Although most routine dermatology services in Spain are provided in outpatient clinics, data on the diagnoses reached by dermatologists in these settings are lacking. Studies carried out by dermatologists tend to be local, retrospective, focused on a single disorder, or conducted in other settings such as primary health care facilities.1–5

A comparative study of common reasons for seeking care at dermatology services in Spanish and immigrant patients in Palma de Majorca found that benign tumors were the most frequent reason, followed by inflammatory conditions and viral skin diseases, with a total of 213 diagnoses reached during the 1-year study period.3 Another study, carried out in Granada in patients over 65 years of age, found that basal cell carcinoma was the most frequent diagnosis, followed by actinic keratosis.2

A recent analysis of the burden of skin disease in the United States found that 26.9% of the population was seen by a physician for a skin condition in 2013, generating high direct and indirect costs, comparable to those associated with conditions such as diabetes or cardiovascular disease.6 In view of the large number of patients seen in dermatology clinics each year, accurate knowledge about the most prevalent diseases would make it possible to analyze the management of these diseases and the costs associated with their treatment, and to distribute the available resources more efficiently in accordance with demand.

The diagnoses made by dermatologists in Spain have not previously been described by a prospective study of a representative sample of the country's dermatologists. The aim of this study was to estimate the prevalence of the diagnoses reached by members of the Spanish Academy of Dermatology and Venereology (AEDV) at dermatology clinics in Spain as well as the volume of activity.

Prevalence data were obtained in the DIADERM study through an anonymous survey of a representative random sample of dermatologists. The project was coordinated by the AEDV's Spanish Group for Epidemiology and Health Promotion in Dermatology, in collaboration with the academy's research unit.

We selected a random sample of dermatologists drawn from the AEDV list (which consisted of 2224 dermatologists as of November 2015) and stratified by geographic area (the AEDV is organized in sections that correspond approximately to Spain's autonomous communities, although the smallest communities are grouped together). The probability of selection was proportional to the number of dermatologists in each geographic area.

To maximize the response rate, the study was advertised through the AEDV's communication channels. All selected dermatologists were personally contacted by the study's regional coordinators and were offered upfront compensation for their participation.7 Several reminders were also sent before the starting date of the study. After they agreed to participate, the selected dermatologists received instructions on how to collect the data required. Each participant recorded all diagnoses reached in the course of their clinical practice on 6 specified days (January 19, 20, and 21 and May 18, 19, and 20).

To estimate percentages of 5 per 1000 with a precision of ±2 per 1000 and a 95% CI using a simple random sample, data from 4778 consultations would be needed (assuming 1 diagnosis per patient). Since consultations were grouped by dermatologist and a certain degree of diagnostic correlation was possible, it was important to consider the sampling method used (design effect). A design effect of approximately 2 was assumed.

An earlier publication estimated that 97% of dermatologists in Spain are AEDV members and that the mean number of daily consultations per dermatologist is 23.6.8 If a dermatologist handles approximately 23 consultations per day, the total number of patients seen in 6 days is 138. Therefore, 35 dermatologists would be needed in order to reach an adequate estimate using a simple random sample. Taking the design effect into account, a total of 70 dermatologists would be needed. Assuming a 60% response rate and with corrections for each geographic area, we decided to contact 124 dermatologists.

At each consultation, the dermatologist collected the following data: diagnosis (free text), whether or not the patient was under 18 years of age, whether or not the consultation was done by teledermatology, reason for consultation, and any secondary diagnoses. If multiple diagnoses were established for a single patient, these diagnoses were marked as grouped. Other variables considered were whether the consultation took place in a public or private health care setting (a consultation was considered to have taken place in a public setting if the cost of medication was covered). Since many dermatologists work in both public and private settings, this aspect was associated not with the dermatologist but with the consultation. The origin of the visit—direct access to the dermatologist without a filter appointment, referral from primary care, referral from another specialist, or follow-up visit—was recorded. The patient destination was also recorded: discharge, follow-up with the dermatologist (including surgery), or follow-up with another physician. Neither personal data nor information about treatments were collected. Diagnosis was the unit of analysis. The dermatologists recorded diagnoses that corresponded to the reason for seeking care as well as any other diagnoses for which treatment—either medical or surgical—was prescribed. The survey was anonymous and no information was collected about the characteristics of the participating dermatologists or about their hospitals or clinics.

The diagnoses were coded by a dermatologist (AML) expert in applying the International Classification of Diseases, 10th revision (ICD-10). Textual descriptions of the diagnoses recorded in the survey and the corresponding codes were stored in a database. Quality control of the data coding was performed using a random sample of surveys. All diagnoses about which doubts existed were reviewed by 3 dermatologists with a particular interest in coding (YG, RT, and IGD), who belonged to the AEDV E-Dermatology and Imaging Group.

The statistical analysis took into account the data collection design and was carried out using the survey module of the Stata software package,9 which incorporates standard error correction for correlated data. A finite population correction was not necessary because the real number of dermatologists yielded values close to 1 for each AEDV section.

The study was classified by the Spanish Agency of Medicines and Medical Devices as a non-postauthorization study and was approved by the research ethics committee of Granada province (October 8, 2014).

