An individual clinician's experience is an important source of knowledge in dermatology. However, when such knowledge becomes the sole basis for clinical decision-making, therapeutic effects are often overestimated. Compounding this problem is the physician's tendency to rely on knowledge acquired during residency training or to find it difficult to incorporate current evidence into routine practice, especially if new information calls beliefs and previous experience into question.1–3 Clinical trials follow an experimental design in which a researcher manipulates exposure to 1 or more treatments in order to compare effects.4,5 The main purpose of this type of study is to assess the efficacy and safety of an intervention that seeks to prevent or cure a health condition or to speed recovery.4,5

Given the importance of randomized clinical trials (RCTs), it might be supposed that they would be easily available to both treating physicians and researchers. Problems arise, however, when health care professionals seek to locate and use information from RCTs. Among the difficulties that have been reported are 1) the novelty of the terminology itself, 2) the underuse of descriptors when trials are indexed in databases, and 3) the high percentage of journals that do not post articles online.6–8 Problems that further interfere with physicians’ use of RCTs are lack of time for reading these articles and the lack of access to the publishing journals.9,10

To help identify RCTs published in the Spanish language in several medical specialties, the Cochrane Collaboration undertook a project to search for them manually. Searching in databases alone reportedly fails to find a significant number of RCTs in the specialties of ophthalmology, public health, anesthesiology and critical care, and general and internal medicine.8,11–14 In addition, online MEDLINE searches can fail to return up to 25% of RCTs available, mainly when the authors have not included the search terms randomized controlled trial or controlled clinical trial in the titles.15

We present the results for dermatology journals included in the Cochrane Collaboration's project on hand searching for RCTs in Spanish, as these findings complement the important earlier work of González-Castro et al.7,16 in identifying trials reported in Actas Dermo-Sifiliográficas between 1948 and 2000 and Medicina Cutánea Ibero-Latino-Americana between 1970 and 2000.

We sought to exhaustively and rigorously search for RCTs in dermatology journals published in the Spanish language.

The only dermatology journal we found to be indexed in both MEDLINE and EMBASE at this time is Actas Dermo-Sifiliográficas. The journals indexed only in EMBASE are Dermatología Revista Mexicana, Revista Argentina de Dermatología, Medicina Cutánea Ibero-Latino-Americana, and Piel.

Included and Excluded Journals

Of the 28 candidate journals, 21 were included. All 28 journals and the reasons for exclusions are shown in Fig. 1.

Figure 1.

Flow chart of the dermatology journal search process and the selection of included and excluded journals.

(0.43MB).

Actas Dermo-Sifiliográficas is the only journal included in this study that specifies and promotes the use of the CONSORT guidelines.17

RCTs Identified

A total of 144 RCTs published since 1969 were found in the 21 included journals (Table 1). Seventy-eight (54%) of the RCTs were published in the 16 Latin American journals searched, and 66 (46%) were found in the 5 Spanish journals (Table 1, which also shows the number of RCTs found in each journal). A large number of the RCTs found in Actas Dermo-Sifiliográficas and Medicina Cutánea Ibero-Latino-Americana were previously gathered and described in 2 earlier publications.7,16

Table 1.

Included Dermatology Journals.

No.	Journal Name	Period Searched	First Year of Publication	Periods Not Searcheda	No. of RCTs Found
	Latin America
1	Dermatología Revista Mexicana	1981–2012	1956	1969–1980, 1984, 1987	21
2	Dermatología Venezolana	1984–2012	1957	1969–1983	13
3	Dermatología Peruana	1996–2012	1996	–	9
4	Revista Asociación Colombiana de Dermatología	1991–2012	1991	–	8
5	Revista Chilena de Dermatología	1985–2012	1985	–	8
6	Dermatología Cosmética, Médica y Quirúrgica	2003–2012	2003	–	4
7	Revista del Centro Dermatológico Pascua	1999–2012	1999	–	4
8	Revista Argentina de Dermatología	1981–2012	1908	1969–1980	4
9	Folia Dermatológica Peruana	1986–2012	1986	–	2
10	Dermatología Argentina	1995–2012	1995	–	2
11	Folia Dermatológica Cubana	2007–2012	2007	–	2
12	Archivos Argentinos de Dermatología	1983–2012	1951	1969–1982	1
13	Actas de Dermatología y Dermatopatología	2001–2009	2001 (publication ceased in 2009)	–	0
14	Revista Dominicana de Dermatología	2010–2012	1971	1971–2009	0
15	Revista Sociedad Ecuatoriana de Dermatología	2003–2010	1991	1991–2002, 2005, 2008, 2009, 2011, 2012	0
16	Gaceta Dermatológica Ecuatoriana	1998	1998	1997–2012	0
Total RCTs in Latin American journals					78
	Spain
17	Actas Dermo-Sifiliográficas	1969–2012	1909	–	30
18	Medicina Cutánea Ibero-Latino-Americana	1969–2012	1966		25
19	Piel	1986–2012	1986	–	10
20	Actualidad Dermatológica	1974–2008	1962	1969–1973,	1
21	Revista Fontilles	2008–2012	1932	1969–2002	0
Total RCTs in Spanish journals					66

Abbreviation: RCT, randomized clinical trial.

a

These years could not be searched because neither print nor digital content was available.

Analysis by 5-year intervals shows that many of the RCTs (89 in total) were published in the last 20 years (from 1993 to 2012). The 25 preceding years (from 1992 to 1968) saw 55 RCTs published (Fig. 2).

Figure 2.

Number of randomized clinical trials published in 5-year periods.

(0.08MB).

