Billing for Tobacco Cessation: Enhancing Data Quality and Revenue Capture

Abstract

OBJECTIVES:

This study aimed to quantify the economic impact of missed billing opportunities for tobacco cessation counseling at an academic medical center to identify what may be a systematic defect in the administration of tobacco cessation services and to highlight opportunities to improve patient outcomes and revenue. Patient surveys show that evidence-based tobacco cessation interventions are provided at low rates despite guidelines supporting the use of these services at every eligible encounter.

STUDY DESIGN:

Retrospective cohort study.

METHODS:

The study analyzed deidentified patient health data from electronic health records at an academic medical center, focusing on primary care encounters from January 1, 2020, to December 31, 2023, involving patients 18 years and older with a history of current tobacco use. Billing data for tobacco cessation counseling (Current Procedural Terminology codes 99406 or 99407) were examined to estimate revenue loss from unbilled eligible encounters.

RESULTS:

Of 1,068,875 primary care visits, 16.8% (179,304) involved tobacco users. However, only 1.0% of these encounters were billed for cessation services, representing an estimated potential revenue loss of $3.2 million over 4 years.

CONCLUSIONS:

These findings identify a significant discrepancy between the billing of tobacco cessation services and the opportunities to do so. Better provision and billing of tobacco cessation counseling can improve patient health outcomes, advance value-based care goals, and enhance financial sustainability.

Notable progress has been made in reducing the prevalence of cigarette smoking among US adults to 9.9%, yet more than 25.2 million adults continue to smoke cigarettes regularly.¹ Data from the most recent CDC National Health Interview Survey (2022) indicated that 67.7% of individuals who smoked were interested in quitting and that 53.3% had attempted to quit in the past year.² However, 38.3% of those attempting to quit smoking utilize evidence-based interventions such as FDA-approved cessation medications or behavioral counseling. This shortfall in intervention use contributes to high relapse rates, with some individuals requiring up to 20 attempts to quit successfully.^3,4 Cigarettes comprise the largest component of tobacco use, so much of the research has been focused on smoking.⁵ However, 18.7% of adults use some form of tobacco regularly.⁵ Among the estimated 46 million tobacco users, 6.5% use e-cigarettes, 3.5% smoke cigars, 2.1% use smokeless tobacco, and 3.4% report using 2 or more tobacco products.^5,6 All tobacco users are eligible for and would benefit from counseling^7,8 and FDA-approved medications⁹; these services are covered by insurance with no cost sharing.¹⁰

Annual surveys conducted by the National Committee for Quality Assurance (NCQA) show that from the patient’s perspective, physicians are advising patients to quit smoking or using tobacco but not consistently directing them to evidence-based treatment.¹¹ For example, NCQA surveys from 2021 revealed that 44% of commercial health maintenance organization (HMO) members who were current smokers or tobacco users did not receive counseling on how to quit and less than 40% had discussions about medications.¹¹ Sardana et al reported that in a large contemporary US registry, 1 in 3 people who used cigarettes and presented for a cardiology visit received smoking cessation assistance.¹² Another study analyzed Optum’s longitudinal ambulatory data across 22 health care organizations between January 2012 and December 2018 to evaluate quit rates for individuals who used cigarettes; adjusted quit rates ranged from 14.3% to 34.5% across 20 health systems, 5% to 66% among 1399 practice sites, and 4% to 87% among 3803 health care providers.¹³ In addition, multiple reports examining administrative data revealed significant variation and overall underperformance in providing tobacco cessation services, which were attributed to barriers such as time constraints and low reimbursement.^14–16

Wide variation in use of cessation services is also demonstrated by multistate Medicaid claims research showing that among 37 states, 0.2% to 32.9% (mean, 9.4%) of fee-for-service (FFS) Medicaid smokers with a past-year attempt to quit had claims for cessation medications.¹⁷ On average, 2.7% of these patients received cessation counseling (range, 0.1%−5.6%).¹⁷ Furthermore, a 3-year study of electronic health records (EHRs) at a multihospital health system showed that only 0.3% of encounters with individuals using tobacco were billed for cessation services.¹⁸ Total potential revenue lost, estimated based on assuming each person identified as using tobacco (ie, eligible for tobacco cessation counseling) is appropriately treated and their insurer billed for the service, was more than $5.9 million.¹⁸

