Using levels of evidence to compare clinical impact from research

Objectives: Impact of medical institutions on clinical decision-making globally might be estimated by the level of evidence of their research articles. The aim of this study was to compare levels of evidence of articles for Pakistan. Methods: We compared levels of evidence of articles from Pakistan, Nigeria, Japan, and the United States (U.S.). Results: Majority (73%) of articles in U.S. general medical journals were high levels (1-2), while majority (66% to 95%) in Japanese, Nigerian, Pakistani, and sub-specialty U.S. journals were lower levels (3-4) (P < 0.001). The number of articles from various regions of Pakistan did not correspond to their population or number of medical colleges/universities, mainly due to the skewing effect of one institution from Karachi which contributed 38% of all high-level articles. Conclusions: A comparison of levels of evidence of articles across institutions might reflect relative potential of clinical impact, and might be useful for institutions, policy makers, and health research planners for priority setting.