Calculating Scores

Composite scores for ease of access, and communication and customer service are all calculated by averaging together the scores for individual questions. The possible responses to each question are ordered from the worst to the best. A question's score is the percentage of responses in the best two categories (in most cases, "Usually" and "Always"). When military treatment facility (MTF) scores are reported, MTF percentages for each question are calculated, summed and divided by the number of questions in the composite. Regional scores are the averages of MTF scores. When only regional scores are reported (in the Retirees and Dependents table, for example), regional percentages for each question are calculated, summed and divided by the number of questions. 

The preventive care composite is the percentage of preventive care that users should have gotten that they actually received. For instance, for a region, the number of mammograms, or pap smears that users in that region reported is summed, weighted by the number of women over 40 or adult women in a standard population, then divided by the number that standard population should have gotten based on their ages and sex.

A rating score is the percentage of users who rate an aspect of care 8 or higher, on a scale of 0-10.

For more information, see Technical Help.

Sample Weights

The results in the TRICARE Beneficiary Reports are based on questionnaires mailed to a stratified random sample of beneficiaries. That means that everyone who responds to the survey stands in for other beneficiaries who were not mailed a questionnaire or who did not respond. When percentages are calculated from these responses, the responses on each returned questionnaire are given a weight corresponding to the number of beneficiaries that respondent represents. The weights are calculated in two stages, sampling, when beneficiaries are assigned a weight based on the number sampled, and a non-response adjustment, which is based on the proportion of the surveys mailed that are returned. Beginning with the report for October, 2005, we use a non-response adjustment based on several factors, such as the beneficiary's age, sex and rank, that were not part of the original sample design. As a result, scores more accurately reflect the TRICARE population.

Report Columns

Scores in the reports are based on responses to questions in the Health Care Survey of DoD Beneficiaries. There are two types of scores: single numeric ratings and composites. The single numeric ratings reflect questions that ask users to rate some aspect of their health care on a scale of 0 to 10. The composites average together responses from several different but related questions. The Survey Questions (CAHPS Version 5.0) shows composites, the individual questions comprising them, and the numeric ratings questions. To see the percent of favorable responses to questions making up a composite, click on the composite column heading in any report.

Finding Scores for Your Region or MTF Service Affiliation

To find scores for your region, service, or combination of region and service, select a beneficiary or enrollment group in the Report Menu on the home page of this site. This will bring up a report page showing all regions and services in the MHS. Many of the scores are composites which average together scores for several aspects of health care. To see the breakdown of a composite score, click on any column heading. In the annual reports this will show you this year's score compared to last year's score. In the quarterly reports this will show you this quarter's score compared to last quarter's score. To see a comparison of beneficiary or enrollment groups in a region or service, click on its row heading. Please note in the annual reports these beneficiary or enrollment group comparisons are unavailable at the MTF level. If you are already in a report and you want to find scores for a different region, beneficiary or enrollment group, click on the Survey Reports link.

Finding Scores for Your MTF (Annual Report Only)

Reports at the MTF level are available only for certain groups: all MHS beneficiaries, active duty beneficiaries, TRICARE Prime enrollees, and enrollees with a military PCM. To find scores for your MTF, select any of these beneficiary or enrollment groups in the Report Menu on the home page of this site. This will bring up a report showing all the regions in the MHS. Click on the region and branch of service containing your MTF to bring up a regional report. The rows in that page are the MTF's in your region and branch of service. Many of the scores are composites which average together scores for several aspects of health care. To see the breakdown of a composite score, click on any column heading. This will also show you this year's score compared to last year's score.

Comparing Scores to Benchmarks

CAHPS scores and benchmarks are adjusted for user age and health status (see Adjusting Scores, below). The precisions of the MHS scores and benchmarks are calculated using the residuals from the regressions used to adjust the scores and accounting for the survey's complex sample design. The precision of a benchmark is calculated based on a sample design in which every health care product has equal weight. T-tests measure the probability that the difference between the score and the benchmark occurred by chance. If p is less than .05, the difference is significant.

Color Significance

The colors of the scores on each page show whether a score differs significantly from its civilian benchmark. A difference is significant if it is large enough that it is unlikely to occur by chance (i.e., the probability that it occurs by chance is less than 5 percent). If a score is significantly higher than the benchmark, it is green and bold. If it is significantly lower than the benchmark, it is red and italicized. Scores that are not significantly different from the benchmark are blue.

Benchmarks

Most of the questions in the HCSDB are based on questions from the Consumer Assessment of Healthcare Providers and Systems (CAHPS) Health Plans Survey. Because many health plans that serve the civilian population use that survey to assess the experience of their enrollees, their CAHPS results can be used as benchmarks for comparison with care delivered in the MHS. 

• Benchmarks for the composites and numeric ratings for FY 2012 and FY 2013 come from the National CAHPS Benchmarking Database (NCBD) for 2011.
• Benchmarks for FY 2014 come from the National Committee for Quality Assurance (NCQA) for 2013. 
• Benchmarks for FY 2015, FY 2016 and FY 2017 come from the NCQA for 2015.
• Benchmarks for FY 2018 and later come from the NCQA for 2017.

Both NCBD and NCQA collect responses to the survey from a national sample of health plans that serve the civilian population. Results from each plan for beneficiaries who responded by mail or internet are averaged together, weighted equally. The benchmarks are adjusted to correspond to the age and health of TRICARE users. Benchmarks for preventive care and smoking are taken from Healthy People 2020. These are goals set by the federal government for the percentage of Americans getting preventive care.

*The Healthy People 2020 Benchmark was changed to 85 from 78 between Q3FY2019 and Q1FY2020.

Adjusting Scores

All the scores except the preventive care scores are adjusted for user characteristics. The HCSDB asks users about their health. The relationship between their health, their age, and their health care ratings and responses to the questions in the composites is measured by regression. The regressions are used to calculate average responses for a region assuming their users are at the MHS average age distribution and health status for that enrollment or beneficiary group. Each beneficiary group and enrollment group is adjusted separately. Then scores are adjusted again, by the difference between the group benchmark and the benchmark for all TRICARE users, so that all scores can be compared to the TRICARE users benchmark.

The non-smoking rate is adjusted by direct standardization. For scores before FY 2018, rates in a beneficiary or enrollment group for each age and sex category are weighted by the number in that beneficiary or enrollment group's age and sex in 2011. For scores in FY 2018 and later, rates in a beneficiary or enrollment group for each age and sex category are weighted by the number in that beneficiary or enrollment group's age and sex in 2016.  A weighted average is then calculated.

For more information, see Technical Help.

Healthy Behavior Score

The healthy behaviors composite is made up of three scores: the non-smoking rate, the counseled to quit rate and the percent not obese. The non-smoking rate is the age and sex adjusted proportion of beneficiaries that currently does not smoke. The counseled to quit rate is the proportion of smokers or tobacco users with physician visits in the last 12 months who were counseled to quit sometimes, usually, or always. The percent not obese is the age and sex adjusted proportion with a body mass index (BMI) below the level classified as obese: less than 30. The BMI is equal to the weight in kilograms divided by the square of the height in meters. Height and weight are self-reported in the HCSDB.