Impact of Expanding Telephonic Codes in a State Medicaid Program During COVID-19

Abstract

Background:

The recent pandemic caused by the 2019 novel coronavirus (COVID-19) resulted in declaration of a national emergency (NE) in March 2020. The Centers for Medicare and Medicaid Services quickly responded with temporary expansion of telehealth coverage policies.

Aim:

To determine the impact of implementing a temporary telephonic code set in a state Medicaid population by comparing the utilization patterns of telehealth claims before and after a NE announcement.

Methods:

This was a retrospective cohort study conducted with the Arizona Medicaid program, the Arizona Health Care Cost Containment System (AHCCCS). Data include telehealth claims submitted to AHCCCS between January and May 2020 by contracted managed care organizations.

Results:

Approximately 2.3 million telehealth claims were analyzed in this study. Utilization of the audio-visual (A/V) modality increased 1,610% and telephonic visits increased 408% compared with pre-NE. Compared with pre-NE, three provider type groups increased their utilization of telephonic visits from 1.8% to 50.8% as a result of the temporary telephonic set post-NE. Rural counties had higher rates of telephonic modality adoption, whereas urban counties adopted the A/V modality more readily. Ten telephonic codes constituted 87% of all telehealth claims, with the majority of those codes used for behavioral health and established office visit types.

Conclusion:

The telephonic modality was adopted more frequently in rural areas and the A/V modality in urban areas. There were several new provider types utilizing telehealth as a result of the temporary telephonic code set implementation.

Introduction

Approximately 68 million individuals are enrolled in a Medicaid health plan in the United States.¹ Individuals insured by Medicaid include “low income families, qualified pregnant women and children, and individuals receiving Supplemental Security Income.”² Within this population, access to care is a major cause of health disparities and poorer health outcomes in general.^3

–6 Telehealth services, developed to bridge gaps in access to health care services, rely on several technological modalities to connect the patient to a provider: asynchronous “store and forward,” synchronous audio-visual (A/V) visits, and telephonic audio only.⁷ Historically, telehealth has been heavily relied upon in rural communities that lack access to local physicians and face barriers of distance, travel time, and ability to travel.^8

–12

Medicaid programs have been slow to adopt telehealth technology. In 2008, less than half of all state Medicaid programs offered telehealth for health care delivery.^13,14 Among programs with telehealth reimbursement policies, nearly all (95%) were limited to behavioral health services.¹⁴ The Arizona Medicaid program—Arizona Health Care Cost Containment System (AHCCCS)—has been reported to have the highest utilization rate with ∼60 claims per 10,000 enrollees as of 2009.¹⁴ Ten years later, in October 2019, AHCCCS expanded its telehealth coverage policy to provide for a broader range of clinical services.

The recent pandemic caused by the 2019 novel coronavirus (COVID-19) resulted in declaration of a national emergency (NE) in March 2020.¹⁵ The Centers for Medicare and Medicaid Services quickly responded with state-level guidelines for a temporary expansion of telehealth coverage policies in an effort to maintain access to health care services and to minimize the disruption of the pandemic and social distancing on the delivery of health care services.¹⁶

This resulted in the temporary addition of 122 telephonic codes to provide reimbursement for clinical visits (standard office visits, mental health evaluations, group counseling, home visits, etc.) delivered through telephonic technology. In March 2020, AHCCCS and other state Medicaid programs implemented the temporary telephonic code set. There are reports on the adoption of telehealth within Medicare^17,18 and commercially insured populations,¹⁹ but less is known about telehealth utilization within Medicaid.

This study examined the impact of the temporary code set on telehealth utilization in the AHCCCS Medicaid program in the months after the declaration. Use of telehealth during the pandemic (post-NE) was compared with utilization patterns before the NE (pre-NE). The objectives of this analysis were to examine the telehealth utilization rates by county (rural vs. urban), provider, and service type, and to identify the temporary telephonic codes with high versus rare utilization. The results of this study may be used to inform policy decisions about telehealth coverage and the permanent inclusion of the Centers for Medicare & Medicaid Services temporary telephonic codes.

