metabias                package:meta                R Documentation

_T_e_s_t _f_o_r _f_u_n_n_e_l _p_l_o_t _a_s_y_m_m_e_t_r_y

_D_e_s_c_r_i_p_t_i_o_n:

     Test for funnel plot asymmetry, based on rank correlation or
     linear regression method.

_U_s_a_g_e:

     metabias(x, seTE, TE.fixed, seTE.fixed,
              method = "rank",
              plotit = FALSE, correct = FALSE)

_A_r_g_u_m_e_n_t_s:

       x: An object of class 'meta', or estimated treatment effect in
          individual studies.

    seTE: Standard error of estimated treatment effect (mandatory if
          'x' not of class 'meta').

TE.fixed: Overall treatment estimate (mandatory if 'x' not of class
          'meta' and 'method' = '"rank"').

seTE.fixed: Standard error of overall treatment estimate (mandatory if
          'x' not of class 'meta' and 'method' = '"rank"').

  method: A character string indicating which test is to be used.
          Either '"rank"', '"linreg"', '"mm"' or '"count"', can be
          abbreviated.

  plotit: A logical indicating whether a plot should be produced for
          method '"rank"', '"linreg"' or '"mm"'.

 correct: A logical indicating whether a continuity corrected statistic
          is used for rank correlation methods '"rank"' and '"count"'.

_D_e_t_a_i_l_s:

     If 'method' is '"rank"', the test statistic is based on the rank
     correlation between standardised treatment estimates and variance
     estimates of estimated treatment effects; Kendall's tau is used as
     correlation measure (Begg & Mazumdar, 1994). The test statistic
     follows a standard normal distribution. By default (if 'correct'
     is FALSE), no continuity correction is utilised (Kendall &
     Gibbons, 1990).

     If 'method' is '"linreg"', the test statistic is based on a linear
     regression of the standardised treatment effect (standard normal
     deviate) on the inverse of the standard error of the treatment
     estimate (Egger et al., 1997). The test statistic follows a t
     distribution with 'number of studies - 2' degrees of freedom.

     If 'method' is '"mm"', the test statistic is based on a weighted
     linear regression using the method of moments estimator of the
     additive between-study variance component (method 3a in Thompson,
     Sharp, 1999). The test statistic follows a t distribution with
     'number of studies - 2' degrees of freedom.

     If 'method' is '"count"', the test statistic is based on the rank
     correlation between a standardised cell frequency and the inverse
     of the variance of the cell frequency; Kendall's tau is used as
     correlation measure (Schwarzer, 2003). The test statistic follows
     a standard normal distribution. By default (if 'correct' is
     FALSE), no continuity correction is utilised (Kendall & Gibbons,
     1990).

_V_a_l_u_e:

     A list with class '"htest"' containing the following components: 

estimate: the estimated degree of funnel plot asymmetry, with name
          '"ks"' or '"bias"' corresponding to the method employed,
          i.e., rank correlation or regression method.

statistic: The value of the test statistic.

parameter: The degrees of freedom of the test statistic in the case
          that it follows a t distribution.

 p.value: The p-value for the test.

null.value: The value of test statistic under the null hypothesis,
          always '0'.

alternative: A character string describing the alternative hypothesis.

  method: A character string indicating what type of test was used.

data.name: A character string giving the names of the data.

_A_u_t_h_o_r(_s):

     Guido Schwarzer sc@imbi.uni-freiburg.de

_R_e_f_e_r_e_n_c_e_s:

     Begg CB & Berlin JA (1994), Operating characteristics of a rank
     correlation test for publication bias. _Biometrics_, *50*,
     1088-1101.

     Kendall M & Gibbons JD (1990), _Rank Correlation Methods_. London:
     Edward Arnold.

     Egger M, Smith GD, Schneider M & Minder C (1997), Bias in
     meta-analysis detected by a simple, graphical test. _British
     Medical Journal_, *315*, 629-634.

     Schwarzer G (2003), Statistical Tests for Bias in Meta-Analysis
     with Binary Outcomes, PhD thesis, University of Dortmund, Germany,
     http://eldorado.uni-dortmund.de

     Thompson SG, Sharp, SJ (1999), Explaining heterogeneity in
     meta-analysis: A comparison of methods, _Statistics in Medicine_,
     *18*, 2693-2708.

_S_e_e _A_l_s_o:

     'funnel', 'metabin', 'metacont', 'metagen'

_E_x_a_m_p_l_e_s:

     data(Olkin95)
     meta1 <- metabin(event.e, n.e, event.c, n.c,
                      data=Olkin95, subset=c(41,47,51,59),
                      sm="RR", meth="I")

     metabias(meta1)
     metabias(meta1, correct=TRUE)

     metabias(meta1, method="linreg")
     metabias(meta1, method="linreg", plotit=TRUE)

     metabias(meta1, method="count")

     ##
     ## Same result:
     ##
     metabias(meta1, method="linreg")$p.value
     metabias(meta1$TE, meta1$seTE, method="linreg")$p.value

