Article

A simple method of sample size calculation for linear and logistic regression.

CSPCC, Department of Veterans Affairs, Palo Alto Health Care System (151-K), California 94304, USA.
Statistics in Medicine (Impact Factor: 2.04). 08/1998; 17(14):1623-34. DOI: 10.1002/(SICI)1097-0258(19980730)17:143.0.CO;2-S
Source: PubMed

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.

3 Bookmarks
 · 
1,164 Views
  • [Show abstract] [Hide abstract]
    ABSTRACT: Assessing prospective program reach is an important step prior to any online-development effort. This study assessed, preliminarily, interest in a proposed online coping skills training program for women living with a partner with a drinking problem, including their (a) overall program interest; (b) relative preference for an online versus face-to-face format; (c) likely use of adjunctive, professional communications; and (d) use under alternate fee structures. In a community sample of women with problem-drinking partners (N = 73), interest was high (73%); 46% preferred the online format; 31% preferred online and face-to-face formats equally. Interest in adjunctive contacts was high, but diminished with fees; interest-level predictors were few. Online skills training could potentially reach a large portion of this frequently hidden, underserved population, and be financially feasible.
    Journal of Technology in Human Services 07/2013; 31(3):185-196.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: An episode of breakthrough bacteremia, which was defined as positive blood cultures despite appropriate antibiotic therapy, imposes a treatment challenge in the neonatal intensive care unit (NICU). All episodes of breakthrough bacteremia from a tertiary level NICU in Taiwan between 2004 and 2011 were analyzed and compared with nonbreakthrough bacteremia. Breakthrough bacteremia was identified in 7.6% (72/942) of neonatal bacteremia, and 43 (59.7%) occurred as recurrent episodes. Gram-negative organisms (41.7%) and fungi (15.3%) accounted for more than half of all microorganisms in breakthrough bacteremia. Compared with nonbreakthrough bacteremia, breakthrough bacteremia was significantly associated with more severe disease, was more likely to require aggressive therapies, and had a higher rate of infectious complications. Previous use of broad-spectrum antibiotics (odds ratio [OR], 7.54; P < .001) and particular microbial etiologies (Pseudomonas aeruginosa: OR, 4.40; P = .025; fungi: OR, 2.70; P = .013) were independent risk factors for developing breakthrough bacteremia. The crude sepsis-attributable mortality rate was greater in breakthrough bacteremia episodes (16.7% vs 6.4%; P = .004), and this condition was independently associated with an increased risk of death (OR, 2.14; 95% confidence interval, 1.04-4.40; P = .040). Breakthrough bacteremia is not uncommon (7.6% of all bacteremia) in NICUs and represents a more severe form of neonatal bacteremia that is independently associated with an increased risk of death. Copyright © 2015 Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights reserved.
    American Journal of Infection Control 01/2015; 43(1):20-5. · 2.33 Impact Factor
  • Source
    Psychological Trauma Theory Research Practice and Policy 01/2015; 7(1):93-101. · 0.89 Impact Factor

Preview

Download
2,695 Downloads
Available from