Skin cancers can be categorized into melanoma and nonmelanoma skin cancers (NMSCs) and are the most frequent types of cancer in Denmark and globally [1,2]. Recent data have shown that NMSC now results in a higher total number of deaths worldwide compared to malignant melanoma (MM), although the individual mortality risk is still higher for MM [3]. For the past decade, there has been a continuous increase in the number of NMSC and MM cases in Denmark [4]. This places a greater burden on a health care system already under pressure from an ageing population, a shortage of health care workers, and the continuous introduction of new, costly treatments [5,6].

In Denmark, primary care providers (PCPs) serve as gatekeepers for all dermatological concerns including skin cancer suspicions. PCPs encounter a broad variety of skin diseases, and studies have suggested that over half of the potentially malignant skin lesions referred to specialists are later diagnosed as benign [7-9]. This could indicate the need for an additional filter function, such as teledermoscopy, to better use the time and resources of health care workers and patients.

Teledermoscopy is a technology within the field of teledermatology where dermoscopic images of skin lesions can be referred to a dermatologist for evaluation [10]. The overall diagnostic accuracy of teledermatology is high, and results from recent studies indicate that accuracy and sensitivity are improving with the emergence of new technology [11]. Some of the advantages of teledermatology include significantly shortened waiting periods for patients and a reduction in the number of referrals for in-person consultations [11-13]. In addition, studies have found that PCPs, dermatologists, and patients have a general positive approach to the implementation of teledermoscopy [14,15]. Still, a few barriers exist when it comes to the implementation of teledermatology in health care systems. Among other things, physicians have reported concerns about lack of clinical information, technical issues, increased workload, and time consumption [13,16].

Only a few studies have measured the time required for consultations using teledermoscopy or teledermatology, and most only report time spent on completing or evaluating telereferrals as additional data points [12,17-19]. Since the area of teledermoscopy is constantly evolving, we wanted to look into a Danish-developed teledermoscopy platform, in order to see the impact and potential gain of implementation for physicians in terms of time consumption.

The overall purpose of this study was to investigate whether teledermoscopy can shorten the diagnostic process for physicians when examining skin lesions suspected of skin cancer.

First, we estimated the amount of time spent investigating these lesions with and without teledermoscopy for PCPs and dermatologists. Then time measurements were performed without teledermoscopy for plastic surgeons and pathologists.

Second, in order to compare the total amount of time physicians spend triaging, diagnosing, and treating skin cancer with and without teledermoscopy, we aimed to construct different diagnostic pathways, reflecting the most common ways through the Danish health care system. Time spent by patients and other staff was not included.

This study did not interfere with the patients’ potential treatment for skin cancer. It was conducted as a quality assurance project, which does not need an ethical review by the regional or national research ethics committee [20]. This study was approved by the Danish Data Protection Agency (reference 23/19285).

Time recordings were carried out at multiple locations in the Region of Southern Denmark and the Capitol Region of Denmark. Locations included 2 different primary care centers, 2 different private practice dermatologists, the department of pathology at Herlev Hospital, and the departments of dermatology, plastic surgery, and pathology at Odense University Hospital (OUH).

Before data collection, staff members from the primary care centers, dermatologist clinics, and the department of plastic surgery had reviewed their schedules in order to trace patients with potential skin cancer. During the designated data collection period, the investigator (RNL) performed the time recordings on days when these patients were scheduled for examination or removal of the skin lesion. We randomly picked the days for data collection at the department of pathology. The dermatopathologists selected the specimens with potential to be skin cancer.

We intended to collect a minimum of 10 recordings at primary care centers and dermatologists with and without teledermoscopy. Furthermore, we aimed to gather a minimum of 10 recordings at the department of plastic surgery and the department of pathology without teledermoscopy. Data were stored in Microsoft SharePoint.

At primary care centers, the eligibility criteria included patients presenting at their PCP with concerns about potential skin cancer, regardless of the PCP’s final assessment as malignant or benign. In dermatologist clinics, the eligibility criteria included patients referred by PCPs for suspected skin cancer. At the department of plastic surgery, the eligibility criteria included patients referred by PCPs or dermatologists for skin tumor excision. At the department of pathology, the eligibility criteria included patients with skin lesions suspected of skin cancer, referred by either PCPs, dermatologists, or plastic surgeons.

At primary care centers, the eligibility criteria excluded patients presenting with inflammatory skin diseases or follow-up of previous examined benign lesions. In dermatologist clinics, the eligibility criteria excluded physical referrals of patients with more than one skin tumor. At the department of plastic surgery, the eligibility criteria excluded re-excisions and excisions with skin grafts. At the department of pathology, the eligibility criteria excluded re-excisions and biopsy-verified lesions.

In this study, teledermoscopy was performed using the teledermoscopy platform Dermloop (Melatech ApS).

In this study setup, where time and data collection resources were limited and skin cancer consultations at PCPs appeared sporadically, it was a major challenge to identify relevant cases and follow them at every step of the diagnostic process. Thus, it was not possible to measure the total time physicians spend examining potential skin cancer for each patient included.

Consequently, we constructed a variety of diagnostic pathways that aimed to reflect real diagnostic processes. For pathways without the use of teledermoscopy, this was done in accordance with guidelines for physicians regarding referral criteria for lesions suspected of skin cancer, published by the Danish Health Authority [22] and the Region of Southern Denmark [23], where so-called cancer patient pathways (CPPs) are described for both MM and NMSC. Pathways with the use of teledermoscopy were constructed based on expertise from a consultant dermatologist (see Figure 1). It must be emphasized that the constructed pathways are simplified and does not cover all possible scenarios. The total amount of time physicians spend examining skin lesions in each of the constructed pathways was calculated by adding together the median amount of time PCPs, dermatologists, plastic surgeons, and pathologists spend on the individual parts of the process.

