Publication bias - a reason for the decreased research output in developing countries

Abstract

Objective: The low level of mental health research from low and middle-income countries, as measured by the relative lack of publications in high impact journals is an inaccurate reflection of research being conducted in these countries. The number of manuscripts submitted for publication is a more accurate measure of the research activities. The study aimed to quantify the number of manuscripts submitted and accepted for publication in high impact psychiatric journals.Method: Editors of 8 psychiatric journals were requested information on the number of manuscripts submitted, the country of origin and the number of manuscripts accepted for publication from April to September 2005. Results: 5,2 % of all manuscripts submitted for publications were from low and middle income countries. The overall acceptance rate of manuscripts was 16,6 % but the acceptance rate for low and middleincome countries was 4,8 %. Manuscripts from high-income countries had a 5,8 times greater odds (2,5 - 4,9) of being accepted for publication than an article from a low and middle-income country. Conclusion: Both the quantity and quality of research from low and middle-income countries must be improved. Interventions to improve the quality of research must be directed towards capacity development, increasing international collaborations, mentoring of researchers and establishing formal psychiatric epidemiology training programs to equip researchers with skills to produce papers that meet the publication criteria of reviewers. Studies must become more innovative and include the changing paradigm in epidemiological research. Manuscripts that describe innovative studies have a greater chance of being published. Keywords: epidemiology, research, psychiatry South African Psychiatry Review Vol. 9(3) 2006: 153-155

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 38 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.