The potential of healthcare claim coding for outcomes measurement
Clinical data is crucial for understanding the outcomes and cost-effectiveness of healthcare interventions. Primary data collection, record abstraction, and chart review are most common in randomized controlled trials (RCTs) evaluating clinical effectiveness. Though the level of detail available is less than that of these other sources, accumulated administrative healthcare claim data contain a near-comprehensive record of a patient’s care because providers collectively submit services and treatment they provide as claims for reimbursement even though not all providers bill for every service they render. As such, claim data have been and are being used more frequently in health economics and outcomes research (HEOR).
Several standardized coding systems are employed for the processing of healthcare claims (see brief descriptions of them in Figure 1). Coding of services and test results on a medical claim can facilitate the evaluation of health outcomes. A subset of Current Procedural Terminology (CPT), CPT Category II codes are a relatively recent addition to CPT made available to healthcare professionals by the American Medical Association (AMA) for tracking performance measurement.1,2 These optional supplemental tracking codes are not used directly nor required for reimbursement; instead, they allow for facilitating data collection about the quality of care without requiring a labor-intensive manual review of the medical record.
The AMA’s most recent listing contains more than 600 CPT Category II codes for categories of patient management, patient history, physical examination, patient safety, diagnostic and screening processes, results, and follow-up, or other outcomes.3 Blood pressure, low-density lipoprotein cholesterol (LDL-C), and hemoglobin A1c test results account for just a few examples of the available codes. Though CPT Category II has been around for several years, its use in HEOR-related analyses is limited. We sought to quantify the presence of codes documenting laboratory test results in claim data for a commonly performed test, cholesterol, to demonstrate current CPT Category II utilization.
Figure 1: Standardized healthcare coding systems
| CPT1,2,3 | HCPCS4 | ICD-105,6 | |
|---|---|---|---|
| Description/ Purpose | Category I: Codes for medical services. Category II: Optional supplemental tracking codes for performance measurement. |
Level I: CPT Category I through the AMA Level II, Alphanumeric: Codes for medical services not included in Level I. |
Clinical Modification (CM): Factors influencing health status and contact with health services. Procedure Coding System (PCS): Inpatient treatment and services, only. |
| Governing Body | American Medical Association (AMA) | Centers for Medicare and Medicaid Services (CMS) | CM: Centers for Disease Control and Prevention (CDC) PCS: CMS |
| Organization/Management | Interested parties may submit application(s) to the AMA. The AMA may also develop codes to address ongoing requirements for documentation. | Interested parties may submit application(s) to CMS. CMS may also develop codes to support claim processing needs. | CM: Developed and maintained by CDC with authorization from the World Health Organization. PCS: Same as HCPCS Level II. |
| Examples | Category I: 83700, Lipoprotein electrophoresis/ phenotyping Category II: 3048F, LDL-C > 130 mg/dl |
Level II: G8736, LDL-C > 130mg/dL | E78.00: Pure hypercholesterolemia, unspecified |
| CPT = Current Procedural Terminology. HCPCS = Healthcare Common Procedure Coding System. ICD-10 = International Classification of Diseases, Tenth Revision. CPT copyright 2022 AMA. All rights reserved. Fee schedules, relative value units, conversion factors, and/or related components are not assigned by the AMA, are not part of CPT, and the AMA is not recommending their use. The AMA does not directly or indirectly practice medicine or dispense medical services. The AMA assumes no liability for data contained or not contained herein. CPT is a registered trademark of the AMA. | |||
Analysis
We analyzed Milliman’s Consolidated Health Cost GuidelinesTM Source Database (CHSD) and Medicare 5% Standard Analytic Files for 2017 through 2020, which collectively contain claim data for approximately 60 million lives in the United States, including those covered by commercial employer-sponsored insurance, Patient Protection and Affordable Care Act (ACA), Medicaid, Medicare fee-for-service (FFS), and Medicare Advantage plans.
