A simple method of sample size calculation for linear and logistic regression.
ABSTRACT A sample size calculation for logistic regression involves complicated formulae. This paper suggests use of sample size formulae for comparing means or for comparing proportions in order to calculate the required sample size for a simple logistic regression model. One can then adjust the required sample size for a multiple logistic regression model by a variance inflation factor. This method requires no assumption of low response probability in the logistic model as in a previous publication. One can similarly calculate the sample size for linear regression models. This paper also compares the accuracy of some existing sample-size software for logistic regression with computer power simulations. An example illustrates the methods.
SourceAvailable from: PubMed Central[Show abstract] [Hide abstract]
ABSTRACT: Commonly when designing studies, researchers propose to measure several independent variables in a regression model, a subset of which are identified as the main variables of interest while the rest are retained in a model as covariates or confounders. Power for linear regression in this setting can be calculated using SAS PROC POWER. There exists a void in estimating power for the logistic regression models in the same setting. Currently, an approach that calculates power for only one variable of interest in the presence of other covariates for logistic regression is in common use and works well for this special case. In this paper we propose three related algorithms along with corresponding SAS macros that extend power estimation for one or more primary variables of interest in the presence of some confounders. The three proposed empirical algorithms employ likelihood ratio test to provide a user with either a power estimate for a given sample size, a quick sample size estimate for a given power, and an approximate power curve for a range of sample sizes. A user can specify odds ratios for a combination of binary, uniform and standard normal independent variables of interest, and or remaining covariates/confounders in the model, along with a correlation between variables. These user friendly algorithms and macro tools are a promising solution that can fill the void for estimation of power for logistic regression when multiple independent variables are of interest, in the presence of additional covariates in the model.Source Code for Biology and Medicine 11/2014; 9:24. DOI:10.1186/1751-0473-9-24
[Show abstract] [Hide abstract]
ABSTRACT: In November 2014, a review of literature concerning prevalence data of Molar Incisor Hypomineralisation (MIH) and Hypomineralised Second Primary Molars (HSPM) was performed. A search of PubMed online databases was conducted for relevant articles published until November 2014. The reference lists of all retrieved articles were hand-searched. Studies were included after assessing the eligibility of the full-text article. Out of 1078 manuscripts, a total of 157 English written publications were selected based on title and abstract. Of these 157, 60 were included in the study and allocated as 52 MIH and 5 HSPM, and 3 for both MIH and HSPM. These studies utilised the European Academy of Paediatric Dentistry judgment criteria, the modified index of developmental defects of enamel (mDDE) and self-devised criteria, and demonstrated a wide variation in the reported prevalence (MIH 2.9-44 %; HSPM 0-21.8 %). Most values mentioned were representative for specific areas. More studies were performed in cities compared with rural areas. A great variation was found in calibration methods, number of participants, number of examiners and research protocols between the studies. The majority of the prevalence studies also investigated possible aetiological factors. To compare MIH and HSPM prevalence and or aetiological data around the world, standardisation of such studies seems essential. Standardisation of the research protocol should include a clearly described sample of children (minimum number of 300 for prevalence and 1000 for aetiology studies) and use of the same calibration sets and methods whereas aetiological studies need to be prospective in nature. A standardised protocol for future MIH and HSPM prevalence and aetiology studies is recommended.European Archives of Paediatric Dentistry. Official Journal of the European Academy of Paediatric Dentistry 04/2015; DOI:10.1007/s40368-015-0179-7
[Show abstract] [Hide abstract]
ABSTRACT: Emergence agitation after intracranial surgery is an important clinical issue during anaesthesia recovery. The aim of this multicentre cohort study is to investigate the incidence of emergence agitation, identify the risk factors and determine clinical outcomes in adult patients after intracranial surgery under general anaesthesia. Additionally, we will deliberately clarify the relationship between postoperative pneumocephalus and agitation. The present study is a prospective multicentre cohort study. Five intensive care units (ICUs) in China will participate in the study. Consecutive adult patients admitted to the ICUs after intracranial surgery will be enrolled. Sedation-Agitation Scale (SAS) or Richmond Agitation-Sedation Scale (RASS) will be used to evaluate the patients 12 h after the enrolment. Agitation is defined as an SAS score of 5-7, or an RASS score of +2 to +4. According to the maximal SAS and RASS score, patients will be divided into two cohorts: the agitation group and the non-agitation group. Factors potentially related to emergence agitation will be collected at study entry, during anaesthesia and operation, during postoperative care. Univariate analyses between the agitation and the non-agitation groups will be performed. The stepwise backward logistic regression will be carried out to identify the independent predictors of agitation. Patients will be followed up for 72 h after the operation. Accidental self-extubation of the endotracheal tube and removal of other catheters will be documented. The use of sedatives and analgesics will be collected. Ethics approval has been obtained from each of five participating hospitals. Study findings will be disseminated through peer-reviewed publications and conference presentations. NCT02318199. Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://group.bmj.com/group/rights-licensing/permissions.BMJ Open 04/2015; 5(4):e007542. DOI:10.1136/bmjopen-2014-007542 · 2.06 Impact Factor