Using these data in future analyses
The data are provided as part of the packaged analysis.
# A tibble: 6 × 13
outcome intervention class moderator study control intervention_me…
<chr> <chr> <chr> <chr> <chr> <chr> <dbl>
1 count omega-3 fatt… omeg… varicoce… abba… placeb… 81.6
2 count vitamin d3 vita… subferti… amin… placeb… 88.3
3 count vitamin/mine… vita… varicoce… zade… placeb… 41.3
4 count zinc/zinc + … mine… varicoce… aziz… placeb… 47.6
5 count zinc/zinc + … mine… varicoce… aziz… placeb… 39.6
6 count folic acid vita… varicoce… aziz… placeb… 49.1
# … with 6 more variables: intervention_sd <dbl>,
# intervention_n <dbl>, control_mean <dbl>, control_sd <dbl>,
# control_n <dbl>, mean_difference <dbl>
# use devtools to install a package from github repository
install.packages("devtools")
# install packaged analysis
devtools::install_github("softloud/spunk")
# load packaged analysis
library(spunk)
# load data
data(spunk_dat)
# columns in data
names(spunk_dat)
[1] "outcome" "intervention" "class"
[4] "moderator" "study" "control"
[7] "intervention_mean" "intervention_sd" "intervention_n"
[10] "control_mean" "control_sd" "control_n"
[13] "mean_difference"
# description of each column
?spunk_dat
The analyses provided in the associated Cochrane review are summaries and overviews. These data provide a means of doing several different future analyses.
Researchers may be interested in investigating one specific intervention from the various interventions in the dataset. Here we provide two tables with summary information about interventions in the first, and class in the second.
See Summary tables in Living analyses for detailed information about the interventions and classes included in this analysis.
These data can be readily accessed by filtering by intervention using
tidyverse::
.
library(tidyverse)
conflicted::conflict_prefer("filter", "dplyr")
spunk_dat %>%
filter(intervention == "folic acid")
# A tibble: 13 × 13
outcome intervention class moderator study control intervention_me…
<chr> <chr> <chr> <chr> <chr> <chr> <dbl>
1 count folic acid vita… varicoce… aziz… placeb… 49.1
2 count folic acid vita… subferti… silv… placeb… 26.1
3 count folic acid vita… oligoast… raig… placeb… 16.3
4 count folic acid vita… subferti… wong… placeb… 50.2
5 volume folic acid vita… subferti… wong… placeb… 3.3
6 motili… folic acid vita… varicoce… aziz… placeb… 51.5
7 motili… folic acid vita… subferti… silv… placeb… 69.6
8 motili… folic acid vita… oligoast… raig… placeb… 33.2
9 motili… folic acid vita… subferti… wong… placeb… 35
10 morpho… folic acid vita… varicoce… aziz… placeb… 53.7
11 morpho… folic acid vita… subferti… silv… placeb… 23.9
12 morpho… folic acid vita… oligoast… raig… placeb… 6
13 morpho… folic acid vita… subferti… wong… placeb… 2.28
# … with 6 more variables: intervention_sd <dbl>,
# intervention_n <dbl>, control_mean <dbl>, control_sd <dbl>,
# control_n <dbl>, mean_difference <dbl>
Filters can be combined. For example, we might only interested in studies that reported morphology comparisons with folic acid and placebo.
spunk_dat %>%
filter(
outcome == "morphology",
intervention == "folic acid",
str_detect(control, "placebo")
)
# A tibble: 4 × 13
outcome intervention class moderator study control intervention_me…
<chr> <chr> <chr> <chr> <chr> <chr> <dbl>
1 morphol… folic acid vita… varicoce… aziz… placeb… 53.7
2 morphol… folic acid vita… subferti… silv… placeb… 23.9
3 morphol… folic acid vita… oligoast… raig… placeb… 6
4 morphol… folic acid vita… subferti… wong… placeb… 2.28
# … with 6 more variables: intervention_sd <dbl>,
# intervention_n <dbl>, control_mean <dbl>, control_sd <dbl>,
# control_n <dbl>, mean_difference <dbl>
Or perhaps researchers are interested in all studies that compared coenzyme-q10 with placebo, but allow for combined interventions.
spunk_dat %>%
filter(
str_detect(intervention, "coenzyme-q10")
)
# A tibble: 4 × 13
outcome intervention class moderator study control intervention_me…
<chr> <chr> <chr> <chr> <chr> <chr> <dbl>
1 count coenzyme-q1… diet… oligoast… gopi… placeb… 33.2
2 count coenzyme-q1… diet… oligoast… gopi… placeb… 31.6
3 motility coenzyme-q1… diet… oligoast… gopi… placeb… 57.4
4 motility coenzyme-q1… diet… oligoast… gopi… placeb… 55.8
# … with 6 more variables: intervention_sd <dbl>,
# intervention_n <dbl>, control_mean <dbl>, control_sd <dbl>,
# control_n <dbl>, mean_difference <dbl>
Class-level analysis data can be extracted in a similar manner.
# A tibble: 15 × 13
outcome intervention class moderator study control intervention_me…
<chr> <chr> <chr> <chr> <chr> <chr> <dbl>
1 count coenzyme q10 coen… asthenoz… bale… placeb… 44.2
2 count coenzyme q10 coen… oligoast… nadj… placeb… 16.1
3 count coenzyme q10 coen… oligoast… safa… placeb… 22.8
4 count coenzyme q10 coen… oligoast… safa… placeb… 22.4
5 volume coenzyme q10 coen… oligoast… nadj… placeb… 3.97
6 volume coenzyme q10 coen… oligoast… safa… placeb… 2.7
7 volume coenzyme q10 coen… oligoast… safa… placeb… 2.7
8 motili… coenzyme q10 coen… asthenoz… bale… placeb… 32.9
9 motili… coenzyme q10 coen… oligoast… nadj… placeb… 41.9
10 motili… coenzyme q10 coen… oligoast… safa… placeb… 24.2
11 motili… coenzyme q10 coen… oligoast… safa… placeb… 31.2
12 morpho… coenzyme q10 coen… asthenoz… bale… placeb… 27.1
13 morpho… coenzyme q10 coen… oligoast… nadj… placeb… 6.52
14 morpho… coenzyme q10 coen… oligoast… safa… placeb… 8.2
15 morpho… coenzyme q10 coen… oligoast… safa… placeb… 15.2
# … with 6 more variables: intervention_sd <dbl>,
# intervention_n <dbl>, control_mean <dbl>, control_sd <dbl>,
# control_n <dbl>, mean_difference <dbl>
These data are then exportable. For example, if we want to export a dataset to a .csv format so that it can be imported into different software platforms.