The prevalence of skin lesions is high in the general population, and they have come to figure among the main reasons for consulting primary care physicians.1,2 Many of these lesions are benign, having no impact on the individual's health, and care is usually sought for purely aesthetic reasons.3 Spanish Royal Decree 1030/2006 of September 15 excludes procedures and treatments that have only cosmetic outcomes from the list of services covered by the public National Health Service (NHS).4 Dermatology-related visits for minor lesions nevertheless accounted for up to 25% of primary care consultations in one Spanish study.3 The primary care family practitioner currently provides the first line of care in the NHS and is the one who decides whether to refer a patient to a dermatologist or not.5,6 The Andalusian section of the Spanish Academy of Dermatology and Venereology (AEDV), working with the Spanish Society for Quality in Health Care (SECA), recently published a consensus paper recommending a care process for persons with cystic lesions or benign tumors.7

The present study aimed to estimate the percentage of primary care referrals to dermatologists for cystic lesions or benign tumors that might be avoided. The results should provide real data for assessing the magnitude of the problem nationwide and for drafting health care policies.

Data were taken from the AEDV's DIADERM study of 2016,8 which surveyed a representative sample of dermatologists practicing in Spain. The 2224 members of the AEDV in November 2014 were stratified according to geographic area and sampled randomly so that the respondents would represent dermatologists across the country. The surveyed dermatologists reported all diagnoses made over a period of 6 days (3 days during cold weather and 3 days during hot weather). An expert coded the diagnoses according to version 10 of the International Classification of Diseases. Each patient visit was associated with a single principal diagnosis.

We extracted the diagnoses that were the main reason for a referral to dermatology by a primary care physician in the NHS. Teledermatology consultations were excluded.

A referral was classified as potentially avoidable if the diagnosis was one of those that could have been discharged from primary care, as established by the 2016 consensus paper of the Andalusian section of the AEDV (Table 1).7 Such cases can be discharged from primary care provided malignancy is not suspected and there is no doubt about the diagnosis, no functional impact, and no history of resistance to treatment in the case of viral infections. Referrals were not classified as potentially avoidable if they met appropriate referral criteria, such as would be the case for an atypical nevus.

A record of the primary care physician's justification for the referral was not available. An indirect indicator of a referral's appropriateness was the decision taken by the dermatologist during the first visit. We assumed that patients who were discharged on the first dermatology visit underwent no dermatologic procedure; if the dermatologist had intervened, there would have been an order for pathology and/or follow-up observation of a scar. The patients with a main diagnosis included in the consensus paper's list and who were discharged by the dermatologist after the first visit might have been referred because of a primary care physician's reasonable doubt that was resolved by the dermatologist; alternatively, their cases might consitute potentially avoidable referrals. On the other hand, if the dermatologist did not discharge the patient, the referral might also have represented either an appropriate one or poor use of resources by the dermatologist.

The DIADERM study8 was classified as observational non-postauthorization research by the Spanish Agency of Medicines and Healthcare Products (AEMP) and approved by the research ethics committee of the province of Granada on October 8, 2014.

Descriptive statistics (absolute frequency and percentage) were compiled and the χ2 test was used to compare qualitative variables using Stata statistical software (version 14.2, Statacorp, Texas, USA).

The DIADERM dataset included 10999 diagnosed cases in 8953 patients. Once we had excluded the teledermatolgy cases, the secondary reasons for consulting a physician, and the cases registered by physicians outside the NHS, there were 5089 diagnoses registered as the main reasons patients consulted a physician available for analysis. Of these, 2171 (42%) were first registered from a primary care clinic of the NHS (Fig. 1). The 2918 remaining diagnoses were excluded because they involved follow-up visits, emergency visits, or referrals from other specialists.

Figure 1.

Flow chart of case selection.

(0.15MB).

Six hundred eighty-six of the 2171 cases (31.6%) were classified as potentially avoidable referrals; 478 of them (22%) were discharged after the first specialist visit (Table 1). The percentage of cases discharged by the dermatologist on the first visit was significantly higher in the group of 686 potentially avoidable referrals than in the others (69.7% vs 39.1%, respectively; P<.001, χ2 test). The percentages of potentially avoidable referrals (Table 2) were similar in all the geographic areas of Spain (P=.74, χ2 test).

The largest group was made up of potentially avoidable referrals for seborrheic keratosis, which accounted for 201 visits (9.3%). Seborrheic keratosis was also the discharge diagnosis reached most often by dermatologists, who recorded it after 180 out of the 201 first visits (89.6%) that named this condition as the main reason for consulting the specialist. Referrals for melanocytic nevi and the group of lesions comprised of common or plantar warts or molluscum contagiosum infections accounted for 8.6% and 6.2% of the referrals, respectively. However, the percentages of first-visit discharges by dermatologists were somewhat lower for these 2 referral diagnoses, at 63% in both cases (118 out of 187 and 85 out of 135 referrals, respectively).