The most frequently studied skin diseases in the Spanish trials were psoriasis (11 RCTs), mycoses (9), and acne vulgaris (8). The most frequent disease contexts in the Latin American trials were common warts (9 RCTs), mycoses (8), acne vulgaris (7), and lower-limb ulcers (6). Details of the RCTs found are listed in Appendix 2 of the online supplementary material.

The MEDLINE search identified 3997 entries, 669 of which were in Actas Dermo-Sifilográficas. Two of these entries were initially considered candidates for inclusion, but were not in fact RCTs, and 4 were. These 4 articles included the word randomized in their indexed title. All these RCTs had also been found by hand searching (Fig. 3).

Figure 3.

Flow chart of the digital search process for each database searched and the selection of randomized clinical trials (RCTs). BVS refers to the Biblioteca Virtual de la Salud (Virtual Health Library).

(0.26MB).

The EMBASE search initially yielded 9584 entries (Fig. 3); 16 of these were for RCTs. Of these, 6 had been previously identified by hand searching and 2 were newly identified ones from the Dermatología Revista Mexicana. These 2 articles had not been found during earlier hand searching because neither print nor digital versions of the table of contents could be found for issue 6 of volume 56 or issue 2 of volume 48 (Fig. 3).

The search of indexes on the Spanish language Virtual Health Library (BVS) initially returned 5140 entries (Fig. 3). After filtering for dermatology, 30 candidates were identified; 2 of these were not RCTs. Thus, the digital searches found only 28 (19.5%) of the 144 RCTs that had already been found by hand searching (Fig. 3).

We were unable to hand search all of the following journals because of missing volumes or issues: Dermatología Revista Mexicana, Dermatología Venezolana, Revista Argentina de Dermatología, Archivos Argentinos de Dermatología, the 2 Ecuadoran journals, the single journal published in the Dominican Republic, and Revista Fontilles (Table 1).

RCTs are the principal units of analysis for systematic reviews, clinical practice guidelines, and other documents that synthesize knowledge. Regulatory agencies also require them before medicines can be approved for use in humans.

Hand searching identified about 80% more RCTs than database searching. This finding is consistent with previous reports for our own field, in which journals like Archives of Dermatology18 and Actas Dermo-Sifiliográficas7,19 were searched. It is also consistent with reports for other medical specialties.8 These results underline the importance of manually checking dermatology journals, as many of the RCTs we found could not have been otherwise identified, possibly because MEDLINE and EMBASE are not sensitive to search terms in Spanish.11,19–22 Alternatively, the reason may be that Actas Dermo-Sifiliográficas is the only Spanish language dermatology journal indexed in MEDLINE or that EMBASE does not index all of the other Spanish journals. Compounding the problem is the inherent difficulty of electronic searching in any language other than English. Searches in other languages have returned 37% fewer entries than searches in English.15

Dermatología Revista Mexicana and Dermatología Venezolana were the Latin American journals that published the largest number of RCTs. In Spain, Actas Dermo-Sifilográficas published the most. We offer no explanation for these observations, but we think it may be that more funding is available for conducting RCTs in the countries where these journals are published. It is also possible that these journals have less stringent policies governing the RCTs they publish than other Latin American journals do.

Our findings showed an increase in the number of RCTs published in Spanish in recent decades. The reason for the increase may lie in the importance currently placed on evidence-based medicine, an approach that obliges researchers to value this type of study above others because it provides more and stronger evidence.

We found 78 Spanish-language RCTs in Latin American journals and 66 in Spanish ones published over the course of 44 years. On average, RCTs were published at a rate of 1 to 2 per year, a statistic that is in sharp contrast with the rate of RCT publication in English. Up to 11 RCTs per year are published by Archives of Dermatology alone, for example.18

The CONSORT reporting guidelines were established in 1996 to standardize the way clinical trials are reported in different journals.17 These standards indirectly encourage greater methodological rigor by obliging the researcher to describe the design explicitly and fully. Only 1 journal in this study, namely Actas Dermo-Sifilográficas, requires authors to follow the CONSORT guidelines. There seems to be a need, therefore, not only to promote the design of RCTs but also to encourage journals to require CONSORT-guided reporting, just as the important English-language dermatology journals do.

That both Spanish and Latin American researchers showed interest in psoriasis, mycoses, and acne was noteworthy. The attention is probably attributable to the greater impact of these diseases among patients, given that clinical importance would drive an effort to identify effective therapies; alternatively, the pharmaceutical industry may be more interested in funding RCTs in these areas.23,24

This study identified Spanish-language dermatology journals. Their full texts proved impossible to find in only a few cases once we applied various means to locate them. This experience underlines the importance of posting full texts online because this strategy not only facilitates the identification of RCTs for systematic reviews but also contributes to making knowledge available worldwide and enhancing the visibility of Spanish-language publications.

One of the strengths of this study is the large number of journals, volumes, and issues we searched exhaustively and systematically. A total of 28 dermatology journals were initially identified. Electronic searching found nearly all the RCTs that had been identified manually. Two references were found by database searching but not by hand searching, since print copies of the issues in question were unavailable. A limitation of our study is our lack of access to all of the articles published from the start of publication (Table 1). However, those early issues probably did not contain RCTs, since this design was little used before the 1970s. Another limitation is that we excluded relevant RCTs that were published in journals that focus on other specialties or in dermatology journals published in other languages. Furthermore, we did not evaluate the quality of the RCTs for this report, although we have recently been working on that task and plan to publish the results shortly.

In conclusion, manual searches identified a large number of RCTs in dermatology journals in Spain and Latin America. However, these journals publish far fewer RCTs than English-language dermatology journals do. Finally, ready access to these RCTs and a large number of other articles of interest is still lacking.

Group of Investigative Dermatology (GRID), Medical Faculty, Universidad de Antioquía, Medellín, Colombia.

The authors declare that they have no conflicts of interest.