tidycms

The goal of tidycms is to easily access public datasets from Centers for Medicare & Medicaid Services (CMS). The package is built on top of httr2. With user-friendly syntax, one can get a tidy tibble in a few lines of code.

Check the CMS API documentation and FAQs.

Installation

You can install the development version of tidycms from GitHub with:

# install.packages("devtools")
devtools::install_github("JiaZhang42/tidycms")

Example

This is a basic example which shows you how to glimpse all available datasets from CMS:

library(tidycms)
cms_datasets <- show_datasets()
cms_datasets
#> # A tibble: 135 × 7
#>    title         identifier temporal modified description reference landing_page
#>    <chr>         <chr>      <chr>    <chr>    <chr>       <chr>     <chr>       
#>  1 Accountable … 9767cb68-… 2014-01… 2024-01… "The Accou… https://… https://dat…
#>  2 Accountable … 5b227bd9-… 2017-01… 2024-04… "The Accou… https://… https://dat…
#>  3 ACO Realizin… 1cd9eded-… 2021-01… 2024-01… "The Accou… https://… https://dat…
#>  4 ACO Realizin… 54551982-… 2021-01… 2024-01… "Accountab… https://… https://dat…
#>  5 ACO Realizin… 6c3532b3-… 2021-01… 2024-01… "The Accou… https://… https://dat…
#>  6 ACO Realizin… e0eba16f-… 2021-01… 2024-01… "The Accou… https://… https://dat…
#>  7 Advance Inve… a3d35ba1-… 2024-01… 2024-01… "The Advan… https://… https://dat…
#>  8 Agency for H… 7cf9662e-… 2015-01… 2020-12… "The Agenc… https://… https://dat…
#>  9 CMS Program … bbcffb70-… 2016-01… 2024-05… "The CMS P… https://… https://dat…
#> 10 CMS Program … 900059a0-… 2016-01… 2024-05… "The CMS P… https://… https://dat…
#> # ℹ 125 more rows

This table contains the basic information of all latest public datasets from CMS. Columns include

• title

• identifier: the “Dataset Type Identifier”, which doesn’t change and always points to the latest version of the dataset. For “Dataset Version Identifier”, visit its landing_page.

• temporal: time period

• modified: last modified date

• description

• reference: documentation link, named “describedBy” by CMS

• landing_page: webpage for the dataset

Alternatively, you can search for a specific dataset by keywords at CMS website search.

Either way, you could get the identifier of the dataset you are interested in. For example, the identifier for the dataset “Accountable Care Organization Participants” is 9767cb68-8ea9-4f0b-8179-9431abc89f11.

Then you can preview the first 10 rows of the dataset:

data_id <- '9767cb68-8ea9-4f0b-8179-9431abc89f11'
aco_participants <- slice_head_data(data_id, 10)
aco_participants
#> # A tibble: 10 × 29
#>    aco_id par_lbn                 aco_name aco_service_area agreement_period_num
#>    <chr>  <chr>                   <chr>    <chr>            <chr>               
#>  1 A2811  ARIZONA COMMUNITY PHYS… Abacus … AZ               3                   
#>  2 A2811  TMC MEDICAL NETWORK     Abacus … AZ               3                   
#>  3 A1596  CARECONNECT HEALTH INC  Account… GA               4                   
#>  4 A1596  Center for Pan Asian C… Account… GA               4                   
#>  5 A1596  Coastal Community Heal… Account… GA               4                   
#>  6 A1596  Curtis V. Cooper Prima… Account… GA               4                   
#>  7 A1596  Diversity Health Cente… Account… GA               4                   
#>  8 A1596  GEORGIA MOUNTAINS HEAL… Account… GA               4                   
#>  9 A1596  Health Education Asses… Account… GA               4                   
#> 10 A1596  J.C. Lewis Primary Hea… Account… GA               4                   
#> # ℹ 24 more variables: initial_start_date <chr>, current_start_date <chr>,
#> #   `re-entering_aco` <chr>, basic_track <chr>, basic_track_level <chr>,
#> #   enhanced_track <chr>, high_revenue_aco <chr>, low_revenue_aco <chr>,
#> #   adv_pay <chr>, aim <chr>, aip <chr>, `snf_3-day_rule_waiver` <chr>,
#> #   prospective_assignment <chr>, retrospective_assignment <chr>,
#> #   aco_address <chr>, aco_public_reporting_website <chr>, aco_exec_name <chr>,
#> #   aco_exec_email <chr>, aco_exec_phone <chr>, aco_public_name <chr>, …

If you want to collect all data from a dataset, you can use the collect_data() function:

data_id <- 'f1a8c197-b53d-4c24-9770-aea5d5a97dfb'
aco_participants_all <- collect_data(data_id, ask_for_confirmation = FALSE)
#> ℹ 1485 rows to be collected

aco_participants_all
#> # A tibble: 1,485 × 9
#>    NPI       `PROVIDER NAME` `ADDRESS LINE 1` `ADDRESS LINE 2` CITY  STATE ZIP  
#>    <chr>     <chr>           <chr>            <chr>            <chr> <chr> <chr>
#>  1 10030813… BAART BEHAVIOR… 617 COMSTOCK RD  STE 5            BERL… VT    0560…
#>  2 10031500… AMS OF WISCONS… 9532 E 16 FRONT… STE 100          ONAL… WI    5465…
#>  3 10033624… BHG XLII LLC    5715 PRINCESS A… <NA>             VIRG… VA    2346…
#>  4 10033689… RTS EDGEWOOD    2205 PULASKI HI… <NA>             EDGE… MD    21040
#>  5 10035716… METRO TREATMEN… 1241 BLANDING B… NEW SEASON TREA… ORAN… FL    3206…
#>  6 10035811… PREMIER CARE O… 2632 WOODMAN CE… <NA>             KETT… OH    4542…
#>  7 10035837… AFFINITY HEALT… 1305 KINGS HWY N <NA>             CHER… NJ    0803…
#>  8 10039416… TARZANA TREATM… 5190 ATLANTIC A… <NA>             LONG… CA    9080…
#>  9 10039471… WEST TEXAS COU… 1108 DOBIE DR S… WTCR PLANO, INC. PLANO TX    7507…
#> 10 10039535… ALLIANCE RECOV… 1116 E PONCE DE… <NA>             DECA… GA    3003…
#> # ℹ 1,475 more rows
#> # ℹ 2 more variables: `MEDICARE ID EFFECTIVE DATE` <chr>, PHONE <chr>