We identified more than 6 million members with a CPT Category I code for a cholesterol test in 2020. We found a CPT Category II LDL-C test result code for 88,567 (1.4%) of these members. The presence of a CPT Category II code was generally consistent across the four years regardless of payer type (see Figure 2).
Figure 2: Number of continuously enrolled members with a CPT category I code for a cholesterol test and CPT category II code for LCL-C test result per 1,000 members overall and by payer type over time
| Payer type | 2017 | 2018 | 2019 | 2020 |
|---|---|---|---|---|
| Overall | 10.4 | 9.8 | 10.0 | 14.3 |
| Commercial | 11.6 | 11.1 | 12.0 | 17.7 |
| Medicaid | 15.2 | 16.7 | 13.8 | 10.6 |
| Medicare Advantage | 9.5 | 9.4 | 8.1 | 14.5 |
| Medicare Fee-for-Service | 4.6 | 2.3 | 1.5 | 1.4 |
| Other | 6.4 | 10.1 | 11.7 | 13.0 |
| CPT = Current Procedural Terminology. LDL-C = low-density lipoprotein cholesterol. CPT Category I codes for cholesterol test included 80061, 82465, 83700, 83701, 83704, 83718, 83721, 83722, and 84478. CPT Category II codes for LDL-C test rest included 3048F, 3049F, and 3050F. See Methods section below for additional detail. | ||||
Implications
In our analysis of a large research data set, we were able to identify a CPT Category II claim code documenting a laboratory test result for 1% to 2% of the members who had a cholesterol test performed. This suggests a very low utilization of CPT Category II codes currently. Furthermore, coding patterns vary by payer type and between years. The potential to expand the utility and consistent use of existing claim coding processes may offer a low-effort, high-value return for HEOR and related analyses, particularly if coding patterns can result in a representative sample of patients.
The most common and readily available large real-world data (RWD) sources are historical medical claim data that contain detail on enrollment, procedures performed and related diagnoses, and pharmacy transactions. They have not typically included the level of detail available in medical records, including laboratory test results. Conversely, the raw data feeds for test results entered manually into electronic health records (EHRs) may not be integrated directly into the data warehouses used for patient outcome assessment. Even if they are, values may be stored in both unstructured and structured data formats, which can further complicate data processing.
Standardized claim codes offer an alternative mode for documenting health status. We were not able to find published literature discussing the uptake of CPT Category II codes over time, nor of their use in observational analyses. Rather we found examples of efforts to increase the use of these codes, particularly among Medicaid providers.7 Some plan administrators are offering a financial incentive for adding these codes to a claim.8-10 Time will tell if these efforts meaningfully impact the use of these codes, at least by Medicaid providers.
Recently, the U.S. Department of Health and Human Services Food and Drug Administration (FDA) issued draft guidance for industry on the use of RWD, including EHRs, registry, and claim data, to support regulatory decision-making for drug and biological produces.11 CPT Category II codes for analyses outlined in the FDA’s guidance would ideally align with national standards on chronic disease management. The currently available options for LDL-C values are listed in Figure 3. Guidelines on cholesterol management recommend a reduction of 50% in LDL-C and outline specific therapy options for LDL-C that remains above 70 mg/dL with treatment.12 Thus, while the current level of granularity available may limit the precision of some studies, the effort to explore healthcare claims data for outcomes-related information may be less than that required for the integration of EHR and/or separate laboratory data feeds for evaluating healthcare interventions.
