Skip to main content

Frequently Asked Questions about the 2000 Area Deprivation Index

What do the ADI values mean?

In the ADI datasets that we have posted, all zip codes, ZCTAs, counties, or block groups have an ADI value, and the the higher numbers indicate “more deprived” areas.

For example, you could say that an ADI of 95.13 is more favorable than an ADI of 117.01.

Is there a cutpoint at which higher ADIs have negative outcomes?

In this Annals of Internal Medicine article that used the ADI that we have posted on HIPxChange, patients from the top most disadvantaged 15% of neighborhoods—i.e., the highest 15% of ADI values—were found to be more likely to be re-hospitalized.

Table 3 in the article shows the cut-points for the ADI scores and what a patient’s risk of readmission is based on the ADI value; you may find this table useful in interpreting the datasets that are posted on HIPxChange.

Why are there some negative ADI values?

While most of the ADI values are positive, there are some negative values in the datasets. The ADI is set to have a mean of 100 and a standard deviation of 20. It is uncommon to have negative values, but not impossible; these represent areas with a very low level of deprivation.

What methodology was used to create the ADI datasets on HIPxChange?

The following 2 articles explain the methodology that was used to create the ADI datasets on HIPxChange:

What are some examples of ways these ADI datasets have been used?

  • Andrew Knighton et al. at Intermountain Healthcare constructed a Utah ADI based on 2013 American Community Survey data at the Census block group level. This paper discusses the methodology for creating the dataset.
  • Andrea Maroko et al. assessed the appropriate geographic scale to apply the ADI to identify and screen patients for social determinants of health. The methods and results are discussed in this paper
  • The Maryland Health Services Cost Review Commission created an updated version of the ADI using American Community Survey data and is using it in its recommendations for the Readmissions Reduction Incentive Program.

I imported the 9-digit ZIP code dataset in SAS and am seeing rows with fewer than 9-digit ZIP codes. What should I do?

Please try using the following code to import the CSV file. This should fix the issue that you are experiencing.

PROC IMPORT
  DATAFILE="[FOLDER PATH OF CSV FILE]\zip4_dep_index.csv"
  DBMS=CSV
  OUT=work.zip4_dep_index
  REPLACE;
  GUESSINGROWS=100000;
RUN;

Will you be posting updated datasets with newer Census data?

We are pleased to announce that the 2013 US Area Deprivation Index (ADI), including both datasets and a customized mapping interface, is now available free through the University of Wisconsin’s new Neighborhood Atlas website.

Do you have a 5-digit ZIP code dataset available, or a ZCTA level dataset?

No. A ZIP code tabulation area (ZCTA) level dataset was previously posted but is no longer provided for the following reasons:

  1. In recent validation work that uses 2009-2013 American Community Survey data, it has become clear that the ADI should not be used at any levels other than those defined by the Census (see diagram of Census levels here). Those with interest in using a ZIP-based methodology may still employ the 9-digit ZIP code crosswalk, which was built to correspond directly to Census block groups and accompanies the Census block group level ADI.
  2. "Employment of ZIP Code Tabulation Areas to link geographic data is a convenient but, ultimately, inferior method for this sort of assessment (Grubesic and Matisziw, 2006). It results in relatively large geographic zones with linkages that can lead to less precise estimates, especially in areas in which concentrated poverty abuts more wealthy regions." - Excerpt drawn directly from Kind et al., Health Affairs, Sept 15, 2016

Reference:

Will you be posting a version of the ADI that uses 5-digit ZIP codes instead of 9-digit ZIP codes?

No; please see the previous question for the rationale.

 
nike air max 90 kids