Materials and Methods

This is a retrospective cohort study of telehealth claims from a state Medicaid program that includes seven Managed Care Organizations (MCOs). The claims database was queried to identify claims between January and May 2020 of telehealth encounters with one of the following code modifiers: asynchronous (GQ modifier), A/V (GT modifier), and telephonic (UD modifier).²⁰ Out-of-state claims and claims without a county designation were excluded. The telehealth abstracted data included the encounter code, code modifiers, MCO, line of business, county, visit month, place of service, provider type, claim volume, and completion factor (defined in a subsequent paragraph).

County metropolitan designations for rural versus urban setting were based on information from the office of management and budget.²¹ Provider types were based on the source of billing, which included groups (Federally Qualified Health Centers [FQHCs], etc.) or individuals not a part of those groups (identified as “physicians,” “APPs” [advanced practice providers], etc.). These determinations are mutually exclusive (i.e., someone identified as a physician did not belong to a group such as an FQHC).

To assess the impact of the temporary telephonic code set on telehealth utilization, the data set was split into two time periods: before (pre-NE) and after the implementation of the temporary code set in March 2020 (post-NE). The pre-NE group wa claims from 2 months before the pandemic (January and February 2020). The post-NE group was claims for 2 months after the NE declaration in March (April and May 2020).

The month of March was excluded from both designations due to the mixture of code sets during transition. The post-NE group included all claims submitted by August 2020. The completeness of claims submission by August 2020 varied between the MCOs. To account for the variability in completeness, the volume of claims for each MCO was corrected using the MCO-specific completion factor.²² The completion factor, calculated by each MCO, estimates the percentage of claims submitted based on historical counts of submitted claims for the same month.

Data were summarized using median (interquartile range [IQR]) and proportions. The IQR provides the 25th percentile and the 75th percentile and can be interpreted as the range of values observed around the median (50th percentile) value. The chi-squared (χ ²) tests were performed to compare the utilization rates between the pre- and post-NE provider types. Data were stratified for comparisons between rural and urban counties and at the provider and service levels.

The difference in utilization rates of modalities was quantified by dividing usage post-NE by the usage pre-NE. In addition, data were analyzed to identify the most and rarely utilized codes within the temporary telephonic code set with subanalyses done at the provider level and by month. Rare utilization was defined as fewer than five claims per month. Data were analyzed with Stata SE 14.0 (College Station, TX), and visualizations generated with GraphPad (San Diego, CA).

This project did not require human subjects review per the University of Arizona's Institutional Review Board (Protocol No. 2008909226).

Results

There were 314,000 telehealth claims submitted in the pre-NE time period. In the post-NE group, there were 1.99 million claims. Utilization of the A/V modality increased 1,610% and telephonic visits increased 408% compared with that pre-NE. The data set represents claims from 52 different provider types and all 15 counties (60% rural). The per county claim counts ranged from 1,300 claims to 1.5 million claims.

CODE UTILIZATION BY PROVIDER TYPE

There was a shift in the provider types that utilized telehealth visits in the post-NE time period. Before the NE, the predominant users of telehealth were outpatient behavioral health clinics (46% of all telehealth claims) and integrated clinics (41%), followed by APPs (8%). Post-NE, two other groups emerged as major utilizers of telehealth: FQHCs and physicians at 8% each, compared with 39%, 27%, and 9%, respectively, of the aforementioned categories. Of note, although physicians, FQHCs, and APPs made up only ∼1.8% of telephonic code set usage pre-NE, they collectively accounted for 50.8% of users of the new temporary telephonic code set.

CODE UTILIZATION BY COUNTY

The percentage change in A/V and telephonic utilization percentages varied between urban and rural counties, which is presented in Figure 1. Urban counties experienced a median increase of 73% in the utilization of A/V modality, with a range of 36% (25th percentile) up to 238% (75th percentile). In rural counties, there was an overall median decrease (−26%) in utilization of A/V (IQR: −47%, 6%). The greatest increases in A/V utilization were in the urban counties of Maricopa (329%) and Pima (239%). The opposite occurred in the utilization rates of telephonic modality in urban and rural counties; respectively, −19% (IQR: −24%, −18%) and 8% (IQR: −3%, 12%). The increase in telephonic utilization was greatest in Navajo county (196%).

Fig. 1.