Stata/BE 18.0 was used to calculate all statistical analyses. For each individual part of the process, the Mann-Whitney U test was used to compare time consumption with and without teledermoscopy. P<.05 was considered significant. Data are presented as median (IQR). All time statements are presented in minutes (min) and seconds (s), that is, min:s. For each of the constructed diagnostic pathways (see Figure 1), the calculated median for every individual part of the pathway was added together (see Figure 2). The absolute difference between the constructed pathways with and without teledermoscopy was calculated.

Data were collected from April 2023 to December 2023. The dataset comprised 118 time recordings and included 17 physicians examining lesions suspected of skin cancer. When asked if it was acceptable for us to perform time recordings during examinations or treatment for potential skin cancer, no patients declined.

Upon evaluation of the data collected, a total of 24 recordings were excluded due to data not fulfilling the inclusion criteria or to various interruptions or errors during the time measurements. The exact number of time recordings excluded at each department was 5, 3, 8, and 8 at the PCP, dermatologists, department of plastic surgery, and department of pathology, respectively. Thus, a total of 94 time recordings were included in this study.

Tasks included in the procedure at the department of plastic surgery, dermatology, and pathology were repeated per lesion. Consequently, time recordings including more than 1 lesion were divided, resulting in each time recording comprising 1 skin lesion.

In order to make time recordings as uniform as possible, the majority of measurements were carried out by the same person (RNL). Due to time constraints, the physician (TV) at the department of dermatology at OUH made some of the recordings on her own (6.8%), following instructions.

Table 1 shows the data for the individual parts of the diagnostic processes with and without teledermoscopy for physicians. PCPs spend significantly longer time with teledermoscopy, whereas the diagnostic process was significantly shortened for dermatologists. The absolute median differences with and without teledermoscopy was 05:09 for PCPs and 01:06 for dermatologists.

Figure 2 shows the total amount of time physicians consumed at each of the constructed diagnostic pathways. Teledermoscopy shortened diagnostic pathway C with 11:31, whereas pathways A, B, D, and E were prolonged with 04:03, 07:02, 07:02, and 04:03, respectively.

Despite the use of different types of microscopes at the two departments of pathology involved, no significant difference between the time recordings at Herlev Hospital and OUH was found (P=.27).

In Denmark, 400.000 dermatological referrals are made per year, out of which about 30% (120,000) deal with suspicion of skin cancer [24,25]. Assuming 75% (90,000) of these referrals are benign and 25% (30,000) are NMSC [7], the amount of time physicians spend in pathway A and B with teledermoscopy can be estimated as shown in Table 2.

The most recent report from the Danish Health Authority regarding CPPs for skin lesions suspected of MM stated that in 58% of the cases, the suspicion was dismissed at some point in the diagnostic pathway [9]. The number of patients diagnosed with MM in 2022 was 3038 [4], which corresponds to a total of 7233 CPPs for suspected MM, with 4195 (58%) dismissed cases in 2022 [9]. The amount of time physicians spend in pathway C, D and E with teledermoscopy can be estimated as displayed in Table 2.

The use of teledermoscopy by PCPs prolonged the diagnostic process, whereas the process was significantly shortened for dermatologists. The constructed pathway C (benign lesion) was shortened for physicians with teledermoscopy, while the diagnostic pathways A (benign lesion), B (NMSC), D (MM), and E (MM) were prolonged by 16%‐58%.

This study has several limitations. First, due to limited resources and different geographic locations, it was not possible to follow each patient throughout the diagnostic process. This resulted in fragmented measurements of physician’s time consumption, involving independent groups of patients. Second, identifying relevant cases was difficult, especially at the PCPs, because patients have a tendency to bring up skin problems as supplementary subjects during consultations. Consequently, we were unable to achieve the intended number of patients and most time recordings were performed on patients with benign skin tumors. Third, our data are very heterogeneous due to the nature of the different departments included and the large spectrum of potential diagnoses. Fourth, patients with multiple potential skin cancers were not included in this study. Finally, this study was mainly based on structures in the Danish health care system. Some of our results might not apply directly to other health care systems.

This is one of few studies measuring clinicians’ time, as suggested by a recent review [30]. A strength of the study is that all time recordings were obtained in a real work setting, which gives the results high external validity. Furthermore, the same person performed the majority of the recordings.

More time recordings would be required to obtain a more accurate picture of physicians’ time consumption, for example, by including more departments and primary care centers, increasing the number of data collectors, or involving medical staff at the departments to a greater extent. In addition, it would increase data transparency if time stamps stating how much time PCPs and pathologists spend on each task were registered during the recordings. Altogether, several modifications to this pilot study’s design would be required to make it feasible for a larger-scale investigation.

Teledermoscopy significantly shortened the diagnostic process for dermatologists but prolonged it for PCPs. Previous studies have concluded that teledermoscopy reduces the number of physical referrals as well as surgical procedures, and our time study indicates that this would altogether result in time saving for dermatologists, plastic surgeons, and pathologists. Part of the time savings attained might be reallocated to other areas of the health care system, such as consultations at PCPs.