This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Dermatology, is properly cited. The complete bibliographic information, a link to the original publication on http://derma.jmir.org, as well as this copyright and license information must be included.
Challenges remain for general practitioners (GPs) in diagnosing (pre)malignant and benign skin lesions. Teledermoscopy (TDsc) supports GPs in diagnosing these skin lesions guided by teledermatologists' (TDs) diagnosis and advice and prevents unnecessary referrals to dermatology care. However, the impact of the availability of TDsc on GPs’ self-reported referral decisions to dermatology care before and after the TDsc consultation is unknown.
The objective of this study is to assess and compare the initial self-reported referral decisions of GPs before TDsc versus their final self-reported referral decisions after TDsc for skin lesions diagnosed by the TD as (pre)malignant or benign.
TDsc consultations requested by GPs in daily practice between July 2015 and June 2020 with a TD assessment and diagnosis were extracted from a nationwide Dutch telemedicine database. Based on GP self-administered questions, the GPs’ referral decisions before and their final referral decision after TDsc consultation were assessed for (pre)malignant and benign TD diagnoses.
GP self-administered questions and TD diagnoses were evaluated for 6364 TDsc consultations (9.3% malignant, 8.8% premalignant, and 81.9% benign skin lesions). In half of the TDsc consultations, GPs adjusted their initial referral decision after TD advice and TD diagnosis. Initially, GPs did not have the intention to refer 67 (56.8%) of 118 patients with a malignant TD diagnosis and 26 (16.0%) of 162 patients with a premalignant TD diagnosis but then decided to refer these patients after the TDsc consultation. Furthermore, GPs adjusted their decision from referral to nonreferral for 2534 (74.9%) benign skin lesions (including 676 seborrheic keratosis and 131 vascular lesions).
GPs adjusted their referral decision in 52% (n=3306) of the TDsc consultations after the TD assessment. The availability of TDsc is thus of added value and assists GPs in their (non)referral for patients with skin lesions to dermatology care. TDsc resulted in referrals of patients with (pre)malignant skin lesions that GPs would not have referred directly to the dermatologist. TDsc also led to a reduction of unnecessary referrals of patients with low complex benign skin lesions (eg, seborrheic keratosis and vascular lesions).
In the Netherlands, patients that are concerned about their skin lesion visit their general practitioner (GP) for advice. GPs assess the skin lesions and decide if a wait-and-see policy is justified, if they can manage the skin condition themselves in their practice, or if the patient should be referred to a dermatologist. In this way, GPs serve as gatekeepers and play a key role in deciding whether a patient is referred to Dutch dermatology care. However, GPs seem to find distinguishing between benign and malignant skin lesions a difficult task [
In general, previous TDsc evaluation studies in primary care settings included all eligible patients with suspicious (pigmented) skin lesions, patients who GPs regularly intend to refer, or patients who were already referred to a hospital or lesion clinic [
In the Netherlands, TDsc has been integrated into GP practices nationwide since 2009 by a Dutch telemedicine provider (Ksyos) and is fully reimbursed by Dutch health insurance companies [
Previous TDsc studies in other settings investigated common TD-provided telediagnoses and the percentage of patients for whom, due to TDsc, a physical referral to a dermatologist could be avoided [
Therefore, for these diagnosis groups, the impact of the availability of TDsc on the GPs’ referral decisions to dermatology care is still unknown. Therefore, this study assessed and compared GPs’ self-reported initial referral decisions before TDsc with their final referral decisions after TDsc for (pre)malignant and benign TD-diagnosed skin lesions.
In the Ksyos-secured TDsc digital health record system, a GP starts the TDsc process with a standardized consultation request and uploads the obtained (detailed, overview, dermoscopic) images of a patient’s skin lesion. After a GP has filled in patient information, such as year of birth, sex, prehistory of skin cancer, structured anamnesis, optional provisional diagnosis, and additional notes, the GP sends the TDsc request to a TD for review. The TD then provides a primary diagnosis (a mandatory and an optional differential diagnosis) in a text entry field and referral recommendations, which may include advice for the GP on the patient management plan.
