Type: | Package |
Title: | Clinical Significance Measures |
Version: | 1.2 |
Author: | Mike Malek-Ahmadi <michael.malekahmadi@bannerhealth.com>, Kjera Schack <kgschack@asu.edu> |
Maintainer: | Mike Malek-Ahmadi <michael.malekahmadi@bannerhealth.com> |
Description: | Provides measures of effect sizes for summarized continuous variables as well as diagnostic accuracy statistics for 2x2 table data. Includes functions for Cohen's d, robust effect size, Cohen's q, partial eta-squared, coefficient of variation, odds ratio, likelihood ratios, sensitivity, specificity, positive and negative predictive values, Youden index, number needed to treat, number needed to diagnose, and predictive summary index. |
License: | GPL-3 |
Encoding: | UTF-8 |
RoxygenNote: | 7.3.1 |
NeedsCompilation: | no |
Packaged: | 2024-07-01 07:12:25 UTC; mikem |
Repository: | CRAN |
Date/Publication: | 2024-07-01 07:30:01 UTC |
Cohen's d Calculation
Description
Calculates a Cohen's d effect size using the means and standard deviations of two independent groups
Usage
cohens_d(Group1_Mean, Group1_SD, Group2_Mean, Group2_SD)
Arguments
Group1_Mean |
Mean for Group 1 |
Group1_SD |
Standard Deviation for Group 1 |
Group2_Mean |
Mean for Group 2 |
Group2_SD |
Standard Deviation for Group 2 |
Value
A single value representing the Cohen's d effect size
Author(s)
Mike Malek-Ahmadi
References
1. Cohen, Jacob (1988). Statistical Power Analysis for the Behavioral Sciences. Routledge. ISBN 978-1-134-74270-7.
2. Malek-Ahmadi M, Perez SE, Chen K, Mufson EJ. Neuritic and diffuse plaque associations with memory in non-cognitively impaired elderly. J Alzheimers Dis 2016;53(4):1641-1652.
Examples
#From Table 2 in Malek-Ahmadi et al (2016)
#comparing groups with (0.75+/-0.35) and without (0.49+/-0.29) neuritic plaques
#on a global cognitive score (z-score).
cohens_d(0.75, 0.35, 0.49, 0.29)
Cohen's q Calculation
Description
Calculates Cohen's q for the effect size of the difference between two correlation values
Usage
cohens_q(corr1, corr2)
Arguments
corr1 |
Correlation for First Group |
corr2 |
Correlation for Second Group |
Value
A single value representing Cohen's q
Author(s)
Mike Malek-Ahmadi
References
1. Cohen, Jacob (1988). Statistical Power Analysis for the Behavioral Sciences. Routledge. ISBN 978-1-134-74270-7.
2. Yang G, Li D, Rao Y, Lu F. The relationship between cortical thickness and language comprehension varies with sex in healthy young adults: a large sample analysis. Neuroreport 2020;31(2):184-188.
Examples
#From Yang et al (2020), Cohen's q for the difference between female and male correlation
#values for vocabulary comprehension and cortical thickness.
cohens_q (0.318, 0.174)
Coefficient of Variation Calculation
Description
Calculates the coefficient of variation for a mean and standard deviation
Usage
cv(Mean, SD)
Arguments
Mean |
Mean for a dataset |
SD |
Standard Deviation for a dataset |
Value
A single value representing the Coefficient of Variation
Author(s)
Mike Malek-Ahmadi
References
1. Everitt B (1998). The Cambridge Dictionary of Statistics. Cambridge, UK New York: Cambridge University Press. ISBN 978-0521593465.
2. Bedeian AG, Mossholder KW. On the use of the coefficient of variation as a measure of diversity. Organizational Research Methods 2000;3(3):285-297.
Examples
#From Bedeian & Mossholder (2000), Table 2 Group A data.
cv(28, 7)
Likelihood Ratio Negative Calculation From a 2x2 Table
Description
Calculates diagnostic test likelihood ratio negative and 95 percent confidence intervals for data from a 2x2 table
Usage
lr_neg(Cell1, Cell2, Cell3, Cell4)
Arguments
Cell1 |
Value for cases with a positive test |
Cell2 |
Value for controls with a positive test |
Cell3 |
Value for cases with a negative test |
Cell4 |
Value for controls with a negative test |
Value
Likelihood Ratio Negative and 95 percent confidence intervals
Author(s)
Mike Malek-Ahmadi
References
1. Grimes DA, Schultz KF. Refining clinical diagnosis with likelihood ratios. Lancet 2005;365:1500-1505.
2. Dujardin B, Van den Ende J, Van Gompel A, Unger JP, Van der Stuyft P. Likelihood ratios: a real improvement for clinical decision making? European Journal of Epidemiology 1994 Feb;10(1):29-36.
