****************
****************
** cs ** Cohort study summaries
****************
****************
Overview: The "cs" command is used to obtain summaries and inference
for cohort or cross-sectional designs.
Usage: "cs Dvar Evar"
Where - Dvar is the disease variable (1=disease, 0=control)
Evar is the exposure variable (1=exposed, 0=unexposed)
Note that the display for "cs" is with disease groups forming
rows and exposure groups forming columns.
Summaries: The "cs" command returns point estimates and confidence intervals
for the risk difference, risk ratio, and (optionally) the odds
ratio. In addition, attributable or prevented fractions for
the exposed and total population is returned.
risk difference = p1 - p0, the difference between the probability
of disease among the exposed minus the probability of disease
among the unexposed.
risk ratio = p1/p0, the ratio of the probability
of disease among the exposed relative to the probability
of disease among the unexposed.
odds ratio = [p1/(1-p1)]/[p0/(1-p0)], the ratio of the odds
of disease among the exposed relative to the odds
of disease among the unexposed.
attributable fraction for the exposed = the percent of cases
among the exposed that could be eliminated if exposure was
eliminated. Computed as (p1-p0)/p1 = [RR-1]/RR.
attributable fraction for population = the percent of cases
within the population that could be eliminated if exposure was
eliminated. Let N1 be the number of exposed, and N0 be the
number of unexposed. We would remove N1*(p1-p0)/p1 cases by
removing exposure. The fraction of total cases would be this
amount relative to the total number of cases, N1+N0. Therefore,
the net effect in the population is [N1*(p1-p0)/p1]/(N1+N0) =
(percent exposed among cases)*(attr fraction exposed).
Options: (1) "cs Dvar Evar [fweight=freq]" -- this is used when the
data are in a "grouped" format with the number of
cases/controls that have a certain covariate combination
given by the variable "freq". See notes pg 52 for cc example.
(2) "cs Dvar Evar, exact" -- the exact option will compute
significance using Fisher's exact test.
(3) "cs Dvar Evar, by(Svar)" -- specifies that 2x2 tables
are to be formed for the strata given by "Svar". Within
stratum statistics are shown and then combined with
Mantel-Haenszel weights.
(4) "cc Dvar Evar, by(Svar) bd" -- the addition of "bd"
specifies that the Breslow-Day test of homogeneity of
risk ratios across the strata be calculated.
****************
****************
** cc ** Case-control summaries
****************
****************
Overview: The "cc" command is used to obtain summaries and inference
for case-control or cross-sectional designs.
Summaries: The "cc" command returns point estimates and confidence intervals
for the the odds ratio and attributable or prevented fractions for
the exposed and total population.
Usage: "cc Dvar Evar"
Where - Dvar is the disease variable (1=disease, 0=control)
Evar is the exposure variable (1=exposed, 0=unexposed)
Note that the display for "cc" is with disease groups forming
rows and exposure groups forming columns.
Options: (1) "cc Dvar Evar [fweight=freq]" -- this is used when the
data are in a "grouped" format with the number of
cases/controls that have a certain covariate combination
given by the variable "freq". See notes pg 52.
(2) "cc Dvar Evar, exact" -- the exact option will compute
significance using Fisher's exact test.
(3) "cc Dvar Evar, by(Svar)" -- specifies that 2x2 tables
are to be formed for the strata given by "Svar". Within
stratum statistics are shown and then combined with
Mantel-Haenszel weights.
(4) "cc Dvar Evar, by(Svar) bd" -- the addition of "bd"
specifies that the Breslow-Day test of homogeneity of
odds ratios across the strata be calculated.
****************
****************
** mcc ** Matched case-control summaries
****************
****************
Overview: The "mcc" command is used to obtain summaries and inference
for matched case-control or cross-sectional designs.
Summaries: The "mcc" command returns McNemar's chi-square test, point
estimates and confidence intervals for the difference, ratio,
and relative difference of the proportion with the exposure,
along with the odds ratio.
Usage: "mcc Evar1 Evar0"
Where - Evar1 is the exposure variable (1=exposed, 0=unexposed)
for the cases.
Evar0 is the exposure variable (1=exposed, 0=unexposed)
for the controls.
Options: (1) "mcc Evar1 Evar0 [fweight=freq]" -- this is used when the
data are in a "grouped" format with the number of
cases/controls that have a certain covariate combination
given by the variable "freq". See notes pg 52.
(2) Note: there is no exact option since the default ouput
also calculates exact binomial p-values.
(3) Note: the by(Svar) option is not available. To obtain
separate analyses one need to use
"mcc Evar1 Evar0 if Svar==1"