Further, a GP is asked to answer 2 similar nonmandatory self-administered questions: (1) “Would you have referred this patient if TDsc was not available?” and (2) “Are you still referring this patient to the dermatologist?”. These questions, which are embedded in the Ksyos system by default, retrieve information about (1) the GP’s initial decision to refer a patient to a dermatologist (Yes, No) when sending the TDsc consultation request to a TD and (2) the GP’s final referral decision (Yes, No) at the time of closing the TDsc consultation after the TD assessment.
As of July 2015, the Ksyos system generates an ICD-10 (International Classification of Diseases, 10th revision) [
No ethical approval was required to evaluate the number of TDsc consultations, since all GPs gave permission through a contract with Ksyos to monitor TDsc quality with these self-administered questions.
For this retrospective database study, TDsc consultations requested by GPs between July 2015 and June 2020 were included in the data analysis. Next, consultations with missing values were excluded. Missing values in the database were defined as a TD report of “no diagnosis” (R69), “no abnormalities” (R68.8), or “nonassessable” (−), or if a GP had not answered both self-administered questions. Data acquired included (1) answers to the GP self-administered questions on referral of a patient to a dermatologist and (2) diagnosis provided by TD during the TDsc consultation. Optional differential diagnoses provided by the TD were omitted from this study. Types of cameras or digital dermoscope used to obtain the images were unknown.
The GP self-administered questions were used to define whether the GPs had or had not adjusted their initial decision to refer a patient to a dermatologist after reviewing the advice and diagnosis of the TD.
In this study, 3 diagnosis groups were defined based on the TD diagnoses and the corresponding ICD-10 codes: malignant, premalignant, and benign. The histopathology and face-to-face diagnoses were not available in our study. Malignant skin lesions included all malignant neoplasms (ICD-10 codes C00-C97) such as melanoma, basal cell carcinoma, and squamous cell carcinoma. Premalignant skin lesions were defined as a separate group and included in situ neoplasms (ICD-10 codes D00-D09), other specified epidermal thickening (ICD-10 code L85.8; eg, keratoacanthomas), and actinic keratosis (ICD-10 code L57.0). Benign skin lesions included the remaining ICD-10 diagnoses. In this group, we specifically focused on seborrheic keratosis (ICD-10 code L82) and vascular lesions (ICD-10 codes D18, I78.1). For each diagnosis group, the GP self-reported initial and final referral decisions were analyzed.
In total, 13,509 TDsc consultations requested by 1185 GPs between July 2015 and June 2020 were provided with a diagnosis by 140 TDs. Of these, 1770 (13.1%) were assessed by the TD as “no diagnosis,” 14 (0.1%) as “no abnormalities,” and 350 (2.6%) as “nonassessable.” Moreover, 5011 (44.1%) TDsc consultations had an absent response on the GP self-administered question(s) and were therefore excluded as a missing value from the data set (
Among the group of malignant diagnoses, the most common were basal and squamous cell carcinoma (n=415, 70.1%) followed by malignant melanoma (n=172, 29.1%). The most commonly provided diagnosis in the premalignant diagnosis group was actinic keratosis (ICD-10 code L57.0; n=434, 77.4%). Among the group of benign diagnoses, the most common was melanocytic nevus (ICD-10 code D22; n=2571, 49.3%), followed by seborrheic keratosis (n=1221, 23.4%).
Flowchart of teledermoscopy (TDsc) consultations requested by general practitioners (GPs) between July 2015 and June 2020 as included in our study. ICD-10: International Classification of Diseases, 10th revision; TD: teledermatologist.
In 3306 (51.9%) TDsc consultations, the GPs adjusted their referral decision (Yes-No, No-Yes) after the TD assessment (
In the premalignant diagnosis group, the GPs indicated that they would not have referred for 162 (28.9%) patients without TDsc. For 26 (16.0%) of these 162 patients with a premalignant TD diagnosis, the GPs changed their decision from nonreferral to referral.