Examples
#From Table 1 in Dujardin et al (1994)
lr_neg(72, 9, 25, 137)
Likelihood Ratio Positive Calculation From a 2x2 Table
Description
Calculates diagnostic test likelihood ratio positive and 95 percent confidence intervals for data from a 2x2 table
Usage
lr_pos(Cell1, Cell2, Cell3, Cell4)
Arguments
Cell1 |
Value for cases with a positive test |
Cell2 |
Value for controls with a positive test |
Cell3 |
Value for cases with a negative test |
Cell4 |
Value for controls with a negative test |
Value
Likelihood Ratio Positive and 95 percent confidence intervals
Author(s)
Mike Malek-Ahmadi
References
1. Grimes DA, Schultz KF. Refining clinical diagnosis with likelihood ratios. Lancet 2005;365:1500-1505.
2. Dujardin B, Van den Ende J, Van Gompel A, Unger JP, Van der Stuyft P. Likelihood ratios: a real improvement for clinical decision making? European Journal of Epidemiology 1994 Feb;10(1):29-36.
Examples
#From Table 1 in Dujardin et al (1994)
lr_pos(72, 9, 25, 137)
Number Needed to Diagnose Calculation From a 2x2 Table
Description
Calculates the Number Needed to Diagnose for data from a 2x2 table
Usage
nnd(Cell1, Cell2, Cell3, Cell4)
Arguments
Cell1 |
Value for cases with a positive test |
Cell2 |
Value for controls with a positive test |
Cell3 |
Value for cases with a negative test |
Cell4 |
Value for controls with a negative test |
Value
Number Needed to Diagnose
Author(s)
Mike Malek-Ahmadi
References
1. Larner AJ. Number Needed to Diagnose, Predict, or Misdiagnose: Useful Metrics for Non-Canonical Signs of Cognitive Status? Dement Geriatr Cogn Disord Extra 2018;8:321–327
Examples
#From Shaikh (2011), page 3, 2x2 table for "Diagnostic Test Evaluation"
#NND is the inverse of the Youden Index (1 / Youden Index)
nnd(105, 171, 15, 87)
Number Needed to Treat Calculation From a 2x2 Table
Description
Calculates number needed to treat and 95 percent confidence intervals for data from a 2x2 table
Usage
nnt(Cell1, Cell2, Cell3, Cell4)
Arguments
Cell1 |
Value for cases with a positive outcome |
Cell2 |
Value for cases with a negative outcome |
Cell3 |
Value for controls with a positive outcome |
Cell4 |
Value for controls with a negative outcome |
Value
Number Needed to Treat and 95 percent confidence intervals
Author(s)
Mike Malek-Ahmadi
References
1. Cook RJ, Sackett DL. The number needed to treat: a clinically useful measure of treatment effect [published correction appears in BMJ 1995 Apr 22;310(6986):1056]. BMJ. 1995;310(6977):452-454.
2. Zar HJ, Cotton MF, Strauss S et al Effect of isoniazid prophylaxi on mortality of tuberculosis in children with HIV: randomised controlled trial. BMJ 2007; 136-9.
Examples
#Mortality data from Zar et al (2007)
nnt(121, 11, 110, 21)
Negative Predictive Value Calculation From a 2x2 Table
Description
Calculates diagnostic test negative predictive value and 95 percent confidence intervals for data from a 2x2 table
Usage
npv(Cell1, Cell2, Cell3, Cell4)
Arguments
Cell1 |
Value for cases with a positive test |
Cell2 |
Value for controls with a positive test |
Cell3 |
Value for cases with a negative test |
Cell4 |
Value for controls with a negative test |
Value
Negative Predictive Value and 95 percent confidence intervals
Author(s)
Mike Malek-Ahmadi
References
1. Trevethan R. Sensitivity, specificity, and predictive values: Foundations, pliabilities, and pitfalls in research and practice. Frontiers in Public Health 2017;5:307.
