Blogs: Get to Know the Gateway: Concordance Tables: Health

Get to Know the Gateway: Concordance Tables: Health

by Jenny Wilkens
Mar 03, 2021
#Get-To-Know-The-Gateway #Concordance-Tables #Health

Being users of the international HRS family of surveys ourselves, we understand just how difficult it can be to find comparable items across surveys. In order to make this task easier, the Gateway has created 23 individual concordance tables covering 13 separate topics comparing the questions asked across surveys. These concordance tables describe whether a question was asked and the waves in which it was asked, as well as detailing any cross-country differences in question wording or answer scales. Best of all, this wealth of information is laid out in an easy-to-read table where you can easily compare measures across surveys or waves.

Let’s deep dive into our health topic, which has 3 concordance tables detailing health, functional limitations, and health behaviors. Our first health concordance table includes doctor diagnosed diseases, self-rated health, vision, hearing, and dental health, and information about falls, pain, and self-reported body measurements. Some of these topics have been built into variables in our Gateway Harmonized datasets since their first release: self-reported health, height, weight, BMI, and whether the respondent has ever been diagnosed with hypertension, diabetes, lung disease, heart problems, stroke, psychiatric problems, arthritis, and cancer.

blog-graphic-1
Fig 1. Health Concordance Table

But did you know that the Gateway has been expanding the variables and topics available in its datasets for the last few years? While not yet incorporated into every dataset (though we assure you that this variable expansion is in the works for the rest), we have been adding a huge number of variables covering all kinds of health topics. Not only do we now include self-rated vision and hearing, but we’ve also included additional doctor diagnosed diseases, the age when diagnosed with the disease, whether they take medication for the disease, and whether the disease limits their activity. We have also added variables capturing the respondent’s experiences with falls and pain, as well as obesity indicators. Did you know that many of these surveys also measure height and weight, along with many other anthropometric and physical measurements? We’re building a whole host of these variables into our Harmonized datasets as well, with its own concordance table coming soon.

If we move on to the functional limitation concordance table, then we can see all of the activities of daily living (ADL), instrumental activities of daily living (IADL), and other functional activities that are asked about in any of these surveys. Despite many of these surveys asking about difficulty with these activities using different answer scales, these responses have all been converted to indicate whether the respondent experiences any difficulty with the activity in the Harmonized variables. While the vast majority of these have been incorporated into the Harmonized datasets since the beginning, we have added some new summary measures to capture the activities that are most commonly asked across studies.

blog-graphic-1
Fig 2. Functional Limitation Concordance Table

For example, we used to offer two IADL summaries, one including using the phone, managing money, and taking medication, and the other adding shopping for groceries and making hot meals. We have now created a few additional IADL summaries, one including managing money, taking medication, shopping for groceries, and making hot meals, which are the four most commonly asked about IADLs across surveys, and one totaling all of the IADLs asked about within a given survey, perfect for longitudinal analyses. And these are just the tip of the iceberg! ADL and IADL activities are also closely linked to care the respondent receives, both from informal sources like family and formal sources like paid caregivers, and we have created Harmonized variables detailing the respondent’s experience with caregiving, though more on that discussion will have to wait for another blog.

When we get to our health behavior concordance table, it’s evident that there’s quite a bit of variation in how these questions are asked across studies. Despite these differences, we have already included as many comparable Harmonized variables as possible describing the respondent’s frequency of smoking, drinking, and exercise. We have also been working on adding Harmonized variables capturing binge drinking, the CAGE alcohol questionnaire, the age the respondent started and quit smoking. These new variables open up so many different avenues for study!

blog-graphic-1
Fig 3. Health Behavior Concordance Table

With the wealth of data available from the international family of HRS surveys, isn’t it great to know that you have a resource to easily see what’s available? We hope you’re able to take advantage of our concordance tables, along with our other tools for comparing question and variable availability across surveys and waves – including our survey search function, working paper series, and graphs and tables. We’re excited about the new variables we’re offering, and we hope you are too! If there are ever any variables you want to have included in our Harmonized datasets that haven’t already been incorporated, send us your suggestion in an email (help@g2aging.org) and we’ll see what we can do!

about the author
Jenny Wilkens is a Senior Research Programmer for The Program for Global Health, Aging & Policy at the Center for Economic and Social Research at the University of Southern California.