RNRoger B. Newson

Roger B. Newson

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Roger B. Newson has a total of 30 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Geographical variation in the prevalence of sensitization to common aeroallergens in adults: the GA2LEN survey

    by R. B. Newson, R. van Ree, B. Forsberg, C. Janson, J. Lötvall, S.‐E. Dahlén, E. M. Toskala, J. Bælum, G. M. Brożek, L. Kasper, et al.

    Background: Geographical variation in the prevalence of sensitization to aeroallergens may reflect differences in exposure to risk factors such as having older siblings, being raised on a farm or other unidentified exposures.Objective: We wanted to measure geographical variation in skin prick test positivity and assess whether it was explained by differences in family size and/or farm exposure. We also compared prevalence in younger and older subjects.Methods: Within the Global Allergy and Asthma European Network (GA2LEN) survey, we measured the prevalence of skin prick positivity to a panel of allergens, and geometric mean serum total immunoglobulin E (IgE), in 3451 participants aged 18–75 years in 13 areas of Europe. Estimated prevalence was standardized to account for study design. We compared prevalence estimates in younger and older subjects and further adjusted for age, gender, smoking history, farm exposure, number of older siblings and body mass index (BMI).Results: Skin prick test positivity to any one of the measured allergens varied within Europe from 31.4% to 52.9%. Prevalence of sensitization to single allergens also varied. Variation in serum total IgE was less marked. Younger participants had higher skin prick sensitivity prevalence, but not total IgE, than older participants. Geographical variation remained even after adjustment for confounders.Conclusion: Geographical variation in the prevalence of skin prick test positivity in Europe is unlikely to be explained by geographical variation in gender, age, smoking history, farm exposure, family size and BMI. Higher prevalence in younger, compared to older, adults may reflect cohort‐associated increases in sensitization or the influence of ageing on immune or tissue responses.


  • Attributable and Unattributable Risks and Fractions and other Scenario Comparisons

    by Roger B. Newson

    Scenarios are alternative versions of the same dataset with the same variables but different observations or values. Applied scientists frequently want to predict how much good an intervention will do by comparing outcomes from the same model between different scenarios. Alternatively, they may want to compare outcomes between different models applied to the same scenario, for instance, when standardizing statistics from different subpopulations to a common gender and age distribution. Standard Stata tools for scenario means and comparisons are margins and pwcompare. A suite of packages is presented for estimating scenario means and comparisons by using margins, together with normalizing and variance-stabilizing transformations implemented by using nlcom.margprev estimates marginal prevalences; marglmean estimates marginal arithmetic means; regpar estimates the difference between two marginal prevalences (the population attributable risk); punaf estimates the ratio between two marginal arithmetic means (the population unattributable fraction); and punafcc estimates a marginal mean between-scenario risk or hazard ratio for case–control or survival data (also known as a population unattributable fraction). The population unattributable fraction and its confidence limits are subtracted from 1 to estimate the population attributable fraction. Formulas and examples are presented, including an example from the Global Allergy and Asthma European Network.


  • The association of asthma, nasal allergies, and positive skin prick tests with obesity, leptin, and adiponectin

    by R. B. Newson, M. Jones, B. Forsberg, C. Janson, A. Bossios, S.‐E. Dahlen, E. M. Toskala, A. Al‐Kalemji, M. L. Kowalski, B. Rymarczyk, et al.

    Summary: Background: Cross‐sectional and longitudinal reports show that obese adults have more asthma than non‐obese adults. A proposed mechanism is via effects of adipokines (leptin and adiponectin) on the immune system.Objective: We wished to measure the associations of asthma and other atopic diseases with serum adipokine levels and to find whether the associations with asthma were strong enough to rule out the possibility that they are secondary to the association of fatness measures with asthma.Methods: The Global Asthma and Allergy Network of Excellence (GA2LEN) clinical follow‐up survey is a clinical survey, embedded in a larger multi‐centre cross‐sectional postal survey, involving, with a case/control design, enrichment of the sample with subjects with asthma and chronic rhinosinusitis (CRS). We recorded serum leptin or adiponectin in 845 men and 1110 women in 15 centres and also anthropometric measures of fatness including body mass index and waist/hip ratio, current asthma, and specific skin prick and IgE sensitisation. We used inverse sampling‐probability‐weighted rank and regression statistics to measure population associations of disease outcomes with adipokines in males and females, adjusting for confounders (area, age, smoking history, and number of elder siblings) and also mutually adjusting associations with adipokines and fatness measures.Results: One thousand nine hundred and fifty‐five subjects aged 16–77 years had information on leptin or adiponectin levels. Leptin and leptin/adiponectin ratio were positively associated with the level of asthma, especially in females (Somers’ D of leptin by asthma score, 0.20; 95% CI, 0.08–0.30; P = 0.00079). These associations were attenuated after adjusting for confounders and became non‐significant after additionally adjusting for fatness measures and multiple comparisons.Conclusions and Clinical Relevance: Asthma levels are positively associated with serum leptin. However, we cannot rule out the possibility that this association is secondary to associations of both with fatness measures.


  • Bonferroni and Holm Approximations for Šidák and Holland–Copenhaver q-Values

    by Roger B. Newson

    I describe the use of the Bonferroni and Holm formulas as approximations for Šidák and Holland–Copenhaver formulas when issues of precision are encountered, especially with q-values corresponding to very small p-values.


