Is Reduction of Pocket Probing Depth Correlated with the Baseline Value                or is it “Mathematical Coupling”?

Tu, Y-K.; Gilthorpe, Mark S.; Griffiths, Gareth

doi:10.1177/154405910208101013

Cited by 34 publications

(37 citation statements)

References 21 publications

(22 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…46 Such adjustments may in turn lead to misleading inferences regarding the adjusted associations between outcomes and explanatory variables. 47 This is particularly likely to be the reason for the reduced (non-significant) regression coefficient for the relationship between gestational age and pulse pressure after adjustment for mean 24-h systolic blood pressure, which was highly correlated with pulse pressure (r ¼ 0.87). This raises the possibility that pulse pressure may be acting as a surrogate for 24-h systolic blood pressure, particularly as there is no significant change in diastolic blood pressure with gestation, which would serve to distort the ratio determining pulse pressure.…”

Section: Discussionmentioning

confidence: 99%

The relationship between gestational age, systolic blood pressure and pulse pressure in children

2008

View full text Add to dashboard Cite

This cross-sectional study investigates the relationship between gestational age and systolic blood pressure and pulse pressure in childhood. Blood pressure was measured in 483 schoolchildren, free from cardiovascular disease, aged between 6 and 16 years. Pulse pressure was estimated as the difference between the 24-h mean systolic and diastolic blood pressure values. Linear regression showed an inverse relationship between gestational age and mean 24-h systolic blood pressure (adjusted regression coefficient mm Hg per week gestation À0.631, 95% confidence interval (CI) À1.21 to À0.04, P ¼ 0.036). Further, linear regression showed a significant negative association between gestational age and log-transformed pulse pressure (adjusted antilog regression coefficient mm Hg per week of gestation À1.39, 95% CI À2.96 to À0.3, P ¼ 0.013), which after gender-specific analyses was found to be restricted to the girls in the study. The results of the present study suggest that low gestational age is associated with elevated systolic blood pressure and pulse pressure in childhood, the latter particularly in girls. This observation provides some support for the developmental origins of adult disease hypothesis-that adverse events in early life may have long-term consequences for cardiovascular health. However, as gestational age itself is unlikely to be the causal event in determining blood pressure control, further investigation is required, particularly with regard to the nutritional, physiological and molecular mechanisms that explain such epidemiological observations.

show abstract

Section: Discussionmentioning

confidence: 99%

The relationship between gestational age, systolic blood pressure and pulse pressure in children

2008

View full text Add to dashboard Cite

show abstract

“…25 Such adjustments may in turn lead to Birth weight and pulse pressuremisleading inferences regarding the adjusted associations between outcomes and explanatory variables. 26 This is particularly likely to be have the reason for the reduced coefficient for the relationship between birth weight and pulse pressure after adjustment for systolic blood pressure, which was highly correlated with pulse pressure. This also raises the possibilities that pulse pressure may be acting as a surrogate for systolic blood pressure, or that systolic blood pressure lies on the causal pathway between fetal growth and pulse pressure.…”

Section: Discussionmentioning

confidence: 99%

The relationship between birth weight and pulse pressure in children: cross-sectional study

Pearce

O’Sullivan

2006

J Hum Hypertens

View full text Add to dashboard Cite

“…Results from the regression analysis revealed that treatment outcomes were significantly associated with baseline HDD in both treatment groups. As there is mathematical coupling [3][4][5] between baseline HDD and the outcome, horizontal bone fill (ie change in HDD), further statistical analyses are warranted to support this purported association. In general, mathematical coupling occurs when one variable directly or indirectly contains the whole or part of another, and the two variables are then analysed using correlation or regression, 4 such as investigating the relation between change or percentage change in variables (eg following an intervention) and their initial/baseline values (ie prior to the intervention).…”

Section: Example Three: Horizontal Bone Fill and Pocket Depth Clinicmentioning

confidence: 99%

“…In general, mathematical coupling occurs when one variable directly or indirectly contains the whole or part of another, and the two variables are then analysed using correlation or regression, 4 such as investigating the relation between change or percentage change in variables (eg following an intervention) and their initial/baseline values (ie prior to the intervention). [3][4][5] Consequently, the statistical procedure of testing the null hypothesis -that the coefficient of correlation or the slope of regression is zero -becomes inappropriate.…”

Section: Example Three: Horizontal Bone Fill and Pocket Depth Clinicmentioning

confidence: 99%

“…Medical statisticians have repeatedly warned against the misuses of correlation and regression within medical and dental research; 1,2 sometimes correlation and simple regression may give rise to spurious results if researchers do not comprehend fully the underlying statistical theory. [3][4][5] One common problem in the use of multiple linear or logistic regression when analysing clinical data is the occurrence of explanatory variables (covariates) which are not independent, ie correlations amongst covariates are not zero. 6 Most textbooks emphasise that there should be no significant associations between covariates, as this gives rise to the problem known as collinearity.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Problems of correlations between explanatory variables in multiple regression analyses in the dental literature

et al. 2005

View full text Add to dashboard Cite

Multivariable analysis is a widely used statistical methodology for investigating associations amongst clinical variables. However, the problems of collinearity and multicollinearity, which can give rise to spurious results, have in the past frequently been disregarded in dental research. This article illustrates and explains the problems which may be encountered, in the hope of increasing awareness and understanding of these issues, thereby improving the quality of the statistical analyses undertaken in dental research. Three examples from different clinical dental specialities are used to demonstrate how to diagnose the problem of collinearity/multicollinearity in multiple regression analyses and to illustrate how collinearity/multicollinearity can seriously distort the model development process. Lack of awareness of these problems can give rise to misleading results and erroneous interpretations. Multivariable analysis is a useful tool for dental research, though only if its users thoroughly understand the assumptions and limitations of these methods. It would benefit evidencebased dentistry enormously if researchers were more aware of both the complexities involved in multiple regression when using these methods and of the need for expert statistical consultation in developing study design and selecting appropriate statistical methodologies.

show abstract

Is Reduction of Pocket Probing Depth Correlated with the Baseline Value or is it “Mathematical Coupling”?

Cited by 34 publications

References 21 publications

The relationship between gestational age, systolic blood pressure and pulse pressure in children

The relationship between gestational age, systolic blood pressure and pulse pressure in children

The relationship between birth weight and pulse pressure in children: cross-sectional study

Problems of correlations between explanatory variables in multiple regression analyses in the dental literature

Contact Info

Product

Resources

About