Cost-Effectiveness of Office-Based Buprenorphine Treatment for Opioid Use Disorder

Abstract

Aim:

To assess the effectiveness and cost-effectiveness of office-based buprenorphine treatment (OBBT) in the U.S.

Design, Setting and Participants:

We performed a model-based analysis of buprenorphine treatment provided in a primary care setting for the U.S. population with OUD.

Intervention:

Buprenorphine treatment provided in a primary care setting.

Measurements:

Fatal and nonfatal overdoses and deaths over five years, discounted lifetime quality-adjusted life years (QALYs), costs.

Findings:

For a cohort of 100,000 untreated individuals who enter OBBT, approximately 9,350 overdoses would be averted over five years; of these, approximately 900 would have been fatal. OBBT compared to no treatment would yield 1.07 incremental lifetime QALYs per person at an incremental cost of $17,000 per QALY gained when using a healthcare perspective. If OBBT is half as effective and twice as expensive as assumed in the base case, the incremental cost when using a healthcare perspective is $25,500 per QALY gained. Using a limited societal perspective that additionally includes patient costs and criminal justice costs, OBBT is cost-saving compared to no treatment even under pessimistic assumptions about efficacy and cost.

Conclusions:

Expansion of OBBT would be highly cost-effective compared to no treatment when considered from a healthcare perspective, and cost-saving when reduced criminal justice costs are included. Given the continuing opioid crisis in the U.S., expansion of this care option should be a high priority.

INTRODUCTION

The Covid-19 pandemic and the continued spread of fentanyl have dramatically worsened the national public health crisis of substance use in the U.S.¹ In 2021, more than 107,000 overdose deaths occurred, setting yet another grim record.² At least 70,000 of these deaths were recorded as opioid-related²; because of limitations in post-mortem data the true total is likely even higher.³

Effective treatments for opioid use disorder (OUD) are available, but most individuals in need do not receive them. In 2019, only about 28% of individuals with OUD received medication for opioid use disorder (MOUD), and an additional 15% received services without medication.⁴ Barriers to accessing MOUD are numerous and include lack of sustained funding for OUD treatment centers, long waiting times, geographic distance to the nearest treatment center, high out-of-pocket payments and low insurance reimbursement, stigma against providers and patients, and limits on the number of patients that a provider can treat.^5–7 In addition, the spread of Covid-19 has interfered with opioid treatment.⁸

The 2022 U.S. National Drug Control Strategy calls for expanding access to MOUD, including office-based buprenorphine treatment (OBBT).⁹ Unlike methadone maintenance, which can only be provided in specially licensed and heavily regulated opioid treatment programs, OBBT can be provided in generalist health care settings with less strict regulatory requirements. OBBT was dramatically expanded during a heroin epidemic in France, which was followed by sharp reductions in heroin-related morbidity.¹⁰ U.S. regulators have somewhat relaxed restrictions on the ability of physicians to prescribe buprenorphine for OUD¹¹, but have not attempted reform on the scale that was undertaken in France.

The U.S. Congress recently passed new legislation repealing the “X-waiver” that used to be required to treat OUD with buprenorphine; hence, evaluating the potential impact of a more significant expansion of OBBT is of major policy importance. This study therefore assesses the cost-effectiveness of significantly expanding OBBT in the U.S. A previous study using a healthcare sector perspective assessed the cost-effectiveness of long-term OBBT among clinically stable patients who had already completed six months of OBBT.¹² That study found OBBT to be cost-effective when measured over a 24-month time horizon, with a cost of $35,100 per quality-adjusted life year (QALY) gained ($44,200 in 2021 dollars). Here we assess a more expansive policy question, namely the cost-effectiveness of OBBT for anyone with OUD who wishes to seek such care, and we estimate lifetime costs and QALYs gained. We further expand on prior work by supplementing a healthcare sector perspective with a limited societal perspective that additionally includes patient costs and criminal justice costs, which are known to be significant.^13–17

