Impactia: automating the study of the scientific production

Introduction

The Andalusian Public Health System Virtual Library (Biblioteca Virtual del Sistema Sanitario Público de Andalucía, BV-SSPA) was set up in June 2006. It consists of a regional government action with the aim of democratizing the health professional access to quality scientific information, regardless of the professional workplace.

Andalusia is a region with more than 8 million inhabitants, with 100,000 health professionals for 41 hospitals, 1,500 primary healthcare centres, and 28 centres for non-medical attention purposes (research, management, and educational centres).

Objectives

The Department of Development, Research and Investigation (R+D+i) of the Andalusian Regional Government has, among its duties, the task of evaluating the hospitals and centres of the Andalusian Public Health System (SSPA)  in order to distribute its funding. Among the criteria used is the evaluation of the scientific output, which is measured using bibliometry.

It is well-known that the bibliometry has a series of limitations and problems that should be taken into account, especially when it is used for non-information sciences, such us career, funding, etc.

A few years ago, the bibliometric reports were done separately in each centre, but without using preset and well-defined criteria, elements which are basic when we need to compare the results of the reports. It was possible to find some hospitals which were including Meeting Abstracts in their figures, while others do not, and the same was happening with Erratum and many other differences.

Therefore, the main problem that the Department of R+D+i had to deal with, when they were evaluating the health system, was that bibliometric data was not accurate and reports were not comparable.

With the aim of having an unified criteria for the whole system, the Department of R+D+i  ordered the BV-SSPA to do the year analysis of the scientific output of the system, using some well defined criteria and indicators, among whichstands out the Impact Factor.

Materials and Methods

As the Impact Factor is the bibliometric indicator that the virtual library is asked to consider, it is necessary to use the database Web of Science (WoS), since it is its owner and editor. The WoS includes the databases Science Citation Index (SCI), Social Sciences Citation Index (SSCI) and Arts & Humanities Citation Index.  To gather all the documents, SCI and SSCI are used; to obtain the Impact Factor and quartils, it is used the Journal Citation Reports, JCR.

Unlike other bibliographic databases, such us MEDLINE, the bibliometric database WoS includes the address of all the authors. In order to retrieve all the scientific output of the SSPA, we have done general searches, which are afterwards processed by a tool developed by our library. We have done nine different searches using the field ‘address’; eight of them including ‘Spain’ and each one of the eight Andalusian Regions, and the other one combining ‘Spain’ with all those cities where there are health centres, since we have detected that there are some authors that do not use the region in their signatures. These are some of the search strategies:

• AD=Malaga and AD=Spain
• AD=Sevill* and AD=Spain
• AD=SPAIN AND (AD=GUADIX OR AD=BAZA OR AD=MOTRIL)

Further more, the field ‘year’ is used to determine the period.

To exploit the data, the BV-SSPA has developed a tool called Impactia. It is a web application which uses a database to store the information of the documents generated by the SSPA. Impactia allows the user to automatically process the retrieved documents, assigning them to their correspondent centres.

In order to do the classification of documents automaticaly, it was necessary to detect the huge variability of names of the centres that the authors use in their signatures. Therefore, Impactia knows that if an author signs as “Hospital Universitario Virgen Macarena”, “HVM” or “Hosp. Virgin Macarena”, he belongs to the same centre. The figure attached shows the variability found for the Empresa Publica Hospital de Poniente.

Besides the documents from WoS, Impactia includes the documents indexed in Scopus and in other databases, where we do bibliographic searches using similar strategies to the later ones.

Aware that in the health centres and hospitals there is a lot of grey literature that is not gathered in databases, Impactia allows the centres to feed the application with these documents, so that all the SSPA scientific output is gathered and organised in a centralized place.
The ones responsible of localizing this gray literature are the librarians of each one of the centres. They can also do statements to the documents and indicators that are collected and calculated by Impactia.

The bulk upload of documents from WoS and Scopus into Impactia is monthly done.

One of the main issues that we found during the development of Impactia was the need of dealing with duplicated documents obtained from different sources. Taking into account that sometimes titles might be written differently, with slashes, comas, and so on, Impactia detects the duplicates using the field ‘DOI’ if it is available or comparing the fields: page start, page end and ISSN. Therefore it is possible to guarantee the absence of duplicates.

Results

The data gathered in Impactia becomes available to the administrative teams and hospitals managers, through an easy web page that allows them to know at any moment, and with just one click, the detailed information of the scientific output of their hospitals, including useful graphs such as percentage of document types, journals where their scientists usually publish, annual comparatives, bibliometric indicators and so on. They can also compare the different centres of the SSPA.

Impactia allows the user to download the data from the application, so that he can work with this information or include them in their centres’ reports.

This application saves the health system many working hours. It was previously done manually by forty one librarians, while now it is done by only one person in the BV-SSPA during two days a month.

To sum up, the benefits of Impactia are:

• It has shown its effectiveness in the automatic classification, treatment and analysis of the data.
• It has become an essential tool for all managers to evaluate quickly and easily the scientific production of their centers. 
• It optimizes the human resources of the SSPA, saving time and money.
• It is the reference point for the Department of R+D+i to do the scientific health staff evaluation.