In a recent article in Communications in Statistics – Theory and Methods, Georgia researchers detail a sampling scheme which can lead to a reduction in sample size and cost in clinical and epidemiological studies of association between a count outcome and risk factor. It shows that inference in two common generalized linear models for count data, Poisson and negative binomial regression, is improved by using a ranked auxiliary covariate, which guides the sampling procedure.
This type of sampling has typically been used to improve inference on a population mean. The novelty of the current work is its extension to log-linear models and derivations showing that the sampling technique results in an increase in information as compared to simple random sampling. Specifically, the work shows that under the proposed sampling strategy the maximum likelihood estimate of the risk factor’s coefficient is improved through an increase in the Fisher’s information. A simulation study is performed to compare the mean squared error, bias, variance, and power of the sampling routine with simple random sampling under various data-generating scenarios. The study also illustrates the merits of the sampling scheme on a real data set from a clinical setting of males with chronic obstructive pulmonary disease.
Empirical results from the simulation study and data analysis coincide with the theoretical derivations, suggesting that a significant reduction in sample size, and hence study cost, can be realized while achieving the same precision as a simple random sample.
Dr. Daniel F. Linder, assistant professor in biostatistics at the Medical College of Georgia, was the lead author. Drs. Jingjing Yin, Haresh Rochani and Hani Samawi, in the department of biostatistics at Georgia Southern University Jiann-Ping Hsu College of Public Health, were co-authors.