In addition to the National Public Health Agenda for Lupus ,NACDD contracted with SciMetrika to conduct a systematic review of the public health literature on lupus that might be relevant to addressing lupus from a public health perspective.
We excluded articles on lupus that addressed only biomarkers, language validation of a survey (not English validation), and clinical diagnosis, were published before 1990, or were editorials on another article (the articles discussed in the editorials, if relevant, were added to the review).
Inclusion criteria. We sought retrievable, peer reviewed articles from 1990 through 2015that addressed lupus and one or more of the following seven topic areas: disease burden, population incidence and prevalence overall and by age, gender, race/ethnicity; comorbidities; dispartiies; access to care and public health.
Burden, incidence, prevalence, and access to care were limited to US studies, while race/ethnicity, comorbidity, disparities, and public health were not limited by country.
Search terms were defined and search syntaxes developed for each topic area to find articles of potential interest. Syntaxes were tailored for each database searched (PubMed, EMBASE, PsycINFO, and the Cochrane Library). SciMetrika identified 2,038 articles of potential interest. After eliminating duplicate articles, 1,541 were reviewed using a two-phase approach to determine relevancy of the articles for each topic area. Phase I reviewed titles and abstracts of the resulting articles for potential relevancy, and added articles to additional topic areas if the article was deemed potentially relevant. Phase II was a full text review of articles determined to be potentially relevant during Phase I. Additional articles suggested by lupus and public health experts or identified during the review of full text were also reviewed in Phase II.
The relevant information from the articles determined to be definitely relevant during Phase II was entered into an Excel spreadsheet during Abstraction. This spreadsheet included citation information, topic/domain area(s), target population, sample size and characteristics, study objective, methodology, results, and conclusions and recommendations.
The 253 articles found to be definitely relevant included 223 original research studies, 18 reviews, 49 commentaries/nonsystematic reviews, and one symposium summary. The 253 articles contributed 292 citations for the seven required topic areas because some articles were relevant to more than one topic area.