Screening for tobacco use and advising a patient to quit are well-established, evidence-based standards of care. Clinical practice guidelines support the delivery of a billable, evidence-based intervention at every eligible clinical encounter, regardless of a patient’s readiness to change.¹⁹ Counseling people who use tobacco for 3 minutes or 10 minutes is a billable service.⁸ Moreover, repeated counseling and interventions with scheduled follow-up visits substantially increase the success of tobacco cessation efforts.²⁰

Much of what is known about clinical performance in tobacco cessation is gleaned from survey data from cigarette users.² EHR studies can increase the level of insight into the processes and outcomes of tobacco cessation interventions.^13,18 Notwithstanding the public health implications, ineffective billing practices for cessation services can be considered an indicator of inefficient workflow, poor quality of care, underperformance of medical claims billing, or, more likely, a combination of these. The economic impact of consistent tobacco cessation counseling is even more significant for value-based contracts because former smokers have lower medical expenses.

This study aimed to improve the understanding of the economic impact of missed opportunities for tobacco cessation counseling using EHR data. By focusing on primary care encounters potentially eligible for billing codes for intermediate and intensive tobacco cessation counseling, we sought to use Current Procedural Terminology (CPT) codes 99406 and 99407 as an indicator for the overall delivery of cessation services. The objective of studying billing practices was to identify what may be a systematic defect in the administration of tobacco cessation services.

METHODS

Study Design, Population, and Data Sources

This retrospective cohort study utilized deidentified patient health data extracted from the Vanderbilt University Medical Center (VUMC) EHR and claims database. VUMC is the largest academic medical center in Tennessee, with 1741 licensed beds across 7 hospitals and 24 outpatient facilities treating 3.2 million patients annually in more than 180 ambulatory locations.²¹

The inclusion criteria (TABLE 1²²) were patients 18 years and older who had ambulatory visits with primary care providers from January 1, 2020, to December 31, 2023. Primary care encounters were identified using department-level classifications (primary care or internal medicine) and were restricted to completed office visits and telemedicine visits. Inpatient, emergency department, procedural-only, and non–primary care specialty visits were excluded. The unit of analysis for billing and revenue calculations was the encounter. The study was exempt from full review by Vanderbilt’s institutional review board (IRB #232175). There was no external funding to support this research.

TABLE 1.

Summary of Eligibility Parameters for Tobacco Cessation Care Billing

Criteria	Details
Eligible encounter codes	Level 2–5 evaluation and management codes, qualifying Current Procedural Terminology or Healthcare Common Procedure Coding System codes22
Tobacco product use	Current use of any tobacco product, including but not limited to cigarettes, cigars, dissolvable, nicotine gels, hookah tobacco, pipe tobacco, roll-your-own tobacco, smokeless tobacco products, vapes, e-cigarettes, hookah pens, and other electronic nicotine delivery systems; or coding of an encounter for tobacco cessation; or prescription of a medication for tobacco cessation

Data Mining, Extraction, and Processing

Demographic data were mined with SlicerDicer (Epic Systems Corporation); visit/encounter and billing data were extracted from the Epic Clarity database (Epic Systems Corporation). SQL (Structured Query Language) queries were tailored to identify patients with a history of current tobacco use. The following steps outline the data processing (full details of the data processing have been described previously¹⁸).

Identification of tobacco users.

Tobacco use status was identified for each ambulatory encounter based on the current or most recent status recorded. Patients met 1 or more of the following criteria: (1) International Statistical Classification of Diseases, Tenth Revision codes indicative of current nicotine dependence, tobacco use of any kind inclusive of vaping, chewing, or dual use; (2) tobacco use identified from the EHR problem lists, encounter diagnoses, and billing records; (3) coding of an encounter for tobacco cessation; or (4) a prescription for a medication with an indication for tobacco cessation.

Encounter selection.

Eligible encounters were those involving primary care clinicians in outpatient or telemedicine settings during the treatment years (1/1/20–12/31/23). Patients 18 years and older on the encounter date were included. Inpatient and emergency department visits and other ambulatory visits were excluded.

Billing data.

Tobacco cessation counseling CPT codes, 99406 or 99407, determined whether the encounter was billed. The difference between these 2 codes is the duration of the counseling session: 99406 indicates brief counseling (sessions of 3–10 minutes), and 99407 indicates intensive counseling (sessions lasting > 10 minutes).

Insurance reimbursement.