Percentage changes in A/V and telephonic claim rates by county and metropolitan designation. *Greater than 100%. A/V, audio-visual.

HIGH VERSUS RARE CODE UTILIZATION

Of the 122 codes in the temporary telephonic code set, 10 (8%) codes accounted for 87% of the claims (Fig. 2). The majority of these highly utilized codes were for behavioral health and established office visit types. Forty (33%) temporary telephonic codes were rarely used and included independent codes or codes within a series (Fig. 3). The rarely used codes could be classified into five main categories (not mutually exclusive): neuro/psych (n = 16), group/family (n = 11), prolonged visit (n = 8), home visit (n = 7), and miscellaneous (n = 6).

Fig. 2.

Top 10 temporary telephonic codes by claim volume.

Fig. 3.

Rarely used temporary telephonic codes.

Discussion

The NE of the COVID-19 pandemic incited a dramatic increase in telehealth utilization, with one study reporting that telehealth visits increased from <1% of all visits to up to 70% of total visits in certain practices.²³ Our study found a dramatic increase in the volume of telehealth utilization.

Compared with the pre-NE time period, the volume of telehealth claims increased 1,610% for A/V and 408% for telephonic visit claims. The temporary code set was entirely telephonic, yet the increase in telehealth was dominated by the rise in use of A/V for telehealth. This predominance could be explained possibly by a couple of factors. First, there may have been a delay in providers and coders adopting the temporary telephonic codes due to confusion on the appropriate usage.

Second, the office of civil rights of the department of health and human services granted temporary allowances for the use of previously disallowed A/V platforms such as FaceTime and Skype.¹⁶ The expanded use of additional platforms would have contributed to the increase in the utilization of this modality. Telehealth codes for A/V currently do not distinguish the platform used, therefore, the role of additional platforms could not be assessed. To address the role of A/V platform options in utilization patterns, telehealth policies should consider additional modifications to the A/V code.

Overall, telehealth in urban counties had a greater increase in A/V utilization than telephonic. The converse occurred in rural counties with the use of telephonic exceeding A/V. The difference reflects a potential disparity in the availability of technical resources such as A/V technology, limited broadband services affecting A/V quality/transmission, and a lack of appropriate auxiliary equipment in rural settings.^24,25 As a result, patients and providers turn to less complex yet more readily available methods of communication such as a cell phone or landline. This likely explains the extreme increase in telephonic utilization for Navajo county.

Behavioral health-centric services have long been the predominate user of telephonic visits. After the NE, the profile of visit types shifted to both behavioral health-centric and physical health-centric services. There were three provider groups (FQHCs, APPs, and physicians) that experienced the greatest increase in telephonic usage, from 1.8% pre-NE to >50% post-NE. This would indicate that implementation of the temporary telephonic code set increased the availability of telehealth options to providers whom otherwise would not have been allowed to provide telephonic-based care.

The utilization rates of the temporary telephonic codes were analyzed to distinguish the predominant codes from those rarely used. Ten (8%) of the temporary codes comprised 87% of all visit claims. One-third of the temporary codes were rarely or never used. There may be several reasons for these differences. First, the most frequently used codes in A/V were already common for in-person visits.

Second, codes that were rarely used may be due to four possible reasons: (1) barriers to utilization of that type of code in the telephonic space (i.e., 9934x—home visits, 99411–99412—group counseling), (2) a lag in the adoption of certain codes that were added during the NE, (3) some codes are used to capture encounter time that are not picked up elsewhere (i.e., 99358–99359—prolonged evaluation and management), or (4) codes that involve higher levels of protection of privacy, especially during a pandemic (S9480—intensive outpatient psychiatric services). Since our analysis was limited to 2 months after the NE, long-term monitoring of code usage is necessary to confirm the codes with rare utilization rates.

From a policy perspective, this analysis has several potential implications. First, given that rural counties trended toward more telephonic usage, this may indicate that there is utility in considering converting portions of the temporary telephonic code set to the permanent codes for rural counties after the COVID-19 pandemic. Second, given the dramatic increase in telephonic claims from some provider categories, there may be utility in keeping a portion of the temporary code set to maintain telehealth access by these provider types.

Third, continued evaluation of rarely used codes as well as monitoring for major changes in code utilization over time would be beneficial to determine which codes to sunset either from a health care outcome perspective or from a cost containment perspective. However, there are always logistical concerns when dealing with new technology or implementing existing technologies, especially with regard to how to incorporate into existing workflows and reimbursement standards. This is concerning for some medical practices since telehealth reimbursement has typically been lower than in-person visits for the same level of service.²⁶ The consequence is risk of financial insolvency due to the increased overhead costs associated with providing both in-person and telehealth at acceptable capacities.²⁷

This study was subject to several limitations. The reporting rates varied between MCOs, with some MCOs experiencing longer lag times in reporting and different rates of completeness. To adjust for these differences, the data were standardized using the MCO predetermined completion factors. Although this is not a perfect method given that pre-NE and post-NE rates may differ, this method was the best available alternative. Some visit codes were submitted as blank or “unknown” and were excluded from the analysis. Moreover, some claim lines included both A/V and telephonic modifiers. Blank and mixed visit claims were infrequent, equally likely across MCOs, and, therefore, an unlikely source of bias.

There is also the possibility that some visits were misclassified, such as an A/V visit being coded as a telephonic visit and vice versa. Lastly, in an ideal situation, the post-NE timeframe would have been matched with the same time period from the previous year to control for issues such as seasonal disease burden and weather-related migratory patient volume (i.e., “snowbirds”). However, this comparison was not possible given the telehealth policy update in October 2019. There was minimal variability in the monthly telehealth claim volume after the October 2019 update through February 2020, and consequently the most recent two months were used to match with the post-NE time period.

Conclusions

Telehealth, including its telephonic modality, increased rapidly in response to health care needs during the COVID-19 pandemic. The adoption of telephonic health care delivery specifically increased more so in rural areas than in urban areas. In addition, several new provider groups were able to utilize telephonic services to provide health care with the new code set.

However, not all codes in the temporary telephonic code set necessarily should be maintained given that 87% of claims came from the top 10 codes and 33% of codes were used fewer than 5 times per month. From a policy perspective, there may be some benefit to maintaining a least portion of the temporary telephonic code set in more rural areas as well as sunsetting rarely used codes to decrease unnecessary administrative burden if they continue to be rarely used.

Footnotes

Authors' Contributions

All listed authors made substantial contributions to the conception and design, or acquisition of data, or analysis and interpretation of data; drafting the article or critical revisions; and have given final approval of the version to be submitted.

Acknowledgments

We thank William Buckley and Rhonda Ellison from AHCCCS for their assistance in this project.

Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.

References

June. 2020. Medicaid & CHIP Enrollment Data Highlights | Medicaid. Available at https://www.medicaid.gov/medicaid/program-information/medicaid-and-chip-enrollment-data/report-highlights/index.html (last accessed October 15, 2020 ).

Eligibility | Medicaid. Available at https://www.medicaid.gov/medicaid/eligibility/index.html (last accessed February 1, 2021 ).

Clark

, Ommerborn MJ

A C

, oull

, Pham

, Haas

. Income inequities and Medicaid expansion are related to racial and ethnic disparities in delayed or forgone care due to cost. Med Care, 2016; 54:555–561.

Okoro

, Zhao

, Fox

, Eke

, Greenlund

, Town

. Surveillance for health care access and health services use, adults aged 18–64 years—Behavioral Risk Factor Surveillance System, United States, 2014. MMWR Surveill Summ, 2017; 66:1–42.

Yue

, Rasmussen

, Ponce

. Racial/ethnic differential effects of Medicaid expansion on health care access. Health Serv Res, 2018; 53:3640–3656.

Seo

, Baggett

, Thorndike

, et al. Access to care among Medicaid and uninsured patients in community health centers after the Affordable Care Act. BMC Health Serv Res, 2019; 19:291.

Telehealth

Programs

. Official web site of the

U.S

. Health Resources & Services

Administration.

2017. Available at https://www.hrsa.gov/rural-health/telehealth (last accessed February 2, 2021 ).

Combi

, Pozzani

, Pozzi

. Telemedicine for developing countries. A survey and some design issues. Appl Clin Inform , 2016; 7:1025–1050.