Of the 124 randomly selected dermatologists, 80 (65%) participated in the first phase of the study and 73 (59%) in the second.

Over the 6-day study period, 11 223 diagnoses were recorded, but 224 were excluded because they were procedures and therefore could not be coded. A total of 10 999 diagnoses in 8953 patients were ultimately obtained, representing an estimated total of 208 141 diagnoses reached in 169 517 patients by all AEDV dermatologists. Hereinafter we will refer to the estimated data based on the total number of AEDV dermatologists.

As for the data coding quality control, a minimal error percentage (0.16%) was found in a sample of 621 diagnoses (5% of the total). In addition, all 764 diagnoses about which coding-related doubts existed (7% of the total) and an additional random sample of 1231 diagnoses (11% of the total) were reviewed by a group of dermatologists with a particular interest in coding, who belonged to the AEDV E-Dermatology and Imaging Group. Errors were found in 79 (4%) of these 1995 diagnoses.

We estimate that more than 600 000 patients are seen by AEDV dermatologists each month in Spain, with an average of 28 patients seen per day per clinic. The most frequent diagnosis was actinic keratosis, followed by basal cell carcinoma and melanocytic nevus. The vast majority of visits took place at clinics and less than 1% of patients were assessed using teledermatology. In one third of cases, the consultation took place in a private health care setting and the dermatologist was the patient's origin as well as destination.

Our findings show that a large number of patients are seen by AEDV dermatologists in Spain each month. Our figures are somewhat higher than those published by the Spanish Ministry of Health in 2014 (the most recent data on real activity currently available),10 which indicate an annual total of 5 177 878 dermatology consultations in public and private health care facilities (hospitals and outpatient clinics). According to the ministry's report, 910 624 consultations (17.5%) took place in private health care settings, a lower percentage than that found in our study (31.7%). One possible explanation for this difference is that some private dermatology clinics are not included in the ministry's information collection systems. In fact, the ministry's statistics on specialized care centers are based on information provided by the centers listed in the national catalogue of hospitals, which omits many private dermatology clinics. According to the ministry's data, dermatology consultations accounted for 4.7% of all consultations, behind higher-volume specialties such as ophthalmology, traumatology, and gynecology but ahead of all others.10 The mean number of daily consultations in our study—28.2 patients per day per dermatologist—was higher than the figure of 23.6 obtained in a previous study.8

The most frequent diagnosis in the DIADERM study was actinic keratosis, accounting for an estimated 8.2% of diagnoses, followed by basal cell carcinoma (8.1%) and melanocytic nevi (7.5%). In a multicenter survey conducted in 2013 and 2014, the prevalence of actinic keratosis in the dermatology outpatient population over the age of 45 years was 28.6%.11 In one local study conducted in Majorca, the most frequent diagnosis was melanocytic nevi, followed by seborrheic keratosis and verruca vulgaris.3 In a study of patients from Granada over 65 years of age, the most frequent diagnosis was basal cell carcinoma, followed by actinic keratosis and seborrheic keratosis.2 In a 2006 study carried out in Barcelona, the most frequent diagnoses were seborrheic keratosis and melanocytic nevi.12

In all these studies, tumors or premalignant disease were the most frequent reasons for seeking care. This may be due to the growing incidence of skin cancer13 or to patients’ motivation to obtain an early diagnosis as a consequence of the AEDV's awareness-raising campaigns.14 Probably as a result of these secondary prevention strategies, we observed that diagnoses of actinic keratosis and basal cell carcinoma were more frequent in follow-up patients than in first-time patients.

In public health care settings, the most common origin for patients seen by a dermatologist was a referral by a primary care physician or by another specialist (accounting for 47.8% of visits); in private settings, the most common origin was personal initiative (74.7%). In both public and private health care settings, follow-up was the patient destination in more than 60% of cases. Given that single-visit care is becoming increasingly common—especially in public settings—this percentage is rather high.

This study had various limitations. First, although the participation rate (greater than 60%) was quite high for surveys of this type, the number of dermatologists taking part decreased in the second phase of the study. Because the sample was random and representative and a total of 10999 diagnoses were obtained, we consider that our findings reflect the reality of clinical practice in Spain quite well, despite the fact that data were collected over just 6 days. Second, surveys tend to have a certain degree of coding error, especially if double data entry is not used. In our study, however, a group of experts conducted a review and found that the data quality was good, with an error rate of less than 5%. Finally, personal details such as age and sex were not collected in our survey. Although these details would have provided interesting information, their collection would made the study procedure more complicated.

In conclusion, this is the first nationwide study in Spain to analyze the diagnoses reached in dermatology clinics by AEDV members. Our findings show a high volume of activity, with the most frequent diagnoses being actinic keratosis, basal cell carcinoma, and melanocytic nevi. This information will be useful for guiding health care planning, resource use, and future studies.

The DIADERM study was promoted by the AEDV's Fundación Piel Sana, which has received financial assistance from Novartis. The pharmaceutical company did not participate in data collection, data analysis, or the interpretation of the results.

The authors declare that they have no conflicts of interest.