About Disparities Data

Healthy People 2030 includes health disparities data for population-based core objectives with available demographic group data. Health disparities are differences in health that are closely linked to social determinants of health.

Addressing health disparities is key to achieving health equity and realizing the Healthy People vision of improving the health and well-being of all.

How to use disparities data 

Watch this video to learn how to interact with Healthy People 2030 objective data, including disparities data.

How we calculate disparities data

The following are key concepts in the assessment of disparities:

Notes:

  • The term “rate” is used as shorthand for rate, percentage, proportion, or any other measurement for which disparities can be meaningfully assessed. Not all Healthy People objectives are measured using a rate.
  • To calculate disparities, demographic group data are needed for at least 2 subgroups for the same period.
  • Whenever possible, disparities are calculated for pre-determined, mutually exclusive (non-overlapping) groups. For example, when “not Hispanic or Latino” race groups are available, racial/ethnic disparities are calculated for the “Hispanic or Latino” group and the race groups listed under “not Hispanic or Latino.”  
  • Rate ratios (RRs) and summary rate ratios (SRRs) presented in Healthy People are constructed such that they are always greater than 1.000.
  • All calculations use unrounded estimates and their standard errors (SE). The values on the Healthy People 2030 website may differ from those calculated using the displayed (rounded to 3 decimals) values of the estimated rates and standard errors.

Highest rate (Rmax)

The highest rate (Rmax) is the highest unrounded group rate for the specified population characteristic (e.g., race and ethnicity or age). When there’s a numerical tie in the unrounded group rates and the unrounded standard errors (SEs) are available, the group rate with the smaller SE serves as Rmax. In the absence of SEs, the denominator of the larger group serves as the tiebreaker. As described in “Measures of variability” below, statistical significance is not used to choose Rmax.

Lowest rate (Rmin)

The lowest rate (Rmin) is the lowest unrounded group rate for the specified population characteristic (e.g., race and ethnicity or age). When there’s a numerical tie in the unrounded group rates and the unrounded standard errors (SEs) are available, the group rate with the smaller SE serves as Rmin. In the absence of SEs, the denominator of the larger group serves as the tiebreaker. As described in "Measures of variability" below, statistical significance is not used to choose Rmin.

Reference rate

The reference rate depends on the desired direction for an objective. 

Some objectives have a desired increase — like the objective to “Increase the proportion of people with a usual primary care provider.” In this case the reference rate is the highest rate (Rmax). 

Other objectives have a desired decrease — like the objective to “Reduce the suicide rate.” In this case the reference rate is the lowest rate (Rmin). 

To find the reference rate in a data table, look for the “Reference rate” label.

Maximal rate difference (MRD)

For the specified population characteristic, the maximal rate difference (MRD) is the difference between the highest rate (Rmax) and lowest rate (Rmin). It’s calculated as:

MRD = Rmax - Rmin

Maximal rate ratio (MRR)

For the specified population characteristic, the maximal rate ratio (MRR) is the ratio between the highest rate (Rmax) and lowest rate (Rmin). 

 For an objective with a desired increaseFor an objective with a desired decrease
FormulaMRR = Rmax/Rmin
Understanding the MRRThe highest group rate Rmax was MRR times the lowest group rate Rmin.
Meaning of × and ÷ before MRR values in disparities tablesThe × emphasizes that the reference rate is Rmax. This means Rmin would need to be multiplied by MRR to equal the reference rate Rmax.The ÷ emphasizes that the reference rate is Rmin. This means Rmax would need to be divided by MRR to equal the reference rate Rmin.

Rate ratio (RR)

The rate ratio (RR) compares 2 group rates for the specified population characteristic. 

When an increase is desired for the objective, the highest rate (Rmax) is the reference rate and serves as the numerator when calculating RR. 

When a decrease is desired for the objective, the lowest rate (Rmin) is the reference rate and serves as the denominator when calculating RR.

For any group g in a specified population characteristic with K groups, RR is calculated as follows:

 For an objective with a desired increaseFor an objective with a desired decrease
FormulaRR = Rmax/Rg, g = 1, 2, … KRR = Rg/Rmin, g = 1, 2, … K
Understanding the RRThe highest group rate Rmax was RR times the comparison group rate (Rg).The comparison group rate (Rg) was RR times the rate of the lowest group rate (Rmin).
Meaning of × and ÷ before RR values in disparities tablesThe × emphasizes that the reference rate is Rmax. This means Rg would need to be multiplied by RR to equal the reference rate Rmax.The ÷ emphasizes that the reference rate is Rmin. This means Rg would need to be divided by RR to equal the reference rate Rmin.