Oest

SER

, Swanson

, Ahmed

, Mohr

. Perceptions and perceived utility of rural emergency department telemedicine services: A needs assessment. Telemed J E Health, 2020; 26:855–864.

10.

Baker

, Stanley

. Telemedicine technology: A review of services, equipment, and other aspects. Curr Allergy Asthma Rep, 2018; 18:60.

11.

Ekeland

, Bowes

, Flottorp

. Effectiveness of telemedicine: A systematic review of reviews. Int J Med Inform, 2010; 79:736–771.

12.

Downes

, Mervin

, Byrnes

, Scuffham

. Telephone consultations for general practice: A systematic review. Syst Rev, 2017; 6:128.

13.

Lin

CCC

, Dievler

, Robbins

, Sripipatana

, Quinn

, Nair

. Telehealth in health centers: Key adoption factors, barriers, and opportunities. Health Aff (Millwood), 2018; 37:1967–1974.

14.

Douglas

, Xu

, Heggs

, Wrenn

, Mack

, Rust

. Assessing telemedicine utilization by using Medicaid claims data. Psychiatr Serv, 2017; 68:173–178.

15.

Proclamation of declaring a national emergency concerning the novel coronavirus disease (COVID-19) outbreak. 2020. Available at https://trumpwhitehouse.archives.gov/presidential-actions/proclamation-declaring-national-emergency-concerning-novel-coronavirus-disease-covid-19-outbreak (last accessed October 29, 2021 ).

16.

Medicare Telemedicine Health Care Provider Fact Sheet |

CMS

. Available at https://www.cms.gov/newsroom/fact-sheets/medicare-telemedicine-health-care-provider-fact-sheet (last accessed September 27, 2020 ).

17.

Whaley

, Pera

, Cantor

, et al. Changes in health services use among commercially insured US populations during the COVID-19 pandemic. JAMA Netw Open, 2020; 3:e2024984.

18.

Office of the Assistant Secretary for Planning and Evaluation

HHS.

Office of the Assistant Secretary for Planning and Evaluation, HHS. ASPE Issue Brief: Medicare Beneficiary Use of Telehealth Visits: Early Data from the Start of the COVID-19 Pandemic. HHS, 2020. Available at https://aspe.hhs.gov/pdf-report/medicare-beneficiary-use-telehealth#:∼:text=the%20analysis%20found%20medicare%20primary,before%20the%20phe%20in%20february (last accessed March 23, 2022 ).

19.

Cantor

, McBain

, Pera

, Bravata

, Whaley

. Who is (and is not) receiving telemedicine care during the COVID-19 pandemic. Am J Prev Med, 2021; 61:434–438.

20.

Table II AHCCCS Telephonic Code Set Permanent from Medical Coding Resources. Available at https://www.azahcccs.gov/PlansProviders/MedicalCodingResources.html (last accessed December 16, 2020 ).

21.

AZDHS | Data Reports and Maps. Arizona Department of Health Services. Available at http://www.azdhs.gov/prevention/health-systems-development/data-reports-maps/index.php (last accessed September 28, 2020 ).

22.

Ryan

, Clay

. An overview of IBNR (incurred but not recorded). Healthc Financ Manage, 1994; 48:18–19.

23.

Wosik

, Fudim

, Cameron

, et al. Telehealth transformation: COVID-19 and the rise of virtual care. J Am Med Inform Assoc, 2020; 27:957–962.

24.

Paul

, Pearlson

, McDaniel

. Assessing technological barriers to telemedicine: Technology-management implications. IEEE Trans Eng Manage, 1999; 46:279–288.

25.

A Rural Perspective on COVID-19: Yavapai

County

, Arizona | telemedicine.arizona.edu. Available at https://telemedicine.arizona.edu/blog/rural-perspective-covid-19-yavapai-county-arizona (last accessed September 30, 2020 ).

26.

Aremu

, Heffernan

, Kvedar

. The difference in practice expense costs between telehealth and in-office care could serve as the basis for differential reimbursement structures. Telemed J E Health, 2022; 28:912–916.

27.

Wilson

, Rampa

, Trout

, Stimpson

. Reimbursements for telehealth services are likely to be lower than non-telehealth services in the United States. J Telemed Telecare, 2017; 23:497–500.