In the benign diagnosis group, 3384 (64.9%) patients with benign skin lesions, of which 784 (64.2%) had seborrheic keratosis and 163 (70.6%) had vascular lesions, would have been referred by the GP without the availability of TDsc. The TD-provided benign diagnoses resulted in a change of the GPs’ decision from referral to nonreferral for 2534 (74.9%) patients. More specifically, GPs adjusted their referral decision to nonreferral after the TD assessment for 676 (86.2%) patients with a seborrheic keratosis TD diagnosis and 131 (80.4%) patients with a vascular lesion TD diagnosis. In addition, the group of “other benign skin lesions” included benign nevi as well as ICD-10 codes for eczema and insect bites.
Number of teledermatologists (TD) diagnoses for the general practitioner (GP) self-administered questions.
Self-administered questions | Malignant skin lesions (N=592), |
Premalignant skin lesions (N=561), |
Benign skin lesions, n (%) | Total TDsca consultations (N=6364), n (%) | ||||
|
|
|
Seborrheic keratosis (N=1221) | Vascular lesions |
Other benign skin lesions (N=3759) | Total benign skin lesions (N=5211) |
|
|
|
474 (80.1) | 399 (71.1) | 784 (64.2) | 163 (70.6) | 2437 (64.8) | 3384 (64.9) | 4257 (66.9) | |
|
Q2c=Yes | 353 (74.5) | 122 (30.6) | 108 (13.8) | 32 (19.6) | 710 (29.1) | 850 (25.1) | 1325 (31.1) |
|
Q2=No | 121 (25.5) | 277 (69.4) | 676 (86.2) | 131 (80.4) | 1727 (70.9) | 2534 (74.9) | 2932 (68.9) |
|
118 (19.9) | 162 (28.9) | 437 (35.8) | 68 (29.4) | 1322 (35.2) | 1827 (35.1) | 2107 (33.1) | |
|
Q2=Yes | 67 (56.8) | 26 (16.0) | 36 (8.2) | 9 (13.2) | 236 (17.9) | 281 (15.4) | 374 (17.8) |
|
Q2=No | 51 (43.2) | 136 (84.0) | 401 (91.8) | 59 (86.8) | 1086 (82.1) | 1546 (84.6) | 1733 (82.2) |
aTDsc: teledermoscopy.
bFirst GP self-administered question: Would you have referred this patient if TDsc was not available?
cSecond GP self-administered question: Are you still referring this patient to the dermatologist?
This retrospective study assessed the impact of the availability of TDsc on GPs’ self-reported decisions to refer patients to the dermatologist. GPs’ self-reported initial referral decisions before the TDsc consultation were compared with their referral decisions after the TDsc consultation for skin lesions diagnosed by the TD as (pre)malignant or benign. This study showed that for these lesions, GPs adjusted their initial referral decision after the TD assessment in half of the TDsc consultations.
For 26 (16%) of 162 patients with a premalignant TD diagnosis and for 67 (56.8%) of 118 patients with a malignant TD diagnosis, GPs adjusted their referral decision after the TDsc consultation from nonreferral to referral. Therefore, without the availability of TDsc, GPs would not have referred these patients with (pre)malignant TD diagnoses directly to the dermatologist.
Furthermore, if the TD provided the diagnosis seborrheic keratosis, GPs adjusted their referral decision in 676 (86.2%) of the 784 TDsc consultations from referral to nonreferral. For the TD diagnosis of vascular skin lesions, GPs adjusted their referral decision in 131 (80.4%) of the 163 TDsc consultations from referral to nonreferral. Therefore, without the availability of TDsc, GPs would have referred these patients with benign skin lesions to a dermatologist.
In a Belgian TDsc study, which included all patients with suspicious skin lesions for TDsc, regardless of whether the GPs intended to refer the patients, GPs photographed all skin lesions as part of the TDsc consultation [
In contrast to our study, a Danish and a Swedish TDsc study included only patients with suspicious skin lesions that the GPs, without the availability of TDsc, would have referred to the dermatologist [
In these 3 TDsc studies, all patients with suspicious skin lesions, along with patients that the GPs initially would have referred for a physical dermatological consultation, were included. By contrast, in our study, which was performed in daily general practice, GPs acted as gatekeepers to dermatology care. GPs decided themselves whether to apply TDsc, justify a wait-and-see policy, manage the skin condition themselves, or refer the patient to a dermatologist.