Potential revenue from missed billing opportunities was classified using Epic financial class mappings. Missed billing opportunity was defined as an encounter involving a tobacco user without a recorded CPT 99406 or 99407 charge. To ensure conservative potential revenue estimations, the standard 2023 Medicare Physician Fee Schedule for CPT code 99406 (brief tobacco cessation counseling) was used as the base rate. Reimbursement rate modeling adjustments for commercial insurance (1.4 times Medicare rate) and Medicaid (0.7 times Medicare rate) represented conservative estimates of payment for a large number of diverse payers.^23,24 Patients with insurance other than Medicare, Medicaid, or commercial coverage or who were uninsured were deemed not eligible and excluded from calculations of missed billing opportunities. All insurance contracts were modeled as FFS.

Statistical Analysis

The prevalence of tobacco use in the patient cohort was calculated by dividing the number of identified tobacco users by the total study population for each year and in aggregate. The percentage of visits billed was computed by dividing the number of billed encounters by the total number of eligible encounters. Potential revenue loss was estimated based on the number of unbilled eligible encounters and the applicable reimbursement rates, representing the dollar value of missed billing opportunities for counseling if all contracts with insurers were FFS.

RESULTS

The primary care population included 248,192 patients with 1,068,875 encounters from January 1, 2020, to December 31, 2023. As shown in TABLE 2, most visits were for patients identified as White (77.4%) and non-Hispanic (72.1%). There was a female predominance (62.0%), and most patients were 45 years and older (64.2%). The sum of percentages for office and telemedicine visits may exceed 100% because patients may have had both an office visit and a telemedicine visit during the study period. Most of the patients had commercial insurance (74.8%).

TABLE 2.

Encounter Data Extracted From Epic Clarity Database, Vanderbilt Health (1/1/20–12/31/23)

	Office		Telemedicine		Total
Characteristic	n	%	n	%	n	%
Race
White	740,983	77.0%	86,371	80.7%	827,354	77.4%
Black or African American	118,334	12.3%	11,673	10.9%	130,007	12.2%
Unknown, declined, or not reported	47,331	4.9%	4335	4.1%	51,666	4.8%
Asian	21,143	2.2%	1346	1.3%	22,489	2.1%
Other	20,140	2.1%	1558	1.5%	21,698	2.0%
Multiple	11,459	1.2%	1444	1.3%	12,903	1.2%
American Indian, Alaska Native, or Native Hawaiian or Other Pacific Islander	2479	0.3%	279	0.3%	2758	0.3%
Ethnicity
Not Hispanic or Latino	693,777	72.1%	76,960	71.9%	770,737	72.1%
Unknown, declined, or not reported	225,110	23.4%	25,934	24.2%	251,044	23.5%
Hispanic or Latino	42,982	4.5%	4112	3.8%	47,094	4.4%
Sex
Female	589,327	61.3%	73,652	68.8%	662,979	62.0%
Male	372,501	38.7%	33,351	31.2%	405,852	38.0%
Unknown	41	0.0%	3	0.0%	44	0.0%
Age in yearsa
18–29	159,664	15.0%	18,746	15.6%	178,410	15.1%
30–44	210,043	19.8%	34,208	28.5%	244,251	20.7%
45–64	351,583	33.1%	39,020	32.5%	390,603	33.1%
≥ 65	339,846	32.0%	27,950	23.3%	367,796	31.1%
Insurance
Commercial (eg, Highmark, Blue Cross)	714,033	74.2%	85,282	79.7%	799,315	74.8%
Medicare	193,998	20.2%	16,232	15.2%	210,230	19.7%
Medicaid	14,517	1.5%	1596	1.5%	16,113	1.5%
Other	39,321	4.1%	3896	3.6%	43,217	4.0%
Total	961,869	90%	107,006	10%	1,068,875	100%

^aAge values can show discrepancies of approximately 5% between cohorts, primarily because some ages are recorded in months and may not be accurately accounted for in the electronic health record. Additionally, certain ages can change during the measurement time frame, so sum totals and cumulative percentages exceed 100%.

The mean prevalence of tobacco use, as previously defined, over the 4-year study was 15.9%. A total of 1,068,875 visits were evaluated, with tobacco users accounting for 16.8% (179,304) of all visits. Billing for cessation services was not commonly reported; just 1.0% of eligible visits were billed. Data on tobacco use from the perspective of all patients and all encounters are presented in TABLE 3.

TABLE 3.