Summary rate ratio (SRR)

The summary rate ratio (SRR) compares a reference rate to an average rate. The average rate (Rave) is the average of all group rates other than the reference rate for a population characteristic. As for the rate ratio (RR), the reference rate is Rmax and serves as the SRR’s numerator when the objective has a desired increase. Similarly, the reference rate is Rmin and serves as the SRR’s denominator when the objective has a desired decrease.  

The SRR is calculated as follows:

 For an objective with a desired increaseFor an objective with a desired decrease
FormulaSRR = Rmax/RaveSRR = Rave/Rmin
Understanding the SRRThe highest group rate Rmax was SRR times the average rate Rave of all other groups.The average rate Rave of all other groups was SRR times the rate of the lowest group rate Rmin.
Meaning of × and ÷ before SRR values in disparities tablesThe × emphasizes that Rmax is the reference rate. This means Rave would need to be multiplied by SRR to equal the reference rate Rmax.The ÷ emphasizes that the reference rate is Rmin. This means Rave would need to be divided by SRR to equal the reference rate Rmin.

Interpreting ratios of rates
Ratios of rates like the MRR, rate ratio (RR), and summary rate ratio (SRR) can be interpreted as percentage differences between 2 rates. For example, if the ratio between rates A and B is 1.432, that means rate A is 1.432 times rate B — or that rate A is 43.2% higher than rate B.

Measures of variability

Individual rates may be subject to uncertainty, as captured by standard errors (SEs), so ordering rates from highest to lowest (or lowest to highest) may also be subject to uncertainty. When available, SEs and 95 percent confidence intervals (95% CIs) are used to account for the uncertainty in the ranking and the interdependence among the ordered values that impact the disparity measures described above.

A resampling/bootstrap procedure is used to calculate the SEs and 95% CIs for the highest rate (Rmax), lowest rate (Rmin), maximal rate difference (MRD), maximal rate ratio (MRR), rate ratio (RR), and summary rate ratio (SRR).

The resampling/bootstrap procedure uses the rate and SE for each group within a population characteristic to randomly draw 25,000 replicate group rates according to a normal distribution. This set of simulated group rates is then used to generate a set of simulated relative rankings, reference rates, and disparity measures. The frequency distribution of each of these simulated estimates is then used to estimate its empirical SE and 95% CI, where the lower limit is the 2.5th percentile and the upper limit is the 97.5th percentile.

Learn More

Resources

  • Braveman, P. (2006). Health disparities and health equity: Concepts and measurement. Annual Review of Public Health, 27, 167–194. https://doi.org/10.1146/annurev.publhealth.27.021405.102103  
  • Harper, S., King, N. B., Meersman, S. C., Reichman, M. E., Breen, N., & Lynch, J. (2010). Implicit value judgments in the measurement of health inequalities. Millbank Quarterly, 88(1), 4–29. https://doi.org/10.1111/j.1468-0009.2010.00587.x  
  • Huang, D. T., Bassig, B. A., Hubbard, K., Klein, R. J., & Talih, M. (2022). Examining progress toward elimination of racial and ethnic health disparities for Healthy People 2020 objectives using three measures of overall disparity. Vital Health Statistics (Series 2, No. 195). National Center for Health Statistics. https://doi.org/10.15620/cdc:121266 
  • Keppel, K., Pamuk, E., Lynch, J., Carter-Pokras, O., Kim, I., Mays, V., Pearcy, J., Schoenbach, V., & Weissman, J. S. (2005). Methodological issues in measuring health disparities. Vital Health Statistics (Series 2, No. 141). National Center for Health Statistics. https://www.cdc.gov/nchs/data/series/sr_02/sr02_141.pdf  
  • Penman-Aguilar, A., Talih M, Huang D, Moonesinghe, R., Bouye, K., & Beckles, G. (2016). Measurement of health disparities, health inequities, and social determinants of health to support the advancement of health equity. Journal of Public Health Management & Practice, 22(1 Supp), S33–S42. https://doi.org/10.1097/phh.0000000000000373   
  • Talih, M., & Huang, D. T. (2016). Measuring progress toward target attainment and the elimination of health disparities in Healthy People 2020. Healthy People Statistical Notes (No. 27). National Center for Health Statistics. https://www.cdc.gov/nchs/data/statnt/statnt27.pdf