CO All Payer Claims Data
Dataset Website
https://civhc.org/get-data/whats-in-the-co-apcd/
Vital Statistics
Sample size: 5.4 million
Sampling frame: Residents of the State of Colorado with insurance
Years of data: 2012 - 2021
Type of data: All payer claims (public and private insurance claims)
Dataset Description
All-payer claims databases (APCDs) assembled by states and include all types of medical claims including pharmacy, dental, and eligibility and provider files collected from both public and private insurance plans. Some types of insurance are not included such as self-insured and some government insurance like VA.
There is generally a state mandate that insurers report data directly to the states. The Colorado APCD is among the most innovative and easy to work with of the state APCDs. Stanford hosts all records and all years of the Colorado APCD from 2012 - 2021. The CO APCD represents a majority of covered lives in the state across commercial health insurance plans, Medicare (Fee-for-Service and Advantage), and Health First Colorado (Colorado’s Medicaid program).
Strengths and limitations of the dataset
There is generally a state mandate that insurers report data directly to the states. Consequently, state APCD data have several important strengths:
They include information on private insurance that many other datasets do not.
They include data from most or all insurance companies operating in any particular State, in contrast to some proprietary datasets.
They include information on care for patients across care sites, rather than just hospitalizations and emergency department visits reported as part of discharge data systems maintained by most States through State governments or hospital associations. They also include large sample sizes, geographic representation, and capture of longitudinal information on a wide range of individual patients.
Demographics
The Colorado APCD is reflective of the population of Colorado residents with health insurance. Like most claims datasets, there is limited information on race. Sex, age and income are reflective of the state of Colorado.
A snapshot of the demographics of the data can be viewed in the table below:
Table 1. Patient Demographics in Colorado APCD
|
|
Full cohort (2012-2021) |
|
|
|
N=8,524,954 |
|
|
|
N |
% |
Race |
|
|
|
|
White |
1,616,570 |
19.0 |
|
Black |
169,132 |
2.0 |
|
Asian |
156,453 |
1.8 |
|
American Indian/Alaska Native |
84,470 |
1.0 |
|
Native Hawaiian/Pacific Islander |
11,942 |
0.1 |
|
Other/Unknown |
6,486,387 |
76.1 |
Ethnicity |
|
|
|
|
Hispanic |
713,667 |
8.4 |
|
Non-Hispanic |
3,442,269 |
40.4 |
|
Unknown |
4,369,018 |
51.2 |
Gender |
|
|
|
|
Female |
4,265,876 |
50.0 |
|
Male |
4,165,877 |
48.9 |
|
Other/Unknown |
93,201 |
1.1 |
Age as of 01/01/2021 |
|
|
|
|
<10 |
876,605 |
10.3 |
|
10-19 |
1,007,451 |
11.8 |
|
20-29 |
1,277,581 |
15.0 |
|
30-39 |
1,370,177 |
16.1 |
|
40-49 |
992,227 |
11.6 |
|
50-59 |
907,722 |
10.6 |
|
60-69 |
943,732 |
11.1 |
|
70-79 |
642,626 |
7.5 |
|
≥80 |
468,096 |
5.5 |
|
Unknown |
38,737 |
0.5 |
Urban/rural (% population of zip code) |
|
|
|
|
>60% urban |
7,195,959 |
84.4 |
|
40-60% urban/rural |
435,278 |
5.1 |
|
>60% rural |
829,020 |
9.7 |
|
Unknown |
|
|
RUCA by zip code |
|
|
|
|
Metropolitan area core |
6,831,573 |
80.1 |
|
Metropolitan area high commuting |
427,350 |
5.0 |
|
Metropolitan area low commuting |
30,401 |
0.4 |
|
Micropolitan area core |
386,193 |
4.5 |
|
Micropolitan high commuting |
97,175 |
1.1 |
|
Micropolitan low commuting |
614 |
0.0 |
|
Small town core |
410,617 |
4.8 |
|
Small town high commuting |
24,740 |
0.3 |
|
Small town low commuting |
1,903 |
0.0 |
|
Rural areas |
268,948 |
3.2 |
|
Not coded |
75 |
0.0 |
|
Zipcode unknown/missing |
45,365 |
0.5 |
Selected Publications
Panhans M. Adverse Selection in ACA Exchange Markets: Evidence from Colorado. Am Econ J Appl Econ. 2019;11(2):1-36. doi:10.1257/app.20170117
Perraillon MC, Liang R, Sabik LM, Lindrooth RC, Bradley CJ. The role of all‐payer claims databases to expand central cancer registries: Experience from Colorado. Health Serv Res. 2022;57(3):703-711. doi:10.1111/1475-6773.13901
Pesa J, Rotter D, Papademetriou E, Potluri R, Patel C, Benson C. Real-world analysis of insurance churn among young adults with schizophrenia using the Colorado All-Payer Claims Database. J Manag Care Spec Pharm. 2022;28(1):26-38. doi:10.18553/jmcp.2022.28.1.26
Sakai-Bizmark R, Jackson NJ, Wu F, et al. Short Interpregnancy Intervals Among Women Experiencing Homelessness in Colorado. JAMA Netw Open. 2024;7(1):e2350242. doi:10.1001/jamanetworkopen.2023.50242
Process for accessing the data and associated costs
The process to access the data is outlined on the PHS Data Portal and on the CIVHC Website. If your work falls under an existing DUA, you need only complete the individual requirements. If you are doing a new DUA, you will have to get approval from Colorado and an IRB (Expedited category 5). Access takes about 1 month for approved questions and 3-6 months for new DUAs. Please reach out to phsdatacore@stanford.edu if you want to apply.
Data Support
PHS has CO APCD office hours, slack channels, AFC data documentation and recorded trainings.
Other Datasets: