Quality and Usability of Common Drug Information Databases: METHODS

30 Dec
2010

A qualitative and quantitative data collection and analysis strategy, with descriptive and inferential statistics, was used to assess the quality, performance, and usability of the databases and users’ preferences, to determine overall database preference. This study satisfied all ethical review requirements of the investigators’ organization.

Selection of Databases and Study Participants

The online drug information databases selected for evaluation were those most commonly used or under consider­ation for purchase or renewal by the various hospital pharmacy practice sites within Vancouver Coastal Health — Providence Health Care at the time of the study (early 2009). The 3 databases evaluated were Clinical Pharmacology (Gold Standard Inc, an Elsevier company, Tampa, Florida), Lexi- Comp Online (Lexi-Comp Inc, Hudson, Ohio), and Micromedex (Thomson Reuters [Healthcare] Inc, Greenwood Village, Colorado) (Table 1). The publishers of the databases provided free trial access for the purposes of this study.
cialis super active

The performance and usability of the databases, as well as users’ preferences, were evaluated during February 2009 with a convenience sample of 26 pharmacists. This sample reflected a cross-section of pharmacist roles from a variety of hospital pharmacy practice sites across the health region. The investigators invited pharmacists to participate in the study, with the number of pharmacists recruited from each site weighted approximately by the number of pharmacists employed at each practice site. Data were also collected regarding the study participants’ characteristics, including length of time in pharma­cy practice, level of pharmacy training attained, and frequency of use of drug information databases before participating in the study (see Appendix 1 for questionnaire).

Table 1. Online Drug Information Databases Evaluated

Characteristic

Clinical Pharmacology

Lexi-Comp Online

Micromedex


Software suite and version*


Clinical Pharmacology (version not
applicable)


Lexi-COMPLETE


(version not applicable)


Micromedex Healthcare
Series, volume

139

(December
2008)


Publisher


Gold Standard Inc (an Elsevier
Company)


Lexi-Comp, Inc


Thomson Reuters (Healthcare) Inc


Platform options


Internet Intranet CD-ROM Mobile device


Internet


Local computer installation Mobile
device


Internet Intranet CD-ROM Mobile device


Supported mobile devices


Palm


Pocket PC/Windows Mobile


Blackberry


iPhone/iPod touch


Palm


Pocket PC/Windows Mobile


Blackberry


iPhone/iPod touch


Palm


Pocket PC/Windows Mobile


Blackberry


iPhone/iPod touch


CD-ROM = compact disk read-only memory. *Drug information databases were accessed during February 2009.

Development of Drug Information Questions

The drug information questions used in this study were selected from those previously published, with balanced representation for different drug classes, disease states, and drug information categories. The questions were modified for relevance to hospital pharmacy practice and were contextual- ized with Canadian brand names as applicable. Ultimately, a set of 15 drug information questions covering 17 different categories of drug information was created for use in this study (Appendix 2).
canadian antibiotics

Evaluation of Database Quality

The study investigators evaluated the quality of the databases in terms of 5 quality indicator categories, namely referencing, grading of evidence cited or recommendations provided, editorial policy for content updates, provision of authorship information, and inclusion of Canadian content. These quality indicators, chosen on the basis of those previously established, were criteria thought to increase a Canadian hospital pharmacist’s confidence in the quality and relevance of drug information provided by an online drug information database. The investigators assessed the quality indicator criteria through a comprehensive review of each database and the respective publishers’ websites, followed by development of consensus ratings for each quality indicator category on a scale of 0 to 3. The quality score for each database was calculated as the mean of investigators’ consensus ratings across all 5 quality indicator categories.

top