METHODS

We defined OBBT as an intervention consisting of a buprenorphine/naloxone combination pharmacotherapy delivered in a primary care setting.¹⁸ OBBT typically includes some brief advice and counseling from a physician or nurse, but does not mandate extensive psychotherapy or counseling as is typically the case in specialist opioid treatment programs. Established patients diagnosed with OUD typically receive a 30-day supply of pills. Because buprenorphine is a partial opioid agonist with high affinity for the mu receptor in the brain, patients are typically advised to abstain from other opioids (e.g., heroin, oxycodone) for 24 hours before taking the first buprenorphine dose to avoid precipitated withdrawal. During the induction period the patient is seen in the primary care clinic weekly; once stable, the patient is seen monthly. Urinalysis is conducted weekly or monthly.

We estimated health outcomes and costs of OBBT using a previously developed continuous-time dynamic compartmental model that simulates the flow of individuals into and out of treatment.¹⁹ Individuals can transition between different health states including out of treatment, on treatment, abstinent (no illicit opioids) and not on treatment, and dead (Figure S1). We distinguish individuals who inject drugs from those who do not because of their different morbidity and mortality risks.

We modeled a representative cohort of 100,000 individuals with OUD in the U.S. population.¹⁹ We considered different combinations of sex and age (18 to 100 years old) and ran the model for each individual’s lifetime for each possible treatment option. Background mortality, baseline healthcare cost, and criminal justice cost varied by sex and age. To assess average per person outcomes we weighted outcomes according to the age and sex distribution of individuals with OUD in the U.S. We used model parameter values from a previous study of non-office-based MOUD¹⁹, adjusted to reflect OBBT (Table 1). We assumed probability distributions for all parameters (Table S1) and calculated mean values and 95% credible intervals for all outcomes. Further model details can be found elsewhere¹⁹ and in the Supplement.

Table 1.

Base Case Parameter Values and Sources

Parameter	Mean	Range	Source
Demographics
Fraction male	0.511	--	36, 37
Fraction female	0.489	--	36, 37
Average male age	43.7	--	36, 37
Average female age	45.3	--	36, 37
Initial fraction with OUD who inject drugs	0.253	[0.214 – 0.294]	37
Transitions
Death and Overdose, annual rates per person
Background mortality	CDC Life Tables	--	38
Non-overdose excess mortality due to OUD, out of treatment	0.00978	[0.00744 – 0.01249]	39
Non-overdose excess mortality due to OUD, in treatment	0.00318	[0.00238 – 0.00406]	39
Overdose, out of treatment	0.1030	[0.0472 – 0.2206]
Overdose, in treatment	0.0438	[0.0205 – 0.0932]	40
Overdose Survival Probability, per overdose 1	0.899	[0.799 – 0.954]	41
Treatment Discontinuation, annual rates per person	1.609	[1.002 – 2.420]	42–48
Other Transitions, annual rates per person
Reentry into treatment (from out of treatment > 1 month)	0.426	[0.367 – 0.489]	49
Becoming abstinent and leaving treatment	0.316	[0.296 – 0.337]	49
Becoming abstinent, from out of treatment	0.0791	[0.00401 – 0.1551]	Estimated
Return to use from abstinence < 1 year2	0.379	[0.331 – 0.430]	49
Return to use from abstinence, year 10+	0.019	[0.00394 – 0.0342]	49
Rate of initiating injection drug use	0.031	[0.020 – 0.043]	50
Costs, 2021 USD 3
Annual background healthcare costs
Baseline, male age 304	2,334	--	51, 52
Excess cost for OUD out of treatment	7,457	[6,729– 8,221]	53
Excess cost for OUD in treatment5	5,973	[3,107– 8,998]	54
Annual criminal justice costs, patient age 306
Out of treatment	40,487	[38,191– 45,659]	14
Out of treatment (IDU)	61,889	[57,792– 66,004]	14
On treatment	9,135	[8,195 – 10,080]	14
On treatment (IDU)	20,436	[18,708– 22,148]	14
Abstinent	5,305	[4,818– 5,796]	14
Abstinent (IDU)	8,107	[7,567 – 8,648]	14
Annual OBBT costs
Cost of OBBT provision	7,212	[6,508 – 7,951]	12
Patient costs	594	[536 – 654]	12
Healthcare cost per overdose	2,681	[1,161– 5,308]	41
Quality-of-life Multipliers for Health States
Out of treatment, month 1	0.670	[0.660 – 0.680]	49, 55
Out of treatment (IDU), month 1	0.660	[0.640 – 0.680]	49, 55
Out of treatment, month > 1	0.670	[0.660 – 0.680]	49, 55
Out of treatment (IDU), month > 1	0.660	[0.640 – 0.680]	49, 55
Induction into treatment	0.725	[0.700 – 0.750]	49, 55
Induction into treatment (IDU)	0.710	[0.700 – 0.720]	49, 55
On treatment	0.725	[0.700 – 0.750]	49, 55
On treatment (IDU)	0.710	[0.700 – 0.720]	49, 55
Abstinence: first year7	0.725	[0.700 – 0.750]	49, 56
Abstinence (IDU): first year	0.710	[0.700 – 0.720]	49, 56
Abstinence: year 10+	0.984	[0.970 – 0.996]	Calculated
Abstinence (IDU): year 10+	0.983	[0.969 – 0.996]	Calculated