Figure 3: Administrative coding for low-density lipoprotein cholesterol (LDL-C)
| CPT category II1,2,3 | HCPCS level II4 | ||
|---|---|---|---|
| 3048F | LDL-C < 100 mg/dL | G8595 | LDL < 100 |
| 3049F | LDL-C between 100 and 129 mg/dL | G8597 | LDL >= 100 |
| 3050F | LDL-C >= 130 mg/dL | G8736 | LDL-C < 100mg/dL |
| G8737 | LDL-C >= 100mg/dL | ||
| G8890 | LDL-C under control | ||
| G8893 | LDL-C ‘not under control’ | ||
| G8891 | Documentation of medical reason(s) for most recent LDL-C not under control | ||
| G9271 | LDL under 100 | ||
| G9272 | LDL 100 and over | ||
| CPT = Current Procedural Terminology. HCPCS = Healthcare Common Procedure Coding System. HCPCS Level II codes were nearly absent for all payer types in our analysis (data not shown). CPT copyright 2022 American Medical Association. All rights reserved. | |||
Limitations
We analyzed 2017-2020 claim data, the most recent available, for a selected list of CPT Category II codes of interest. Findings may differ for other CPT Category II codes. The data set analyzed represents a sample of healthcare administrative claims data for nearly 60 million insured lives from all 50 U.S. states and may not be generalizable to all individuals with similar health insurance coverage nationally. It is possible that the payer and geographic distributions changed year to year. We did not measure coding patterns for individuals who are uninsured.
Methods
Payer types for individuals continuously enrolled during each year studied were categorized as commercial—health maintenance organization (HMO), preferred provider organization (PPO), ACA, and other—with upwards of 28 million enrollees; Medicaid (HMO, PPO, Medicaid-Medicare dual eligibility, other) with more than 4 million enrollees; Medicare Advantage with more than 5.5 million enrollees; and Medicare FFS with upwards of 3 million enrollees. Findings were not risk-adjusted or acuity-adjusted.
Endnotes
1AMA. CPT® (Current Procedural Terminology). Retrieved March 21, 2022, from https://www.ama-assn.org/amaone/cpt-current-procedural-terminology.
2AMA. Criteria for CPT® Category II codes. Retrieved March 21, 2022, from https://www.ama-assn.org/practice-management/cpt/criteria-cpt-category-ii-codes#:~:text=CPT%20Category%20II%20codes%20are,and%20other%20health%20care%20professionals.
3AMA. Criteria for CPT® Category II codes. Retrieved March 21, 2022, from https://www.ama-assn.org/practice-management/cpt/category-ii-codes.
44 CMS. HCPCS – General Information. Retrieved March 21, 2022, from https://www.cms.gov/Medicare/Coding/MedHCPCSGenInfo.
5NCHS. International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM). Retrieved March 21, 2022, from https://www.cdc.gov/nchs/icd/icd10cm.htm.
6CMS. ICD-10. Retrieved March 21, 2022, from https://www.cms.gov/Medicare/Coding/ICD10.
7Mississippi Division of Medicaid. Follow-up Information Regarding Use of CPT Category II Codes. Retrieved March 21, 2022, from https://medicaid.ms.gov/follow-up-information-regarding-use-of-cpt-category-ii-codes/.
8Health Plan of Nevada. CPT Category II Code Reimbursements. Retrieved March 21, 2022, from https://myhpnmedicaid.com/-/media/Files/HPNMedicaid/pdf/Provider/Forms-Memos-Letters/Medicaid-CPT-II-CODING-INCENTIVE-Memo-Jan2020.ashx?la=en&hash=2B4E32859315A26ADF506CBA52F84438.
9Anthem Medicaid. Re: CPT Category II Coding Incentive Program. Retrieved March 21, 2022, from https://providers.anthem.com/docs/gpp/KY_CAID_CategoryIIRequirements.pdf?v=202111182146.
10Healthy BlueSM. 2022 Quality Incentive Program. Retrieved March 21, 2022, https://provider.healthybluesc.com/docs/inline/SCHB_PE_QualityIncentiveProgram.pdf?v=202202251641.
11FDA. Real-World Data: Assessing Electronic Health Records and Medical Claims Data to Support Regulatory Decision-Making for Drug and Biological Products Draft Guidance for Industry. Retrieved March 21, 2022, from https://www.fda.gov/regulatory-information/search-fda-guidance-documents/real-world-data-assessing-electronic-health-records-and-medical-claims-data-support-regulatory.
122018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Blood Cholesterol: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Retrieved March 21, 2022, from https://www.ahajournals.org/doi/10.1161/CIR.0000000000000625.