  • Sensible Parameters for Univariate and Multivariate Splines

    by Roger B. Newson

    The package bspline, downloadable from Statistical Software Components, now has three commands. The first, bspline, generates a basis of Schoenberg B-splines. The second, frencurv, generates a basis of reference splines whose parameters in the regression model are simply values of the spline at reference points on the X axis. The recent addition, flexcurv, is an easy-to-use version of frencurv that generates reference splines with automatically generated, sensibly spaced knots. frencurv and flexcurv now have the additional option of generating an incomplete basis of reference splines, with the reference spline for a baseline reference point omitted or set to 0. This incomplete basis can be completed by adding the standard unit vector to the design matrix and can then be used to estimate differences between values of the spline at the remaining reference points and the value of the spline at the baseline reference point. Reference splines therefore model continuous factor variables as indicator variables (or “dummies”) model discrete factor variables. The method can be extended in a similar way to define factor-product bases, allowing the user to estimate factor-combination means, subset-specific effects, or even factor interactions involving multiple continuous or discrete factors.


  • From Resultssets to Resultstables in Stata

    by Roger B. Newson

    The listtab package supersedes the listtex package. It inputs a list of variables and outputs them as a table in one of several formats, including TEX, LATEX, HTML, Microsoft Rich Text Format, or possibly future XML-based formats. It works with a team of four other packages: sdecode, chardef, xrewide, and ingap, which are downloaded from the Statistical Software Components archive. The sdecode package is an extension of decode; it allows the user to add prefixes or suffixes and to output exponents as TEX, HTML, or Rich Text Format superscripts. It is called by another package, bmjcip, to convert estimates, confidence limits, and p-values from numeric to string. The chardef package is an extension of char define; it allows the user to define a characteristic for a list of variables, adding prefixes or suffixes. The xrewide package is an extension of reshape wide; it allows the user to store additional results in variable characteristics or local macros. The ingap package inserts gap observations into a dataset to form gap rows when the dataset is converted to a table. Together, these packages form a general toolkit to convert Stata datasets (or resultssets) to tables, complete with row labels and column labels. Examples are demonstrated using LATEX and also using the rtfutil package to create Rich Text Format documents. This article uses data from the Avon Longitudinal Study of Parents and Children, based at Bristol University, UK.


  • Frequentist Q-values for Multiple-test Procedures

    by Roger B. Newson

    Multiple-test procedures are increasingly important as technology increases scientists’ ability to make large numbers of multiple measurements, as they do in genome scans. Multiple-test procedures were originally defined to input a vector of input p-values and an uncorrected critical p-value, interpreted as a familywise error rate or a false discovery rate, and to output a corrected critical p-value and a discovery set, defined as the subset of input p-values that are at or below the corrected critical p-value. A range of multiple-test procedures is implemented using the smileplot package in Stata (Newson and the ALSPAC Study Team 2003, Stata Journal 3: 109–132; 2010, Stata Journal 10: 691–692). The qqvalue command uses an alternative formulation of multiple-test procedures, which is also used by the R function p.adjust. qqvalue inputs a variable of p-values and outputs a variable of q-values that are equal in each observation to the minimum familywise error rate or false discovery rate that would result in the inclusion of the corresponding p-value in the discovery set if the specified multiple-test procedure was applied to the full set of input p-values. Formulas and examples are presented.


  • Open Access

    Frequentist Q-values for Multiple-test Procedures

    by Roger B. Newson

    Multiple-test procedures are increasingly important as technology increases scientists’ ability to make large numbers of multiple measurements, as they do in genome scans. Multiple-test procedures were originally defined to input a vector of input p-values and an uncorrected critical p-value, interpreted as a familywise error rate or a false discovery rate, and to output a corrected critical p-value and a discovery set, defined as the subset of input p-values that are at or below the corrected critical p-value. A range of multiple-test procedures is implemented using the smileplot package in Stata (Newson and the ALSPAC Study Team 2003, Stata Journal 3: 109–132; 2010, Stata Journal 10: 691–692). The qqvalue command uses an alternative formulation of multiple-test procedures, which is also used by the R function p.adjust. qqvalue inputs a variable of p-values and outputs a variable of q-values that are equal in each observation to the minimum familywise error rate or false discovery rate that would result in the inclusion of the corresponding p-value in the discovery set if the specified multiple-test procedure was applied to the full set of input p-values. Formulas and examples are presented.


  • Comparing the Predictive Powers of Survival Models Using Harrell's C or Somers’ D

    by Roger B. Newson

    Medical researchers frequently make statements that one model predicts survival better than another, and they are frequently challenged to provide rigorous statistical justification for those statements. Stata provides the estat concordance command to calculate the rank parameters Harrell's C and Somers’ D as measures of the ordinal predictive power of a model. However, no confidence limits or p-values are provided to compare the predictive power of distinct models. The somersd package, downloadable from Statistical Software Components, can provide such confidence intervals, but they should not be taken seriously if they are calculated in the dataset in which the model was fit. Methods are demonstrated for fitting alternative models to a training set of data, and then measuring and comparing their predictive powers by using out-of-sample prediction and somersd in a test set to produce statistically sensible confidence intervals and p-values for the differences between the predictive powers of different models.