2. Safari S, Baratloo A, Elfil M, Negida A. Evidence Based Emergency Medicine Part 2: Positive and negative predictive values of diagnostic tests. Emerg (Tehran) 2015;3(3):87-88.
Examples
#From Figure 2 in Safari et al (2015)
npv(15, 6, 25, 34)
Odds Ratio Calculation From a 2x2 Table
Description
Calculates an odds ratio and 95 percent confidence intervals for data from a 2x2 table
Usage
odds_ratio(Cell1, Cell2, Cell3, Cell4)
Arguments
Cell1 |
Value for cases with the factor/exposure of interest |
Cell2 |
Value for cases without the factor/exposure of interest |
Cell3 |
Value for controls with the factor/exposure of interest |
Cell4 |
Value for controls without the factor/exposure of interest |
Value
Odds ratio and 95 percent confidence intervals
Author(s)
Mike Malek-Ahmadi
References
1.Mufson EJ, Malek-Ahmadi M, Perez SE, Chen K. Braak staging, plaque pathology, and APOE status in elderly persons without cognitive impairment. Neurobiol Aging 2016;37:147-153.
Examples
# From Table 1 in Mufson et al (2016), using data for sex (Male/Female)
#and Braak stage group classification (I-II/III-V).
#Female/Braak III-V = 46, Female/Braak I-II = 14, Male/Braak III-V = 32,
#Male/Braak I-II = 31.
odds_ratio(46, 14, 32, 31)
Partial Eta Squared Calculation
Description
Calculates partial eta squared effect size for ANOVAs
Usage
partial_eta_sq(SS.Between, SS.Error)
Arguments
SS.Between |
Sum of Squares Between from ANOVA Output |
SS.Error |
Sum of Squares Error from ANOVA Output |
Value
A single value representing partial eta squared
Author(s)
Mike Malek-Ahmadi
References
1. Levine TR, Hullett CR. Eta squared, partial eta squared, and misreporting of effect size in communication research. Human Communication Research 2002;28:612-625.
Examples
#From Levine & Hullett (2002), Example 1 in Table 1
partial_eta_sq(2500, 800)
Positive Predictive Value Calculation From a 2x2 Table
Description
Calculates diagnostic test positive predictive value and 95 percent confidence intervals for data from a 2x2 table
Usage
ppv(Cell1, Cell2, Cell3, Cell4)
Arguments
Cell1 |
Value for cases with a positive test |
Cell2 |
Value for controls with a positive test |
Cell3 |
Value for cases with a negative test |
Cell4 |
Value for controls with a negative test |
Value
Positive Predictive Value and 95 percent confidence intervals
Author(s)
Mike Malek-Ahmadi
References
1. Trevethan R. Sensitivity, specificity, and predictive values: Foundations, pliabilities, and pitfalls in research and practice. Frontiers in Public Health 2017;5:307.
2. Safari S, Baratloo A, Elfil M, Negida A. Evidence Based Emergency Medicine Part 2: Positive and negative predictive values of diagnostic tests. Emerg (Tehran) 2015;3(3):87-88.
Examples
#From Figure 2 in Safari et al (2015)
ppv(15, 6, 25, 34)
Predictive Summary Index Calculation From a 2x2 Table
Description
Calculates the Predictive Summary Index for data from a 2x2 table
Usage
psi(Cell1, Cell2, Cell3, Cell4)
Arguments
Cell1 |
Value for cases with a positive test |
Cell2 |
Value for controls with a positive test |
Cell3 |
Value for cases with a negative test |
Cell4 |
Value for controls with a negative test |
Value
Predictive Summary Index
Author(s)
Mike Malek-Ahmadi
References
1. Linn S, Grunau PD. New patient-oriented summary measure of net total gain in certainty for dichotomous diagnostic tests. Epidemiol Perspect Innov 2006;3:11.