¹For calculations, see Fairley et al.^{19. Fairley M, Humphreys K, Joyce VR, et al. Cost-effectiveness of treatments for opioid use disorder. JAMA Psychiatry. 2021;78(7):776–77.}

²Rates of recurrence from abstinence from years 2, …, 9 of abstinence were linear interpolations of the year 1 and year 10 values

³All costs were updated to 2021 dollars using the Consumer Price Index

⁴Baseline healthcare costs were age- and sex-specific.

⁵Estimated based on the (conservative) assumption that patients with OUD who are on treatment incur 20% higher healthcare costs on average than those not on treatment due to increased access to healthcare

⁶Criminal justice costs for other ages were calculated based on Krebs et al.^{49. Krebs E, Enn B, Evans E, et al. Cost-effectiveness of publicly funded treatment of opioid use disorder in California Ann Intern Med. Jan 2 2018;168(1):10–19.}

⁷Utility values for years 2, …, 9 were linear interpolations of the year 1 and year 10+ utility values

Abbreviations: IDU = injection drug user; OBBT = office-based buprenorphine treatment; OUD = opioid use disorder

A number of studies have examined outcomes among individuals receiving OBBT in different settings. For example, a randomized controlled trial of 53 clinically stable patients who had already received six months of OBBT found that 51% were retained for one year and 38% were retained for two years.²⁰ A retrospective study of 255 patients at a primary care practice in Baltimore found that 56.9% of patients remained in treatment after one year, with 64.7% of months being opioid-negative on uranalysis.²¹ A retrospective study examined 382 patients receiving OBBT at an urban academic primary care practice, where care was provided collaboratively by nurse care managers and generalist physicians, and found that 49% remained in treatment after 12 months and an additional 2% had been tapered off after six months of adherence and absence of illicit drug use; of those remaining in treatment, 91.1% were opioid-negative on uranalysis.²² A retrospective cohort study of 957 patients receiving OBBT in a primary care clinic for homeless individuals found a retention rate of 11.3% and an abstinence rate of 2.9% at 12 months.²³ An observational cohort study of 41 patients in Rhode Island, 16 of whom had transferred from methadone to OBBT, found that 59% remained in treatment after six months.²⁴

