Hello,
I am using NHANES data with Stata 14.1.
I have a question about setting my weights using svyset. I am analyzing latent tuberculosis infection results. There are two tests in the dataset that do this (tuberculin skin test and Quantiferon Gold test).
In the literature I have seen that people weight their data by Medical Examination Center sample 2-year weights that adjust for nonparticipation in the physcial examination and survey. However, I have also seen that people analyzing the outcome that I am weight the data by nonparticipation to either of these tests (Mancuso et al; Bennett et al; Shea et al -- citations below). This is because the missing TST and QFT-GIT data are considered a type of survey non-response which can bias the estimates.
I am able to weight the data by the first 2-year sample weight part but have been unable to additionally weight the data by the nonparticipation bias from either the tuberculin skin test of the Quantiferon test. Therefore, I feel my analysis may be off because I am weighting incorrectly.
When I press svyset on my current dataset this is what I currently get. .
The current names for the variables for the outcome variables are "quantnew" for the Quantiferon test and "ppd10" for the tuberculin skin test.
Can someone help me with this final step?
Thanks
Leo
Mancuso, J.D., Diffenderfer, J.M., Ghassemieh, B.J., Horne, D.J. and Kao, T.C., 2016. The prevalence of latent tuberculosis infection in the United States. American journal of respiratory and critical care medicine, (ja).
Bennett, D.E., Courval, J.M., Onorato, I., Agerton, T., Gibson, J.D., Lambert, L., McQuillan, G.M., Lewis, B., Navin, T.R. and Castro, K.G., 2008. Prevalence of tuberculosis infection in the United States population: the national health and nutrition examination survey, 1999–2000. American journal of respiratory and critical care medicine, 177(3), pp.348-355.
Shea, K.M., Kammerer, J.S., Winston, C.A., Navin, T.R. and Horsburgh, C.R., 2013. Estimated rate of reactivation of latent tuberculosis infection in the United States, overall and by population subgroup. American journal of epidemiology, p.kwt246.
I am using NHANES data with Stata 14.1.
I have a question about setting my weights using svyset. I am analyzing latent tuberculosis infection results. There are two tests in the dataset that do this (tuberculin skin test and Quantiferon Gold test).
In the literature I have seen that people weight their data by Medical Examination Center sample 2-year weights that adjust for nonparticipation in the physcial examination and survey. However, I have also seen that people analyzing the outcome that I am weight the data by nonparticipation to either of these tests (Mancuso et al; Bennett et al; Shea et al -- citations below). This is because the missing TST and QFT-GIT data are considered a type of survey non-response which can bias the estimates.
I am able to weight the data by the first 2-year sample weight part but have been unable to additionally weight the data by the nonparticipation bias from either the tuberculin skin test of the Quantiferon test. Therefore, I feel my analysis may be off because I am weighting incorrectly.
When I press svyset on my current dataset this is what I currently get. .
The current names for the variables for the outcome variables are "quantnew" for the Quantiferon test and "ppd10" for the tuberculin skin test.
Can someone help me with this final step?
Thanks
Leo
Mancuso, J.D., Diffenderfer, J.M., Ghassemieh, B.J., Horne, D.J. and Kao, T.C., 2016. The prevalence of latent tuberculosis infection in the United States. American journal of respiratory and critical care medicine, (ja).
Bennett, D.E., Courval, J.M., Onorato, I., Agerton, T., Gibson, J.D., Lambert, L., McQuillan, G.M., Lewis, B., Navin, T.R. and Castro, K.G., 2008. Prevalence of tuberculosis infection in the United States population: the national health and nutrition examination survey, 1999–2000. American journal of respiratory and critical care medicine, 177(3), pp.348-355.
Shea, K.M., Kammerer, J.S., Winston, C.A., Navin, T.R. and Horsburgh, C.R., 2013. Estimated rate of reactivation of latent tuberculosis infection in the United States, overall and by population subgroup. American journal of epidemiology, p.kwt246.