This study found that approximately 32% of referrals from primary care to dermatology in the Spanish NHS concerned the conditions included in the Andalusian consensus paper on cystic lesions or benign tumors7 and might therefore be classified as avoidable referrals in the absence of diagnostic doubt or functional impact. Seborrheic keratosis, melanocytic nevus, the group comprising common or plantar warts and molluscum contagiosum, and acrochordons and cysts account for 28.6% of the skin conditions referred from primary care to dermatology. As expected, a significantly higher proportion of visits for these conditions were discharged by the dermatologist after the first visit, suggesting that they may have been potentially avoidable. Such referrals for minor skin problems consume human and material resources but provide scant added value in terms of the population's health and may even violate current legislation.5

The consensus statement of the Andalusian section of the AEDV7 and the model for the practice of dermatology issued by the department of health of the government of Catalonia (Departament de Salut, Generalitat de Catalunya)9 are examples of efforts intended to increase the ability of primary care physicians to resolve benign lesions and skin conditions that are not complex in order to improve efficiency.

The percentage of dermatology referrals from primary care that our study identified as potentially avoidable was similar to the percentage of requests for treatment for aesthetic purposes in a study by Macaya-Pascual et al3 in an outpatient clinic of the NHS in L’Hospitalet de Llobregat, near Barcelona. Similarly, a study by Lowell et al2 reported that 29.4% of referrals to a dermatologist were for benign tumors.

Avoidable referrals in large numbers put pressure on the routine practice of dermatology, taking time that could be dedicated to other skin diseases that have greater impact on health.5 According to the previously cited study by Macaya-Pascual et al,3 the diagnosis and treatment of these conditions in a single dermatology practice in a Spanish outpatient clinic that attends 7500 visits each year would cost some €317000 per year. This figure points to a need for strategies to make the system more efficient nationwide by applying consensus-based recommendations, such as are outlined in the Andalusian statement,7 and by improving contact with primary care physicians through teledermatology.10 This approach, which is already in place in many centers, offers an effective way to manage communication between care levels and would probably help to reduce the number of potentially avoidable referrals.11

Our study's main strength is that it is based on the DIADERM registry, which reflects a stratified random selection of cases from dermatologists working in all the territories where members of the AEDV practice.8 The information, therefore, is a representative reflection of dermatology practice throughout Spain.

An important limitation of the study is that we have no information about the circumstances justifying the primary care physicians’ referrals, which would be appropriate in the presence of diagnostic doubt or functional impact. Therefore we can only speculate on the maximum impact of applying the recommendations of the Andalusian consensus statement.7 Cases discharged because they were resolved during the first visit to the dermatologist may have involved either avoidable referrals or appropriate ones (because of the primary care physicians’ doubts). Cases in which the dermatologist scheduled a second visit or a surgical procedure might also have been either appropriate or inappropriate use of resources. Thus, it is possible that the proportion of truly avoidable referrals, according to the definitions of the consensus statement,7 might be as low as 0% or as high as 31.6% of first visits to the dermatologist. More likely, however, the real percentage lies between the extremes. Another limitation derives from our study's focus on cystic lesions and benign tumors (including common warts, molluscum contagiosum, and onychogryphosis as listed by the Andalusian recommendations7). If a similar model were applied to all the reasons for referral, including infectious and inflammatory conditions, the potential impact could be greater. Finally, although the Andalusian section of the AEDV's consensus statement was drafted with the collaboration of the SECA and knowledgeable, clinically experienced dermatologists, it must be admitted that neither primary care physicians nor patients were consulted. Nonetheless, administrators of public agencies, who are charged with representing patients’ needs, did participate. In any case, we believe that future consensus statements should involve both the aforementioned stakeholders in the care process.12

In conclusion, this is the first study to analyze potentially avoidable referrals for cystic lesions and benign tumors from primary care to dermatology throughout Spain. The findings show that these referrals account for a high proportion of the dermatology caseload and that the most common diagnoses are seborrheic keratosis, melanocytic nevus, and the group comprising warts and molluscum contagiosum. We have provided information that can be useful for health care planning and training to optimize available resources as well as for designing future studies. Conclusions should take into consideration the limitations discussed above, the real possibilities of the health care situation of Spain, and the need for continuity between primary and specialist care.

The DIADERM study, which is supported by the Healthy Skin Foundation of the Spanish Academy of Dermatology and Venereology (AEDV), received funding from Novartis. This pharmaceutical laboratory, however, did not participate in gathering or analyzing data, interpreting results, or drafting the final report.

The authors declare that they have no conflicts of interest.