Statnote 25: Stepwise multiple regression

An investigator may also wish to select a small subset of the X variables which give the best prediction of the Y variable. In this case, the question is how many variables should the regression equation include? One method would be to calculate the regression of Y on every subset of the X variables and choose the subset that gives the smallest mean square deviation from the regression. Most investigators, however, prefer to use a ‘stepwise multiple regression’ procedure. There are two forms of this analysis called the ‘step-up’ (or ‘forward’) method and the ‘step-down’ (or ‘backward’) method. This Statnote illustrates the use of stepwise multiple regression with reference to the scenario introduced in Statnote 24, viz., the influence of climatic variables on the growth of the crustose lichen Rhizocarpon geographicum (L.)DC.

Probing three-way interactions in moderated multiple regression:Development and application of a slope difference test

Researchers often use 3-way interactions in moderated multiple regression analysis to test the joint effect of 3 independent variables on a dependent variable. However, further probing of significant interaction terms varies considerably and is sometimes error prone. The authors developed a significance test for slope differences in 3-way interactions and illustrate its importance for testing psychological hypotheses. Monte Carlo simulations revealed that sample size, magnitude of the slope difference, and data reliability affected test power. Application of the test to published data yielded detection of some slope differences that were undetected by alternative probing techniques and led to changes of results and conclusions. The authors conclude by discussing the test's applicability for psychological research. Copyright 2006 by the American Psychological Association.

Mathematical modelling of patient risks in a hospital environment, using multiple regression analysis

The aim of this research work was primarily to examine the relevance of patient parameters, ward structures, procedures and practices, in respect of the potential hazards of wound cross-infection and nasal colonisation with multiple resistant strains of Staphylococcus aureus, which it is thought might provide a useful indication of a patient's general susceptibility to wound infection. Information from a large cross-sectional survey involving 12,000 patients from some 41 hospitals and 375 wards was collected over a five-year period from 1967-72, and its validity checked before any subsequent analysis was carried out. Many environmental factors and procedures which had previously been thought (but never conclusively proved) to have an influence on wound infection or nasal colonisation rates, were assessed, and subsequently dismissed as not being significant, provided that the standard of the current range of practices and procedures is maintained and not allowed to deteriorate. Retrospective analysis revealed that the probability of wound infection was influenced by the patient's age, duration of pre-operative hospitalisation, sex, type of wound, presence and type of drain, number of patients in ward, and other special risk factors, whilst nasal colonisation was found to be influenced by the patient's age, total duration of hospitalisation, sex, antibiotics, proportion of occupied beds in the ward, average distance between bed centres and special risk factors. A multi-variate regression analysis technique was used to develop statistical models, consisting of variable patient and environmental factors which were found to have a significant influence on the risks pertaining to wound infection and nasal colonisation. A relationship between wound infection and nasal colonisation was then established and this led to the development of a more advanced model for predicting wound infections, taking advantage of the additional knowledge of the patient's state of nasal colonisation prior to operation.

Statnote 24: Multiple regression

In previous statnotes, the application of correlation and regression methods to the analysis of two variables (X,Y) was described. These methods can be used to determine whether there is a linear relationship between the two variables, whether the relationship is positive or negative, to test the degree of significance of the linear relationship, and to obtain an equation relating Y to X. This Statnote extends the methods of linear correlation and regression to situations where there are two or more X variables, i.e., 'multiple linear regression’.

Measuring the degree of spatial correlation between histological features in thin sections of brain tissue

Histological features visible in thin sections of brain tissue, such as neuronal perikarya, blood vessels, or pathological lesions may exhibit a degree of spatial association or correlation. In neurodegenerative disorders such as AD, Pick's disease, and CJD, information on whether different types of pathological lesion are spatially correlated may be useful in elucidating disease pathogenesis. In the present article the statistical methods available for studying spatial association in histological sections are reviewed. These include tests of interspecific association between two or more histological features using χ2 contingency tables, measurement of 'complete' and 'absolute' association, and more complex methods that use grids of contiguous samples. In addition, the use of correlation matrices and stepwise multiple regression methods are described. The advantages and limitations of each method are reviewed and possible future developments discussed.

Correlation between visual function and visual ability in patients with uveitis

Aim: To investigate the correlation between tests of visual function and perceived visual ability recorded with a quality of life questionnaire for patients with uveitis. Methods: 132 patients with various types of uveitis were studied. High (monocular and binocular) and low (binocular) contrast logMAR letter acuities were recorded using a Bailey-Lovie chart. Contrast sensitivity (binocular) was determined using a Pelli-Robson chart. Vision related quality of life was assessed using the Vision Specific Quality of Life (VQOL) questionnaire. Results: VQOL declined with reduced performance on the following tests: binocular high contrast visual acuity (p = 0.0011), high contrast visual acuity of the better eye (p = 0.0012), contrast sensitivity (p = 0.005), binocular low contrast visual acuity (p = 0.0065), and high contrast visual acuity of the worse eye (p = 0.015). Stepwise multiple regression analysis revealed binocular high contrast visual acuity (p <0.01) to be the only visual function adequate to predict VQOL. The age of the patient was also significantly associated with perceived visual ability (p <0.001). Conclusions: Binocular high contrast visual acuity is a good measure of how uveitis patients perform in real life situations. Vision quality of life is worst in younger patients with poor binocular visual acuity.