Primary Care Tobacco Cessation Counseling by Year

	2020	2021	2022	2023	Total	Mean
Patients
Total	108,605	135,403	148,323	160,728	553,059	138,264.75
Tobacco user	13,951 (12.8%)	20,057 (14.8%)	24,585 (16.6%)	29,094 (18.1%)	87,687 (15.9%)	21,921.75 (15.9%)
Billed	443 (3.2%)	431 (2.1%)	297 (1.2%)	367 (1.3%)	1538 (1.8%)	384.5 (1.8%)
Visits
Total	208,255	262,944	283,773	313,903	1,068,875	267,218.75
Tobacco user	28,145 (13.5%)	41,173 (15.7%)	49,897 (17.6%)	60,089 (19.1%)	179,304 (16.8%)	44,826 (16.8%)
Billed	543 (1.9%)	525 (1.3%)	349 (0.7%)	440 (0.7%)	1857 (1.0%)	464.25 (1.0%)

TABLE 4 shows the modeling for reimbursement of payable visits by insurance type, which includes Medicare, Medicaid, and various commercial insurers. Visits by patients with no insurance were excluded. Potential revenue lost across payer types totaled $3.2 million over the 4-year study time frame, with a mean of $805,553.46.

TABLE 4.

Estimated Reimbursement for Missed Tobacco Cessation Billing Opportunities Based on Payer Type for Tobacco Users

Payer	2020 encountersa (potential revenue)	2021 encountersa (potential revenue)	2022 encountersa (potential revenue)	2023 encountersa (potential revenue)	Total encountersa (potential revenue)	Mean encountersa (potential revenue)
Medicare	8282 ($123,981.54)	12,282 ($183,861.54)	15,045 ($225,223.65)	16,719 ($250,283.43)	52,328 ($783,350.16)	13,082 ($195,837.54)
Medicaid	668 ($7000.64)	1434 ($15,028.32)	1574 ($16,495.52)	1684 ($17,648.32)	5360 ($56,172.80)	1340 ($14,043.20)
Commercial	17,527 ($367,365.92)	25,314 ($530,581.44)	31,316 ($656,383.36)	39,521 ($828,360.16)	113,678 ($2,382,690.88)	28,419.5 ($595,672.72)
Total	26,477 ($498,348.10)	39,030 ($729,471.30)	47,935 ($898,102.53)	57,924 ($1,096,291.91)	171,366 ($3,222,213.84)	42,841.5 ($805,553.46)

^aEach encounter represents an estimation of missed revenue based on eligibility for Current Procedural Terminology code 99406, assuming either that counseling occurred but was not billed or that counseling did not occur when it should have.

DISCUSSION

The substantial gap between the prevalence of tobacco users (15.9% 4-year mean) and the low proportion of billed cessation services (1.0% 4-year mean) highlights significant missed opportunities to enhance patient outcomes through counseling and to improve revenue. Repetition, reinforcement, and follow-up of tobacco cessation interventions have been shown to increase quit rates.¹⁹

In 2023, the prevalence of tobacco use in Tennessee was reported at 19.9%, with Nashville having a rate of 21.0%.^21,25 Thus, the calculated prevalence of 15.9% in this study may indicate that not all tobacco users were identified. However, the progressive increase in prevalence of tobacco use in the study data from 12.8% in 2020 to 18.1% in 2023 could indicate an improvement in screening for tobacco use.

Although the data in this study does not directly capture counseling delivery independent of billing, national benchmarks suggest that cessation interventions occur more frequently than reflected by CPT billing. In the National Health Interview Survey results, 7.3% were counseled and 36.3% used medications,² suggesting that a substantial portion of unbilled visits are likely reimbursable. Moreover, the NCQA data reveal that more than 50% of tobacco users receive advice to quit.¹¹ If the observed billing gap is the result of unmet documentation standards, this suggests that opportunities to provide cessation services could be missed entirely. Since repetition and increasing intensity of cessation services are likely to enhance both clinical and financial outcomes, the observed billing frequency (1.0% of eligible encounters) is inconsistent with expected clinical practice. More likely, this reflects a combination of missed counseling opportunities, incomplete documentation, and/or failure to submit charges. Improved revenue generation, however, depends on the payer contracts. Our reimbursement estimates were based on FFS payment for all insurance types, so if all missed opportunities were billed, the additional revenue over the 4-year study period would be $3.2 million.

