Missing data in trial-based cost-effectiveness analysis review dataset – Data Codebook

Persistent identifier

10.17037/DATA.272

Description

This dataset contains information extracted from studies published in the NIHR Health Technology Assessment journal between the 1st January 2013 and 31st December 2015. The study team subsequently used it as a basis to critically reviews the extent of and methods used to address missing data in recently published trial-based CEA. Detailed methodology and variables definition are reported in the accompanying Health Economics paper, “Missing data in trial-based cost-effectiveness analysis: an incomplete journey”.

Data Codebook

The following variables are contained in the missing_data_review dataset.

Variable Name Variable Label Answer Label Answer Code Variable Type
author Author Open ended   String
year Year Open ended   Integer
futime FU time (months) Open ended   double
fucat Follow-up design     Numeric
    Continuous 1  
    One follow-up 2  
    Repeated assesments 3  
nfu Number of assesments, including baseline Open ended   Numeric
effectcat Primary effectiveness measure type     Numeric
    QALY 1  
    Binary 2  
    Clinical scale 3  
    Time to recovery 4  
pcc Proportion of complete cases Open ended   double
n Number complete cases Open ended   Integer
n_ Number of participants Open ended   Integer
reportcc Report exact CC     Numeric
    No 0  
    Yes 1  
arm1 Proportion CC arm1 Open ended   double
arm2 Proportion CC arm2 Open ended   double
diffarm Missingness differs between arms     Numeric
    Yes 1  
    No 2  
    No missing 3  
    Unclear 8  
cceffect CC effect Open ended   double
cccost CC cost Open ended   double
missinfo2 Missingness cost vs. effect     Numeric
    None 1  
    Both missing 2  
    Cost more missing 3  
    Effect more missing 4  
    Unclear 8  
md Included - missing data >5% Open ended   Numeric
primcat Primary analysis category     Numeric
    CC 1  
    MI 2  
    Other 3  
    Mixed 4  
    Unclear 5  
primcat3 Primary analysis     Numeric
    CC 1  
    MI 2  
    Other 3  
mi Conducted multiple imputation     Numeric
    No 0  
    Yes 1  
cc Conducted complete-case analysis     Numeric
    No 0  
    Yes 1  
sa Conducted sensitivity analysis     Numeric
    No 0  
    Yes 1  
other1 Secondary approach to MD 1 Open ended   String
other2 Secondary approach to MD 2 Open ended   String
other3 Secondary approach to MD 3 Open ended   String
mnarsa MNAR SA     Numeric
    No 0  
    Yes 1  
a1 Recommendations indicator A1.Maximise response rate     Numeric
    Yes 1  
    No 2  
    Unclear 3  
a2 A2.Alternative data sources     Numeric
    Yes 1  
    No 2  
    Unclear 3  
a3 A3.Monitor completeness     Numeric
    Yes 1  
    No 2  
    Unclear 3  
b1 B1.Assumption for primary analysis     Numeric
    Yes 1  
    No 2  
    Unclear 3  
b2 B2.Appropriate primary method     Numeric
    Yes 1  
    No 2  
    Unclear 3  
c1 C1.Discuss departures from the primary assumption     Numeric
    Yes 1  
    No 2  
    Unclear 3  
c2 C2.Consider broad range of assumptions     Numeric
    Yes 1  
    No 2  
    Unclear 3  
c3 C3.Method valid under these assumptions     Numeric
    Yes 1  
    No 2  
    Unclear 3  
d1 D1.Missing data by endpoint, arm and time-point     Numeric
    Yes 1  
    No 2  
    Unclear 3  
d2 D2.Discuss reasons for missing data     Numeric
    Yes 1  
    No 2  
    Unclear 3  
d3 D3.Describe methods used and assumptions     Numeric
    Yes 1  
    No 2  
    Unclear 3  
d4 D4.Conclusions in light of missing data     Numeric
    Yes 1  
    No 2  
    Unclear 3