The effect of corneal thickness and corneal curvature on pneumatonometer measurements

Purpose. The purpose of this study was to investigate the influence of corneal topography and thickness on intraocular pressure (IOP) and pulse amplitude (PA) as measured using the Ocular Blood Flow Analyzer (OBFA) pneumatonometer (Paradigm Medical Industries, Utah, USA). Methods. 47 university students volunteered for this cross-sectional study: mean age 20.4 yrs, range 18 to 28 yrs; 23 male, 24 female. Only the measurements from the right eye of each participant were used. Central corneal thickness and mean corneal radius were measured using Scheimpflug biometry and corneal topographic imaging respectively. IOP and PA measurements were made with the OBFA pneumatonometer. Axial length was measured using A-scan ultrasound, due to its known correlation with these corneal parameters. Stepwise multiple regression analysis was used to identify those components that contributed significant variance to the independent variables of IOP and PA. Results. The mean IOP and PA measurements were 13.1 (SD 3.3) mmHg and 3.0 (SD 1.2) mmHg respectively. IOP measurements made with the OBFA pneumatonometer correlated significantly with central corneal thickness (r = +0.374, p = 0.010), such that a 10 mm change in CCT was equivalent to a 0.30 mmHg change in measured IOP. PA measurements correlated significantly with axial length (part correlate = -0.651, p < 0.001) and mean corneal radius (part correlate = +0.459, p < 0.001) but not corneal thickness. Conclusions. IOP measurements taken with the OBFA pneumatonometer are correlated with corneal thickness, but not axial length or corneal curvature. Conversely, PA measurements are unaffected by corneal thickness, but correlated with axial length and corneal radius. These parameters should be taken into consideration when interpreting IOP and PA measurements made with the OBFA pneumatonometer.

Macular pigment optical density is related to blood glutathione levels in healthy individuals

Purpose. To assess the relationship between macular pigment optical density (MPOD) and blood markers for antioxidant defense in otherwise healthy volunteers. Methods. Forty-seven healthy volunteers were subjected to blood analysis to detect the level of circulating glutathione in its reduced (GSH) and oxidized (GSSG) forms. The level of MPOD was measured using heterochromatic flicker photometry. Systemic blood pressure (BP) parameters, heart rate (HR), body mass index (BMI), and plasma levels of total, HDL, and LDL cholesterol and triglycerides (TGs) were also determined. Results. A simple correlation model revealed that the level of MPOD correlated significantly and positively with both GSH (P < 0.001) and t-GSH (P < 0.001) levels but not with those of GSSG (P > 0.05). Age, sex, systemic BP parameters, HR, BMI, and plasma levels of cholesterol and TGs did not have any influence on either MPOD or glutathione levels (all P > 0.05). In addition, a forward stepwise multiple regression analysis showed MPOD to have a significantly and independent correlation with GSH levels (ß = 0.63; P < 0.001). Conclusions. In otherwise healthy older individuals, there is a positive correlation between local and systemic antioxidant defense mechanisms.

Relationships between beta-amyloid (A beta) deposits and blood vessels in patients with sporadic and familial Alzheimer's disease

The density of diffuse, primitive and classic beta-amyloid (A beta) deposits was studied in relation to the incidence of blood vessels in the superior frontal gyrus of nine cases of sporadic Alzheimer's disease (SAD), two cases of familial Alzheimer's disease (FAD) with amyloid precursor protein (APP) mutations (APP717, Val --> Ile), and eight cases of FAD not linked to chromosomes 21, 14 or 1. Stepwise multiple regression was used to determine for each patient whether variations in the density of A beta deposits along the cortex were significantly correlated with the incidence of blood vessels. In the majority of FAD and SAD cases, the density of the diffuse and primitive type A beta deposits was not related to blood vessels. However, the incidence of the larger diameter (> 10 microns) blood vessels was positively correlated with the density of the classic A beta deposits in eight (89%) SAD and two (20%) FAD cases. The data suggest that the densities of vessels and deposits were not significantly correlated between cases but only within cases, suggesting a strictly local effect. In addition, the spatial association between classic A beta deposits and blood vessels may be more apparent in SAD compared with FAD cases.

Alzheimer's disease: the relationship between the density of senile plaques, neurofibrillary tangles and A4 protein in human patients

The numerical density of senile plaques (SP) and neurofibrillary tangles (NFT) as revealed by the Glees silver method was compared with SP and NFT revealed by the Gallyas method and with amyloid (A4) deposits in immunostained sections in 6 elderly cases of Alzheimer's disease. The density of NFT was generally greater and A4 lower in tissue from hippocampus compared with the neocortex suggesting that A4 deposition was less important than the degree of paired helical filament (PHF) related damage in the hippocampus. The density of Glees SP was positively correlated Gallyas SP weakly correlated with A4 deposit number. A stepwise multiple regression analysis which included A4 deposit and Gallyas SP density and accounted for 54% of the variation in Glees SP density. Hence, different populations of SP were revealed by the different staining methods. The results suggested that the Glees method may stain a population of SP in a region of cortex where both amyloid deposition and neurofibrillary changes have occurred.