These and other studies have examined patients in different settings (e.g., newly starting on MOUD vs. not; clinically stable vs. not; homeless vs. broader population with OUD) with different OBBT program characteristics (e.g., treatment provided exclusively by the primary care physician vs. treatment provision by nurses and the primary care physician), and have found a wide range of values for OBBT effectiveness. None has compared OBBT to provision of buprenorphine treatment or other forms of MOUD in a specialty clinic. A study comparing office-based methadone treatment for OUD to clinic-based methadone treatment found that methadone treatment in the primary care setting was slightly more effective (90.5% retention in treatment after 6 months vs. 82.6%, and 95.3% of samples opioid-free on urinalysis vs. 85.7%).²⁵ We therefore considered a range of values for OBBT effectiveness. Specifically, we varied the rate at which patients are retained on treatment and the rate at which patients reenter treatment. In the base case we assumed that these values would be the same as for clinic-based buprenorphine treatment.¹⁹ We also considered values that were 50% lower and 50% higher than base case values.¹⁹ Thus, for example, 50% lower effectiveness corresponded to a 50% higher patient discontinuation rate and a 50% lower rate of reentry into treatment compared to base case values.

We calculated the mean number of fatal and nonfatal overdoses and total deaths over five years for the cohort of individuals with OUD. We used a life-long time horizon for our estimates of cost-effectiveness in accordance with guidelines for the conduct of economic analyses.²⁶ For each health state we assigned a quality-of-life multiplier based on published estimates (Table 1). For individuals who are out of treatment and abstinent, we assumed that their quality of life increases over time, eventually approaching that of individuals without OUD.

We included healthcare, patient, and criminal justice costs. All costs are expressed in 2021 dollars, converted using the Consumer Price Index. Healthcare costs included age-, sex-, and OUD-state-specific costs of healthcare, nonfatal and fatal overdoses, and treatment. For individuals who become abstinent and leave treatment, we assumed that their healthcare costs diminish over time, eventually approaching that of individuals without OUD.

We assumed that OBBT cost includes drug-specific ongoing costs associated with buprenorphine and its administration. A study of OBBT among 53 clinically stable patients estimated monthly cost of OBBT provision to be $601 ($498 in 2010 dollars).¹² A study of buprenorphine treatment provided in specialty clinics estimated monthly cost to be $551 ($530 in 2019 dollars).¹⁹ Costs will vary in different healthcare systems and settings but these studies suggest that OBBT provision is not necessarily less expensive than clinic-based buprenorphine treatment provision. A study comparing office-based methadone treatment for OUD to clinic-based methadone treatment found that methadone treatment in the primary care setting was approximately 50% more expensive than in the specialty clinic setting, primarily because of increased physician and other clinician time.²⁵ Thus, we assumed a base value of $601 per month for OBBT, and as a conservative estimate increased this value to 50% higher in sensitivity analysis. In our limited societal analysis we also included patient costs for receiving OBBT, which were estimated in a previous study to be $49.50 per month ($41 in 2010 dollars, comprising patient time, travel time, and transportation cost).¹²

Multiple studies have found reduced criminal justice costs for individuals on MOUD.^13–17 Accordingly, we estimated criminal justice costs for individuals with OUD receiving OBBT and those not on treatment. Costs were specific to the patient’s age and health state.¹⁴

We calculated incremental cost-effectiveness ratios (ICERs) for OBBT compared to no treatment, as we assumed that individuals who begin OBBT are not currently receiving treatment. We used a healthcare sector perspective for costs and QALYs as well as a limited societal perspective that additionally included patient costs and criminal justice costs, and discounted all values to the present at 3% annually.²⁶ An Impact Inventory Analysis and CHEERS checklist are provided in the Supplement (Tables S2, S3).