To support a business case that approaches this level of economic return, the complete costs need to be identified. This involves many considerations, most significantly use of time by the clinical staff. Other barriers frequently identified in surveys on implementation of tobacco cessation services include lack of training, perceived lack of efficacy, low patient motivation, insufficient use of health information technology, and low rates of reimbursement.²⁶

Any tobacco cessation program selected should be rooted in a strong systems-change protocol—for example, based on elements outlined in the CDC’s Tobacco Cessation Change Package,²⁷ which is part of the American Heart Association’s Million Hearts initiative. Systems change for tobacco cessation is widely supported,^28,29 although adoption has been slow.³⁰ Several models for system change are emerging and include leadership buy-in, training, optimal use of the EHR, electronic referrals to state quit lines, and detailed program evaluation.

Revenue estimates must be tailored for institutions operating in value-based contracts, as calculations may require adjustments to remove patients with these types of provider contracts. National data suggest that up to 40% of Medicare Advantage and 17% of commercial plans share the risk of excess medical expense and the savings associated with improved health status.³¹ Because tobacco cessation is one of the few preventive services that is both cost-effective and produces a return on investment, its impact may be even greater in value-based contracts.^32,33

Mundt et al highlighted this point for value-based contracts from both clinical and economic perspectives. They described a Medicaid HMO in Wisconsin that invested in an intensive, clinic-based tobacco cessation program that included staff for face-to-face counseling.³⁴ Quit rates increased 3-fold over baseline.³⁵ Program expenses per user of cigarettes, including clinician time, cessation counselor salaries, and medication costs, averaged $3.80 per member per month, and a reduction of $42 per cigarette smoker per month was demonstrated in the first year.³⁴ Additional savings are likely to accrue in future years.

Levy et al examined the cost-effectiveness of a different model relying entirely on telephonic services.³⁶ A detailed model for capturing cost was also developed.³⁷ The cost model can be adopted to a variety of settings and can capture detailed expenses such as salary, space, outreach, and training. One important caveat to the findings of both Mundt et al³⁴ and Levy et al³⁶ is the need to adjust pharmacy costs to zero because medications are generally carved out of both FFS and value-based contracts. If such costs are included in the contract, then varenicline (Chantix) expenses should be adjusted to reflect its availability in generic form.

Time pressure on clinicians has been consistently identified as a barrier to delivery of effective treatment, which is why the focus on systems change adds professionally trained staff and resources. In both Levy et al³⁶ and Mundt et al,³⁴ tobacco treatment specialists provided telephonic or face-to-face counseling.

The use of EHR review to assess tobacco cessation billing can generate significant insights into the performance of cessation interventions. Furthermore, the examination of variation in provider performance at the clinic or individual provider level can lead to well-targeted improvement opportunities.¹³

Health systems should not overlook the systemic barriers that inhibit the effective delivery of tobacco cessation interventions.⁴ Several known solutions to these barriers are readily available. Notably, standardized approaches to identifying tobacco users and training primary care clinicians more effectively (in both cessation techniques and appropriate billing practices) can enhance service provision and health outcomes.²⁷ The best outcomes can be achieved by adopting a systems change approach,³⁰ ensuring that clinicians advise tobacco users to quit and follow through with documented, billable interventions supporting cessation efforts.

Limitations

Our findings are limited to a single health care system and may not generalize to other settings. Not all tobacco users were identified, and some users identified may have recently quit. The types of tobacco used were not available. Additionally, the analysis relied on EHR billing data, which may underreport counseling sessions provided but not documented or billed. Referrals to internal or external cessation services were not captured. Lastly, the model assumes that all contracts were FFS and that every ambulatory encounter with a tobacco user was billed.³⁴

CONCLUSIONS

The findings from this study demonstrate a discrepancy between the frequency of visits by tobacco users and the proportion of tobacco cessation services that were billed. Beyond opportunities to enhance billing, measurement of counseling provides a metric for the effectiveness of tobacco cessation interventions and can serve to improve both clinical and economic outcomes.

TAKEAWAY POINTS.

• An evidence-based billable tobacco cessation service can be provided at every office visit with a tobacco user.

• In a study of deidentified electronic health records at an academic medical center, billing insurance carriers for tobacco cessation services occurred in only 1% of eligible office encounters.

• Underbilling can reflect missed prevention opportunities for tobacco cessation counseling, poor billing practices, or a combination of both.

• The economic value of the missed counseling opportunities—calculated using insurance type and a fee-for-service payment model—amounts to more than $800,000 annually in potential lost revenue.