Measuring the degree of spatial correlation between pathological lesions in Alzheimer's disease

The pathological lesions characteristic of Alzheimer's disease (AD), viz., senile plaques (SP) and neurofibrillary tangles (NFT) may not be randomly distributed with reference to each other but exhibit a degree of sptial association or correlation, information on the degree of association between SP and NFT or between the lesions and normal histological features, such as neuronal perikarya and blood vessels, may be valuable in elucidating the pathogenesis of AD. This article reviews the statistical methods available for studying the degree of spatial association in histological sections of AD tissue. These include tests of interspecific association between two or more histological features using chi-square contingency tables, measurement of 'complete' and 'absolute' association, and more complex methods that use grids of contiguous samples. In addition, analyses of association using correlation matrices and stepwise multiple regression methods are described. The advantages and limitations of each method are reviewed and possible future developments discussed.

The levels of ribitol, arabitol and mannitol in individual lobes of the lichen Parmelia conspersa (Ehrh. ex Ach.) ACH.

The levels of the soluble carbohydrates ribitol, arabitol and mannitol were measured in individual lobes of the lichen Parmelia conspersa (Ehrh. ex Ach.) Ach. Lobes were collected from a north and a south facing slate rock surface in South Gwynedd, Wales, U.K. on 4 days during 1990-1991. On each day sampled, the most significant variation in soluble carbohydrate levels was between the individual lobes of a thallus. In addition, carbohydrate levels were significantly greater on the south facing rock surface on 2 of the 4 days sampled. Factorial analyses of variance suggested that the levels of individual carbohydrates varied significantly between days but not between north and south facing rock surfaces. Mannitol levels varied less between days than arabitol levels. Levels of ribitol, arabitol and mannitol were positively correlated in individual lobes. A stepwise multiple regression suggested that on the north facing rock surface, arabitol and mannitol levels could be explained by variations in the level of ribitol. By contrast, on the south facing rock surface, the levels of fungal carbohydrates were less dependent on the level of ribitol and there was evidence of a relationship between arabitol and mannitol. Variations in carbohydrate production, allocation and metabolism could help to explain lobe growth variation in foliose lichens and the radial growth of lobes over a longer period of time. Greater carbohydrate production rather than differences in allocation and metabolism may explain the increased growth and frequency of P. conspersa on south facing rock surfaces in South Gwynedd. © 1994.

Alzheimer's disease: Are cellular neurofibrillary tangles linked to beta/A4 formation at the projection sites?

The density of senile plaques (SP) and cellular neurofibrillary tabgles (NFT) revealed by the Glees and Gallyas stains; and beta/A4 deposits revealed by immunocytochemical staining, was estimated in the hippocampus and adjacent gyri in Alzheimer's disease (AD). Stepwise multiple regression was used to detemine whether the density of cellular NFT was related to the density of SP or beta/A4 deposits totalled over the projection sites. Cellular NFT density was only weakly correlated with the density of Glees SP and beta/A4 deposits at some of the projection sites. However, beta/A4 deposit density in a tissue was strongly correlated with the density of beta/A4 deposits at the projection sites suggesting that the lesions could spread through the brain. Hence, although there is a strong correlation between the density of beta/A4 deposits in different parts of the hippocampal formation there is little association between SP or beta/A4 and cellular NFT. These results do not provide strong evidence that beta/A4 protein is the cause of the neuritc changes in AD.

Predicting changes in measured introcular pressure with non-contact tonometry after LASIK

Purpose: Optometrists are becoming more integrally involved in the diagnosis of and care for glaucoma patients in the UK. The correlation of apparent change in non contact tonometry (NCT) IOP measurement and change in other ocular parameters such as refractive error, corneal curvature, corneal thickness and treatment zone size (data available to optometrists after LASIK) would facilitate care of these patients. Setting: A UK Laser Eye Clinic. Methods: This is a retrospective study study of 200 sequential eyes with myopia with or without astigmatism which underwent LASIK using a Hansatome and an Alcon LADARvision 4000 excimer laser. Refraction keratometry, pachymetry and NCT IOP mesurements were taken before treatmebnt and agian 3 months after treatment. The relationship between these variables anfd teh treatment zones were studied using stepwise multiple regression analysis. Results: There was a mean difference of 5.54mmHg comnparing pre and postoperative NCT IOP. IOP change correlates with refractive error change (P < 0.001), preoperative corneal thickness (P < 0.001) and treatment zone size (P = 0.047). Preoperative corneal thickness correlates with preoperative IOP (P < 0.001) and postoperative IOP (P < 0.001). Using these correlations, the measured difference in NCT IIOP can be predicted preoperatively or postoperatively using derived equations.Conclusion: There is a significant reduction in measured NCT IOP after LASIK. The amount of reduction can be calculated using data acquired by optometrists. This is helpful for opthalmologists and optometrists who co-manage glaucoma patients who have had LASIK or with glaucoma pateints who are consideraing having LASIK.