Model validation was performed by comparing a range of outcomes for 30-year-old males (reflecting the average age in many published studies) over 50 years, including survival curves, probability of fatal overdose, survival-adjusted number of nonfatal overdoses, percentage of time spent in treatment, and healthcare cost to external estimates. Full details are provided elsewhere.¹⁹

RESULTS

Base Case Analysis

In the absence of treatment, 42,656 overdoses (4,133 fatal, 38,524 nonfatal) and 12,661 deaths occurred in the cohort of 100,000 patients over five years, and 11.58 discounted lifetime QALYs were experienced per person (Table 2). Under base case assumptions about OBBT, 33,311 overdoses (3,227 fatal, 30,084 nonfatal) and 10,894 deaths occurred in the cohort over five years, with 12.65 discounted lifetime QALYs experienced per person. Thus, OBBT averted 9,345 overdoses over five years, 906 of which would have been fatal, and increased life expectancy by 1.07 net present QALYs per person.

Table 2.

Results: Health Outcomes and Costs (Mean Value and 95% Credible Interval)

	Over 5 Years, per 100,000 Individuals with Opioid Use Disorder				Lifetime, per Person Discounted
Treatment Option	Fatal Overdoses	Nonfatal Overdoses	Total Overdoses	Deaths1	QALYs	Costs ($1000s, 2021)				Healthcare Perspective ICER ($/QALY gained)2
						Health Care	Patient Costs	Criminal Justice	Total
No treatment	4,133 [3,175, 5,270]	38,524 [15,598, 87,141]	42,656 [19,363, 91,596]	12,661 [10,941, 14,657]	11.58 [10.06, 12.93]	266 [248, 285]	0 [0, 0]	371 [290, 470]	637 [550, 742]	--
OBBT
Base cost, same efficacy as CBBT	3,227 [2,508, 4,074]	30,084 [12,221, 67,991]	33,311 [15,182, 71,485]	10,894 [9,590, 12,399]	12.65 [11.55, 13.76]	284 [263, 308]	1.4 [1, 1.9]	286 [227, 335]	572 [506, 647]	$16,822
Base cost, 50% lower efficacy than CBBT3	3,613 [2,802, 4,566]	33,677 [13,684, 76,183]	37,289 [16,988, 80,076]	11,647 [10,185, 13,328]	12.09 [10.81, 13.26]	275 [256, 296]	0.7 [0.5, 1.0]	329 [260, 411]	605 [530, 693]	$17,647
50% higher cost than base cost4, same efficacy as CBBT	3,227 [2,508, 4,074]	30,084 [12,221, 67,991]	33,311 [15,182, 71,485]	10,894 [9,590, 12,399]	12.65 [11.55, 13.67]	292 [269, 317]	1.4 [1, 1.9]	286 [228, 355]	580 [514, 655]	$24,299
50% higher cost than base cost, 50% lower efficacy than CBBT	3,613 [2,802, 4,566]	33,677 [13,684, 76,183]	37,289 [16,988, 80,076]	11,647 [10,185, 13,328]	12.09 [10.81, 13.26]	279 [259, 300]	0.7 [0.5, 1.0]	329 [260, 411]	609 [533, 697]	$25,490

¹From all causes

²Incremental cost-effectiveness ratio (ICER) compared to No treatment

³Efficacy was varied by simultaneously changing the rate of treatment discontinuation and the rate of treatment reentry

⁴The health care cost (cost of treatment provision) was varied

Abbreviations: CBBT = specialty clinic-based buprenorphine treatment, OBBT = office-based buprenorphine treatment, QALY = quality-adjusted life year

With no treatment, total discounted lifetime healthcare costs were $266,200 per person. With OBBT provision, these costs were $284,400 per person. Thus, using a healthcare perspective, OBBT cost $17,000 per QALY gained compared to no treatment (Figure 1a).

Figure 1.

Lifetime per person costs and QALYs for no treatment and for office-based buprenorphine treatment under different assumptions about cost and effectiveness. Efficacy and cost were assumed to be 50% higher or lower than base case values.

