The Impact of Patient Complexity on Healthcare Utilization

• Conditions: Well Child Visits in First 15 Months of Life NQF 1392; Emergency Department Utilization; Diabetes Mellitus NQF 0059; Alcohol and Drug Screening; Primary Care Quality Metrics; Colorectal Cancer Screening NQF 0034

• Registration Number: NCT03327896
• Lead Sponsor: OCHIN, Inc.

Brief Summary

Healthcare providers are routinely being assessed for metrics designed to assess the quality of the care they deliver. There is growing consensus that these measurements, which typically assess the percentage of patients meeting a specific standard of care, should be adjusted for the clinical complexity of the providers. This study will assess whether adjusting for the social complexity of the patient panel adds significantly to adjustment for clinical complexity in explaining apparent differences in quality of care provided by Primary care providers and clinics.

Detailed Description

The analysis will be conducted in 2 stages. In the first stage, indicators of social complexity (as assessed by characteristics of the patients' geocoded addresses) will be tested for significant association with patient level outcomes (meeting or not meeting a specific standard of care). These indicators will be assessed in 4 cohorts: 2 cohorts of patients identified in electronic health records of Clinical Research Networks of community health centers, and 2 cohorts of patients insured by Medicaid in 2015 (Oregon and Florida Medicaid cohorts). Due to limitations in the available data, some quality indicators can only be evaluated in a subset of the cohorts. The consistency of specific social complexity indicators, referred to as Community Vital Signs (CVS), will be evaluated across the cohorts and quality metrics.

In the second stage, CVS indicators that contributed consistently to models at the patient level will be evaluated for their effect on provider ranking on quality metrics if the metrics are adjusted for the social complexity of the providers' patient panel in addition to the panels' clinical complexity.

Study Timeline

• Study Start: 2017-01-15
• Primary Completion: 2017-08
• Status Verified: 2017-10
• Study First Submitted: 2017-10-27
• Study First Submitted QC: 2017-10-27
• Last Update Submitted: 2017-10-27
• Study First Posted: 2017-11-01
• Last Update Posted: 2017-11-01
• Study Completion: 2018-08

Recruitment & Eligibility

• Status: UNKNOWN
• Sex: All
• Target Recruitment: 200000

Inclusion Criteria

Not provided

Exclusion Criteria

Exclusion criteria for patients: Patients with no geocoded address on file will be excluded from the analysis.

Exclusion criteria for patients for Medicaid based Outcomes: Patient with Medicaid coverage gaps of >45 days will be excluded from theses analyses. Exclusion criteria for providers: Providers with <20 patients in any Quality metric denominator.

Study & Design

• Study Type: OBSERVATIONAL
• Study Design: Not specified

Primary Outcome Measures

Name	Time	Method
Provider Ranking as Assessed by Health Care Quality Metrics	1 year (2015)	Comparison of adjusted provider quality rankings with metrics adjusted for Charlson Morbidity Scores and Patient Panel Neighborhood Deprivation Indicators

Secondary Outcome Measures

Name	Time	Method
Well Child Visits in first 15 Months of life (patient level)	15 months	Number of well child visits in first 15 months of life in children turning 15 months in 2015
Emergency Department Visits (Patient level)	1 year (2015)	Age stratified Emergency department visit rates in patients in Medicaid cohorts
Colorectal Cancer Screening (Patient level)	1 year	Documented colorectal cancer screening in 2015 (Oregon Medicaid cohort only)
Uncontrolled Hemoglobin A1c in Diabetics (patient level)	1 year	Last recorded HbA1c greater than 9 in patients with recorded diagnosis of Diabetes (EHR cohorts only)
Avoidable Emergency Department Visits (Patient level)	1 year (2015)	Age stratified Avoidable Emergency department visit rates in patients in Medicaid cohorts
Screening for alcohol and drug misuse (patient level)	1 year (2015)	Documented screening for alcohol or drug misuse in patients aged 12 and over