Previous findings show that TDsc is especially valuable for the triage of patients with benign skin lesions. The relatively fast TD assessment of skin lesions diagnosed as evidently benign reassures and avoids nervous waiting for both patients and practitioners [
Remarkably, the GPs in our study also applied TDsc to request TD advice concerning nonpigmented benign diagnoses, such as eczema, psoriasis, and insect bites, which is in accordance with 2 other TDsc studies in a virtual lesion clinic and primary health care center setting [
The TDsc service evaluated in our study is unique compared to other systems because it asks GPs to enter their initial referral decision at the start of the TDsc consultation request and their final referral decision after the TDsc consultation. In a retrospective TDsc study by our research group 5 years ago in the same nationwide context and with the same Dutch TDsc system, we found that the GPs adjusted their initial referral decision after TDsc in half of the consultations [
In an Italian study, GPs were also asked to assess photographed skin lesions and decide whether they would refer the patient to a dermatologist [
The strengths of this large retrospective study include that TDsc consultations were conducted in daily GP practice and were not simulated in a study setting. The GP referral decisions were noted both before and after the TDsc consultation, which allowed us to verify whether GPs adjusted their initial referral decisions after the TDsc consultation. In doing so, we gained insight into GP referral decisions for different diagnosis groups after the TD assessment in daily GP practice.
On the other hand, the first limitation of our study is that the TDs did not always report their diagnosis in the TDsc system and that we omitted data on the differential diagnosis. This might have resulted in an underestimation of the absolute number of (pre)malignant and benign diagnoses for which TDsc was applied by the GPs. It is possible that TDs were unable to provide a diagnosis because the GPs provided insufficient patient information in the TDsc consultation [
The second limitation of our study is that the GPs were not obliged to fill in the self-administered questions regarding their referral decisions; thus, these self-administered questions were not always filled in. For these TDsc consultations, we could not compare the GP referral decision before and after the TDsc consultation. In addition, we do not know if the GP interpreted these questions regarding their referral decision as originally intended in the TDsc system. The reasons why GPs decided not to physically refer patients with a TD-diagnosed (pre)malignant skin lesion are still unknown. Additionally, clinical follow-up data on these patients are lacking. The Dutch guideline for suspicious skin abnormalities recommends that GPs refer malignant skin lesions to the dermatologist [
The third limitation is that only the TDsc consultation data extracted from the Ksyos system were accessible for our study. Although Ksyos is the largest store-and-forward telemedicine provider in the Netherlands, the overall number of TDsc consultations in the Netherlands might be higher.
The fourth limitation is that no data concerning the histopathological diagnoses were available for our study. In practice, it is considered unethical to acquire, purely for research purposes, the histopathology of patients with benign skin lesions who not have been referred by the GP to the dermatologist (Q1=No and Q2=No; Q1=Yes and Q2=No). Vestergaard et al [
This study showed that GPs adjusted their initial referral decision of patients with skin lesions in half of the studied TDsc consultations after the TD assessment. The availability of TDsc remains thus of added value to support GPs in gatekeeper health care systems in their decision to refer patients to a dermatologist for an in-person consultation. This study has shown that GPs initially did not intend to refer patients with (pre)malignant skin lesions for an in-person dermatological consultation and that the availability of TDsc aids in the referral of these patients. In addition, TDsc supports GPs in the prevention of unnecessary physical referrals to the dermatologist for patients with low complex benign skin lesions (eg, seborrheic keratosis and vascular skin lesions), easing the burden on dermatology care.
general practitioner
International Classification of Diseases, 10th revision
teledermatologist
teledermoscopy
The authors would like to thank Dr JP van der Heijden for reviewing the manuscript.
ET, FvS, MWB, and LWP conceptualized the study. All authors were involved in the study design. ET and FvS were responsible for data acquisition. All authors were involved in the analysis and interpretation of data. ET, FvS, and LWP wrote the manuscript from the first version onward. MWB and MWJ were further involved in critical revision of the manuscript.
ET and FvS are PhD researchers at the Amsterdam University Medical Center (UMC) and employed by Ksyos.