Longitudinal versus cross-sectional estimation of lung function decline–further insights
ABSTRACT This paper explores the extent to which differences in longitudinal versus cross-sectional inference may be influenced by the choice of statistical models. Using lung function data on 524 working men, we first compare the goodness-of-fit and implication for longitudinal decline of a variety of cross-sectional models. We then compare the predicted longitudinal patterns from these models with those observed over a period of four years. In general, both approaches provide qualitatively, if not quantitatively, similar messages concerning the relative effects of smoking and age on lung function decline. Nonetheless, we acknowledge the existence of real selection and cohort effects. Although we recognize the utility of cross-sectional designs, we discourage quantitative comparisons between studies, especially longitudinal versus cross-sectional.
- SourceAvailable from: ehrn.co.za
- Annals of the New York Academy of Sciences 02/1991; 624:195-208. · 4.31 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Longitudinal designs are important in medical research and in many other disciplines. Complete longitudinal studies, in which each subject is evaluated at each measurement occasion, are often very expensive and motivate a search for more efficient designs. Recently developed statistical methods foster the use of intentionally incomplete longitudinal designs that have the potential to be more efficient than complete designs. Mixed models provide appropriate data analysis tools. Fixed effect hypotheses can be tested via a recently developed test statistic, FH. An accurate approximation of the statistic's small sample non-central distribution makes power computations feasible. After reviewing some longitudinal design terminology and mixed model notation, this paper summarizes the computation of FH and approximate power from its non-central distribution. These methods are applied to obtain a large number of intentionally incomplete full-span designs that are more powerful and/or less costly alternatives to a complete design. The source of the greater efficiency of incomplete designs and potential fragility of incomplete designs to randomly missing data are discussed.Statistics in Medicine 01/1992; 11(14‐15):1889 - 1913. DOI:10.1002/sim.4780111411 · 2.04 Impact Factor