Restricted cubic spline function is an ideal model in trend approximation, which is widely used in doseresponse meta-analysis. The spline function, based on parameter technique, is a smoothly joined piecewise polynomial of each knot, with a cubic polynomial in each sub-interval of the slope which fits well in the non-linear trend by changing the number and (or) the sites of the knots. We have introduced the methodology of linear and non-linear slope model in dose-response meta-analysis in the previous article, and in this one, we will give a more detailed discussion on restricted cubic spline function mainly in the following aspects: model building, parameters pooling and knots selecting.
Dose-response meta-analysis, as a subset of meta-analysis, plays an important role in dealing with the relationship between exposure level and risk of diseases. Traditional models limited in linear regression between the independent variables and the dependent variable. With the development of methodology and functional model, Nonlinear regression method was applied to dose-response meta-analysis, such as restricted cubic spline regression, quadratic B-spline regression. However, in these methods, the term and order of the independent variables have been assigned that may not suit for any trend distribution and it may lead to over fitting. Flexible fraction polynomial regression is a good method to solve this problem, which modelling a flexible fraction polynomial and choosing the best fitting model by using the likelihood-ratio test for a more accurate evaluation. In this article, we will discuss how to conduct a dose-response meta-analysis by flexible fraction polynomial.
ObjectiveTo develop reporting guideline for dose-response meta-analysis (DMA), so as to help Chinese authors to understand DMA better and to promote the reporting quality of DMA conducted by them. MethodPubMed, EMbase, The Cochrane Library, CNKI, and WanFang Data were searched from Jan 1st 2011 to Dec 30th 2015 to collect DMA papers published by Chinese authors. The number of these publications by years, whether and what kind of reporting guideline was used, and whether the DMA method claimed in these publications was correct were analysed. Then we drafted a checklist of items for reporting DMA, and organized a discussion meeting with experts from the fields of DMA, evidence-based medicine, clinical epidemiology, and clinicians to collect suggestions for revising the draft reporting guideline for DMA. ResultsOnly 33.73% of the publications clarified it is a DMA on the title and 48.02% of them reported risk of bias. Almost 38.49% of the publications didn't use any reporting guidelines. Fourteen of them claimed an incorrect use of methodology. We primarily took account for 47 potential items related to DMA based on our literature analysis results and existing reporting guidelines for other types of meta-analyses. After the discussion meeting with 6 experts, we revised the items, and finally the G-Dose checklist with 43 items for reporting DMA was developed. ConclusionThere is a lack of attention on reporting guidelines in Chinese authors and evidence suggests these authors may be at risk of incomplete understanding on reporting guidelines. It is strongly recommended to use reporting guidelines for DMA and other types of meta-analyses in Chinese authors.
Dose-response meta-analysis is being increasingly applied in evidence production and clinical decision. The research method, synthesizing certain dose-specific effects across studies with the same target question by a certain types of weighting schedule to get a mean dose-response effect, is to reflect the dose-response relationship between certain exposure and outcome. Currently, the most popular method for dose-response meta-analysis is based on the classical "two-stage approach", with the advantage that it allows fixed- or random-effect model, according to the amount of heterogeneity in the model. There are two types of random-effect model available for dose-response meta-analysis, that is, the generally model and the coefficient-correlation-adjusted model. In this article, we briefly introduce two models and illustrate how they are applied in Stata software, which is expected to provide theoretical foundation for evidence-based practice.
Dose-response meta-analysis serves an important role in investigating the dose-response relationship between independent variables (e.g. dosage) and disease outcomes. Traditional dose-response meta-analysis model is based on one independent variable to consider its own dose-specific effect on the outcome. However, for drug clinical trials, it generally involves two-dimensions of the treatment, such as dosage and course of treatment. These two-dimensions tend to be associated with each other. When neglecting their correlations, the results may be at risk of bias. Moreover, taking account of the "combined effect” of dosage and time on outcome has more clinical value. Therefore, in this article, based on traditional dose-response meta-analysis model, we propose a three-dimension model for dose-response meta-analysis which considers both the effect of dosage and time, to provide a solution for the above-mentioned problems in a traditional model.