Article type
Year
Abstract
Background: Clinical medicine is becoming increasingly interested in clinical prediction rules (CPRs) and their potential for improving patient care. However, identifying CPRs for primary care from electronic databases is difficult because several terms are used interchangeably to describe them.
Objectives: The aim of the current work is to develop an international register of CPRs relevant to primary care to be distributed through the Cochrane Primary Health Care Field.
Methods: The register is being developed by identifying relevant articles from various sources including: (1) the MEDLINE database [1980-2008] using an electronic search string, specifically developed and tested in-house; (2) other electronic resources (e.g. JAMA rational clinical examination series); and (3) contacting authors in the field for published and unpublished work. Relevant articles are classified according to a set of criteria including condition specific codes, level of evidence and quality assessment.
Results: To date, the register consists of 300 articles that contain CPRs relevant to primary care, covering 16 different broad clinical areas. The largest number of articles was identified for cardiovascular conditions (n = 93) and the least identified for eye conditions (n = 1). These broad areas are further classified according to specific conditions. The register identified CPRs across all levels of evidence. The majority of articles were concerned with validating rules (narrow validation n = 99, broad validation (n = 53). Few articles were concerned with formal impact analysis (n = 8). Assessment of the quality of the published articles produced mixed results.
Conclusions: Although the register will need to be regularly updated, it offers many potential benefits for primary care practice. Classification of articles allows the clinician to objectively assess the utility of the CPR for use with patients. The register will also assist researchers to identify clinical areas in which CPRs have not yet been derived, as well as CPRs that have not undergone formal impact analysis.
Objectives: The aim of the current work is to develop an international register of CPRs relevant to primary care to be distributed through the Cochrane Primary Health Care Field.
Methods: The register is being developed by identifying relevant articles from various sources including: (1) the MEDLINE database [1980-2008] using an electronic search string, specifically developed and tested in-house; (2) other electronic resources (e.g. JAMA rational clinical examination series); and (3) contacting authors in the field for published and unpublished work. Relevant articles are classified according to a set of criteria including condition specific codes, level of evidence and quality assessment.
Results: To date, the register consists of 300 articles that contain CPRs relevant to primary care, covering 16 different broad clinical areas. The largest number of articles was identified for cardiovascular conditions (n = 93) and the least identified for eye conditions (n = 1). These broad areas are further classified according to specific conditions. The register identified CPRs across all levels of evidence. The majority of articles were concerned with validating rules (narrow validation n = 99, broad validation (n = 53). Few articles were concerned with formal impact analysis (n = 8). Assessment of the quality of the published articles produced mixed results.
Conclusions: Although the register will need to be regularly updated, it offers many potential benefits for primary care practice. Classification of articles allows the clinician to objectively assess the utility of the CPR for use with patients. The register will also assist researchers to identify clinical areas in which CPRs have not yet been derived, as well as CPRs that have not undergone formal impact analysis.