AHRQ Quality Indicators

Overview

Code on GitHub

The AHRQ Quality Indicators data mart implements measures maintained by the Agency for Healthcare Research and Quality. AHRQ describes the Quality Indicators as standardized, evidence-based measures that use administrative healthcare data to measure and track quality, outcomes, and potential access issues.

This package currently implements the Prevention Quality Indicators in Inpatient Settings (PQIs). PQIs identify hospital admissions for ambulatory care-sensitive conditions, which are conditions where timely, high-quality outpatient care may prevent hospitalization or reduce disease severity.

Install this standalone package alongside Core using the data mart installation guide.

Methodology

The PQI models run on top of the Core Data Model and use inpatient encounters, diagnoses, procedures, patient demographics, and member months to produce numerator, denominator, exclusion, and rate outputs. The implementation follows AHRQ's PQI measure concepts and value sets while expressing the logic in dbt models that run across Tuva-supported data warehouses.

AHRQ maintains the full Quality Indicators documentation, including PQI technical specifications and supporting resources.

The individual PQIs currently implemented are:

PQI Number	PQI Name	Description
01	Diabetes Short-Term Complications Admission Rate	Hospitalizations for a principal diagnosis of diabetes with short-term complications per 100,000 population, ages 18 years and older.
03	Diabetes Long-Term Complications Admission Rate	Hospitalizations for a principal diagnosis of diabetes with long-term complications per 100,000 population, ages 18 years and older.
05	COPD or Asthma in Older Adults	Hospitalizations with a principal diagnosis of chronic obstructive pulmonary disease or asthma per 100,000 population, ages 40 years and older.
07	Hypertension Admission Rate	Hospitalizations with a principal diagnosis of hypertension per 100,000 population, ages 18 years and older.
08	Heart Failure Admission Rate	Hospitalizations with a principal diagnosis of heart failure per 100,000 population, ages 18 years and older.
11	Community Acquired Pneumonia Admission Rate	Hospitalizations with a principal diagnosis of community-acquired bacterial pneumonia per 100,000 population, ages 18 years and older.
12	Urinary Tract Infection Admission Rate	Hospitalizations with a principal diagnosis of urinary tract infection per 100,000 population, ages 18 years and older.
14	Uncontrolled Diabetes Admission Rate	Hospitalizations for uncontrolled diabetes without short-term or long-term complications per 100,000 population, ages 18 years and older.
15	Asthma in Younger Adults Admission Rate	Hospitalizations for a principal diagnosis of asthma per 100,000 population, ages 18 to 39 years.
16	Lower-Extremity Amputation Among Patients with Diabetes Rate	Hospitalizations for diabetes and a lower-extremity amputation procedure, excluding toe amputations, per 100,000 population, ages 18 years and older.

Outputs

Model	Description
ahrq_quality_indicators.pqi_summary	One qualifying numerator encounter per PQI and data source, with supporting encounter context.
ahrq_quality_indicators.pqi_rate	PQI numerator, denominator, and rate output by year and data source.
ahrq_quality_indicators.pqi_num_long	Long-format numerator records by PQI.
ahrq_quality_indicators.pqi_denom_long	Long-format denominator records by PQI.
ahrq_quality_indicators.pqi_exclusion_long	Long-format exclusion records by PQI.

Example SQL

PQI Encounters by Measure

select
      data_source
    , pqi_number
    , pqi_name
    , count(*) as pqi_encounters
from ahrq_quality_indicators.pqi_summary
group by
      data_source
    , pqi_number
    , pqi_name
order by pqi_encounters desc;

PQI Rates

select
      data_source
    , year_number
    , pqi_number
    , num_count
    , denom_count
    , rate_per_100_thousand
from ahrq_quality_indicators.pqi_rate
order by
      data_source
    , year_number
    , pqi_number;

Calculate Rates from Numerator and Denominator Tables

with numerator as (
    select
          data_source
        , year_number
        , pqi_number
        , count(encounter_id) as numerator_count
    from ahrq_quality_indicators.pqi_num_long
    group by
          data_source
        , year_number
        , pqi_number
)

, denominator as (
    select
          data_source
        , year_number
        , pqi_number
        , count(person_id) as denominator_count
    from ahrq_quality_indicators.pqi_denom_long
    group by
          data_source
        , year_number
        , pqi_number
)

select
      denominator.data_source
    , denominator.year_number
    , denominator.pqi_number
    , denominator.denominator_count
    , coalesce(numerator.numerator_count, 0) as numerator_count
    , 100000.0 * coalesce(numerator.numerator_count, 0)
        / nullif(denominator.denominator_count, 0) as rate_per_100000
from denominator
left join numerator
    on denominator.data_source = numerator.data_source
    and denominator.year_number = numerator.year_number
    and denominator.pqi_number = numerator.pqi_number
order by
      denominator.data_source
    , denominator.year_number
    , denominator.pqi_number;