, two dermatologists (an attending and a resident) examined 214 patients in person and produced a consensus diagnosis. They compared their findings with those of other dermatologists who used SAF techniques and uncompressed and compressed high-definition video. To create a baseline, the two in-person dermatologists also separately assessed a selection of 134 individuals. The consensus kappa values for in-person attendees and residents were 1.0 for partial agreement and 0.91 for primary agreement. The SAF method yielded comparable kappa values of 0.76 and 0.87 for primary agreement, 0.76 and 0.89 for partial agreement with uncompressed video, and 0.72 and 0.88 for primary agreement and partial agreement with compressed video.^[15]

In a visual specialty like dermatology, virtual doctor–patient encounters offer a great chance to improve care delivery; nonetheless, they were notably neglected until March 2020.^[17] The use of inpatient TD, which has been demonstrated to be on par with live hospitalist evaluation for triage, diagnosis, and management of hospitalized patients for a wide range of illnesses, has increased since the COVID-19 epidemic.^[18] In particular, telemedicine has shown itself to be a cost-effective, efficient, accessible to underprivileged groups, and a dependable consultation method compared to in-person dermatology consultations^[17] [Figure 3].

Close

Figure 3:

Comparison between in-person and remote teledermatology.

Export to PPT

Management and follow-up in TD

The TD modality (SAF, real-time, hybrid), consultation, follow-up, and referral procedures all have a role in defining an effective business and reimbursement model. Patient-assisted follow-up care at home for chronic patients eliminates the need for lengthy doctor’s appointments during work hours. When they have time, patients can take pictures with their smartphone and send them to their doctor.^[16]

It has been demonstrated that an effective TD consultation for NMSC can reduce wait times for biopsy or diagnostic excision, provide a prompt consultation for rural residents, and address the high demand for diagnosis and lesion-specific follow-up for NMSC and its antecedents.^[19] The following is a work that shows a follow-up of patients with Alopecia areata using TD. Patients with alopecia areata were divided into two groups. The equipment required for a telemedicine follow-up was given to one group. The follow-up for the other group was traditional. One year was the length of the follow-up period. The images acquired using telemedicine and outpatient trichoscopy showed a high level of compatibility. Patients visited in the outpatient clinic and those monitored by telemedicine had similar satisfaction levels.^[20] Another study examined all of the TD referrals from the emergency room of our Singaporean tertiary hospital between June 2015 and December 2019. The dermatological issues, triage, and recommended treatments were analyzed. Follow-up plans were recorded. Between June 2015 and December 2019, 147 emergency department patients were sent to TD; 11 (7.5%) were admitted, and 136 (92.5%) were recommended for discharge with a dermatological diagnosis and treatment plan. If required, a follow-up visit to the dermatological specialty clinic was planned. 110 of the 136 patients who were released for follow-up were among the 129 (94.9%) patients who were given an outpatient appointment at the dermatology clinic. Twenty patients (18.2%) had their TD diagnoses changed following an in-person evaluation, while 90 patients (81.8%) had their original TD diagnosis. With the right specialty-directed care, follow-up outpatient appointments that are adequately triaged, and return guidance, patients can be mostly treated in the outpatient setting.^[21]

Consensus guidelines for the surveillance of asymptomatic patients with cutaneous melanoma were published by the American Cancer Society in 2011.^[20] The expert panel recommends that follow-up be done at least once a year; however, the period should be shortened to 6 or 3 months if there is a family history of melanoma, multiple primary melanomas, atypical naevi, or new primary melanoma. Later in June 2013, medical documents were reviewed, and all data related to follow-up dermatological visits, follow-up dermatological examinations, and reports of skin biopsies or excisions were gathered. Following a study dermatologist’s assessment of each patient, the following recommendations for care were made: Two patients had short-term follow-up of lesions of questionable clinical significance; 48 patients had monthly skin self-examinations; and 20 patients had suspected malignant lesions excised. A total of 23 lesions were removed: two were removed from one patient, and one was removed from each of 17 individuals.^[22] For newly released patients in need of follow-up care, TD may be a crucial tool for ensuring accessible, adaptable, and convenient care^[23] [Figure 4].

Close

Figure 4:

Management and follow-up aspects of teledermatology.

Export to PPT

New insights into TD

In the future, Teledermatology consultation (TC) is anticipated to be essential in the treatment of infectious and chronic diseases such as leprosy, melasma, sexually transmitted infections, and dermatological disorders linked to HIV/AIDS. More accessibility for patients in remote or resource-constrained locations, enhanced image sharing, and more seamless virtual consultations have all been made possible by information technology advancements. The use of TC in therapeutic settings has been further refined by traditional and comparative investigations. With a focus on patient safety and diagnostic dependability, Kanthraj gave a summary of proper case selection, available TD tools, and the “do’s and don’ts” of online consultations. Digital lesion counting was used by Shetty et al. to create an image-based acne scoring system, proving that these techniques can be trustworthy substitutes for in-person assessments of acne vulgaris.^[24] Similar to this, another study mentioned in Pangti et al., 2023 developed a system for rating tinea corporis that evaluated lesion area, erythema, and border clarity.^[25] This allowed for precise follow-up therapy through virtual consultations that were on par with in-person evaluations. In addition, TC was quite helpful during the epidemic for inpatient treatment and for keeping track of elderly and chronic patients. Pangti and Gupta emphasized the value of TD in hospital settings and called for the expansion of virtual dermatology services across the country.^[25] In another instance, Anisha et al. described how teleconsultation through phone calls and mobile apps was used to successfully manage a senior patient with contact dermatitis and several comorbidities.^[26] According to their findings, TD can help disadvantaged populations by lowering hospital visits, guaranteeing continuity of care, and preserving patient satisfaction even under trying conditions. Bhatia et al., showed how to efficiently employ popular mobile applications such as WhatsApp for taking and sending dermatological photographs.^[27] Since then, the Indian National Medical Commission has created official rules covering important topics such as patient privacy, confidentiality, medical ethics, and reimbursement practices.^[27] Collectively, these advancements demonstrate how TD is becoming a crucial, standardized, and safe part of contemporary dermatological treatment, guaranteeing its sustained growth after the pandemic.