by Chi-Kwok CHAN
Citations according to Clarivate Analytics Science Citation Index (SCI) database are recognized as one of the standard measures of academic influence. In 2014, “Nature” reported the top cited 100
research articles of all disciplines of all times. The top article (also called Lowry article) had been cited 305,000 times and the article ranked 100th had had 12,119 citations before the end of 2013. For
comparison, the top Urology article has 5,587 citations to date and was authored by H Feldman in Journal of Urology in 1994, which is a longitudinal observational study showing the correlation between erectile dysfunction and medical and psychosocial illnesses {J Urol 1994; 151(1): 54-61}. To the urologists, 5587-citation is not a shame. Watson and Crick paper in 1953, describing the double
helix structure of DNA, only earned 5207 citations before 2014, whereas the original article reporting special theory of relativity by Albert Einstein has only been cited 5,455 times by the end of 2018 !!!
Of the 100 articles cited by “Nature”, frankly speaking, I am barely able to comprehend only five of them. However, all of these five articles have had immense and far-fetched influence on how the medical or urological research is being carried out across the globe nowadays. The article ranked 11th was written by Kaplan & Meier (both were PhD graduates of Princeton University) in 1958 – Kaplan-Meier product limit estimator for survival (55,507 citations to date), followed by an paper co-authored by JR Landis and GG Koch in 1977 (Biometrics 33(1): 159 – 174) on the refinement of using Kappa-statistics (1st described by Galton in 1892 and later by Smeeton in 1895 !) in measuring agreement between categorical data (50,281 citations to date). The 24th article was published by David COX in 1972, titled “Regression Models and Life-Tables” ( i.e. Cox proportional hazard regression model with time varying co-variates ) (Journal of the Royal Statistical Society. Series B (Methodological), Vol. 34, No. 2, 187 – 220) (49012 citations to date). The 29th article was jointly written by Marin BLAND and Douglas ALTMAN and published in Lancet in 1986 (Lancet 1986; 327: 307 – 310). Of note, it is just 3-page long !!!!!!! ( i.e. Bland-Altman plot to assess the agreement between two measurements of continuous parameters) (42,718 citations to date). The 64th article was by DB DUNCAN in 1955 (Biometrics 11: 1-42) ( yes, the post-hoc test performed after finishing ANOVA) (29415 citations to date)
To the medical statisticians, the most influential work goes to the joint effort by Kaplan & Meier (American Statistical Association). However, their work was only appreciated 11 years later by the medical community (1969) and even further behind by the urologists because of the limited access to the computing statistical package at that time. The first urological article using K-M method to describe the survival outcome was reported by the urologists of Mayo Clinic in the gold journal “Urology” in 1982, showing that men with high grade and high stage prostate cancer treated with radical prostatectomy and pelvic lymphadenectomy (with or without adjuvant hormonal therapy) could live longer than men with similar disease but just treated by radiotherapy / hormonal therapy or their combination. {Urology 1982 19(3): 238 – 247}.