2. Shaikh SA. Measures Derived from a 2 x 2 Table for an Accuracy of a Diagnostic Test. J Biomet Biostat 2011, 2:5
Examples
#From Shaikh (2011), page 3, 2x2 table for "Diagnostic Test Evaluation"
psi(105, 171, 15, 87)
Robust effect size for comparison of means between two groups
Description
Calculates the robust effect size for a two-group comparison using the means, standard deviations, and sample sizes for each group
Usage
robust_effect_size(M1, M2, SD1, SD2, N1, N2)
Arguments
M1 |
Mean for Group 1 |
M2 |
Mean for Group 2 |
SD1 |
Standard deviation for Group 1 |
SD2 |
Standard deviation for Group 2 |
N1 |
Sample Size for Group 1 |
N2 |
Sample Size for Group 2 |
Value
Robust Effect Size
Author(s)
Kjera Schack
References
Vandekar S, Tao R, Blume J. A Robust Effect Size Index [published correction appears in Psychometrika. 2020 Dec;85(4):946]. Psychometrika. 2020;85(1):232-246. doi:10.1007/s11336-020-09698-2
Examples
#From Table 2 in Malek-Ahmadi et al (2016)
#comparing groups with (0.75+/-0.35, n=45) and without (0.49+/-0.29, n=78) neuritic plaques
#on a global cognitive score (z-score).
robust_effect_size(0.75, 0.49, 0.35, 0.29, 45, 78)
Sensitivity Calculation From a 2x2 Table
Description
Calculates diagnostic test sensitivity and 95 percent confidence intervals for data from a 2x2 table
Usage
sensitivity(Cell1, Cell2, Cell3, Cell4)
Arguments
Cell1 |
Value for cases with a positive test |
Cell2 |
Value for controls with a positive test |
Cell3 |
Value for cases with a negative test |
Cell4 |
Value for controls with a negative test |
Value
Sensitivity and 95 percent confidence intervals
Author(s)
Mike Malek-Ahmadi
References
1. Trevethan R. Sensitivity, specificity, and predictive values: Foundations, pliabilities, and pitfalls in research and practice. Frontiers in Public Health 2017;5:307.
2. Weissberger GH, Strong JV, Stefanidis KB, Summers MJ, Bondi MW, Stricker NH. Diagnostic accuracy of memory measures in Alzheimer's dementia and mild Cognitive Impairment: a Systematic Review and Meta-Analysis. Neuropsychol Rev. 2017;27(4):354-388.
Examples
#Sensitivity calculation from Figure 11, Line 22 of Weissberger et al
sensitivity (121, 50, 13, 199)
Specificity Calculation From a 2x2 Table
Description
Calculates diagnostic test specificity and 95 percent confidence intervals for data from a 2x2 table
Usage
specificity(Cell1, Cell2, Cell3, Cell4)
Arguments
Cell1 |
Value for cases with a positive test |
Cell2 |
Value for controls with a positive test |
Cell3 |
Value for cases with a negative test |
Cell4 |
Value for controls with a negative test |
Value
Specificity and 95 percent confidence intervals
Author(s)
Mike Malek-Ahmadi
References
1. Trevethan R. Sensitivity, specificity, and predictive values: Foundations, pliabilities, and pitfalls in research and practice. Frontiers in Public Health 2017;5:307.
2. Weissberger GH, Strong JV, Stefanidis KB, Summers MJ, Bondi MW, Stricker NH. Diagnostic accuracy of memory measures in Alzheimer's dementia and mild Cognitive Impairment: a Systematic Review and Meta-Analysis. Neuropsychol Rev. 2017;27(4):354-388.
Examples
#Specificity calculation from Figure 11, Line 22 of Weissberger et al
specificity (121, 50, 13, 199)
Youden Index Calculation From a 2x2 Table
Description
Calculates the Youden Index for data from a 2x2 table
Usage
youden_index(Cell1, Cell2, Cell3, Cell4)
Arguments
Cell1 |
Value for cases with a positive test |
Cell2 |
Value for controls with a positive test |
Cell3 |
Value for cases with a negative test |
Cell4 |
Value for controls with a negative test |
Value
Youden Index
Author(s)
Mike Malek-Ahmadi
References
1. Ruopp MD, Perkins NJ, Whitcomb BW, Schisterman EF. Youden Index and optimal cut-point estimated from observations affected by a lower limit of detection. Biom J 2008;50(3):419-430.
2. Shaikh SA (2011) Measures derived from a 2 x 2 table for an accuracy of a diagnostic test. J Biomet Biostat 2:128
Examples
#From Shaikh (2011), page 3, 2x2 table for "Diagnostic Test Evaluation"
youden_index(105, 171, 15, 87)