With no treatment, discounted lifetime criminal justice costs were $371,000 per person. No patient cost was incurred as no treatment was received. Thus, using a limited societal perspective, $637,200 in discounted lifetime costs (i.e., $266,200 + $371,000) were incurred per person. For OBBT, discounted per person patient cost was $1,400 and criminal justice cost was $286,000, leading to a total cost of $571,800 (i.e., $284,400 + $1,400 + $286,000) per person. Using the limited societal perspective, OBBT was cost-saving compared to no treatment (Figure 1b) because of the significant savings in criminal justice cost among individuals receiving OBBT.

Sensitivity Analysis

If OBBT costs the same as clinic-based buprenorphine treatment but is half as effective in retaining people on treatment and inducing patients to re-enter treatment, 12.09 QALYs are experienced per person, a gain of 0.51 QALYs compared to no treatment (Table 2). Healthcare cost was $275,300 per person, and total cost was $605,100 per person. In this case, incremental healthcare cost was $9,200 per person, so OBBT cost $18,000 per QALY gained using the healthcare perspective. The total cost of $605,100 per person was less than the total cost of $637,100 per person for the case of no treatment so OBBT was cost-saving compared to no treatment when using the limited societal perspective.

If OBBT has effectiveness similar to that of clinic-based buprenorphine treatment but costs 50% more, the same number of QALYs (12.65) are experienced as in the base case, but healthcare cost increases to $292,000 per person and total cost increases to $579,500 per person (Table 2). Under these assumptions, OBBT compared to no treatment costs $24,200 per QALY gained when using the healthcare perspective, and is cost-saving when using the limited societal perspective.

Finally, if OBBT is half as effective and 50% more expensive than clinic-based buprenorphine treatment, 12.09 QALYs are experienced per person. Lifetime per person healthcare cost is $279,100 and total cost is $608,900 (Table 2). Under these assumptions, OBBT compared to no treatment costs $25,500 per QALY gained when using the healthcare perspective, and is cost-saving when using the limited societal perspective.

DISCUSSION

Our analysis shows that expanding OBBT to untreated individuals with OUD leads to a reduction in overdoses and deaths, and a substantial gain in lifetime QALYs per person. We estimate that for a cohort of 100,000 untreated individuals who enter OBBT, approximately 9,350 overdoses would be averted over five years; of these, approximately 900 would have been fatal. Using a healthcare perspective, OBBT cost less than $26,000 per QALY gained, which is less than the US GDP per capita, and thus would be considered highly cost-effective.²⁷ When we additionally consider savings in criminal justice costs among treated individuals, such expansion is cost-saving, even under pessimistic assumptions about OBBT cost and effectiveness.

Effective expansion of OBBT will require significant long-term efforts at the federal and state levels to reduce barriers to such treatment. These include the development of sustainable reimbursement models, expansion of Medicaid in states that have not done so, tailored strategies including education and training, and availability of resources to support primary care physicians who provide OBBT, and efforts aimed at reducing stigma among individuals seeking treatment for OUD.^28–30 The challenges of expanding OBBT will likely vary by healthcare system and setting. While not every individual with OUD will be willing to receive OBBT, other cost-effective service options are available (e.g., contingency management and overdose education and naloxone distribution)¹⁹ that can be expanded alongside OBBT to better care for the population of individuals with OUD.

Our analysis adds to the existing evidence in two important aspects. First, in assessing cost-effectiveness of OBBT, we used a life-long time horizon as recommended by the Second Panel on Cost Effectiveness in Health and Medicine.²⁶ Although our estimates are consistent with prior work that used a 24-month time horizon, use of shorter time horizons can underestimate or overestimate cost-effectiveness. Second, as also recommended by the Panel, we use a broader perspective than the healthcare sector. For substance use treatment, this is particularly relevant as successful treatment reduces costs in the criminal justice sector.^13–17 Including these cost savings has a substantial impact on our assessment of cost-effectiveness, with treatment reducing total costs. These savings are important from a societal perspective, although the health systems providing care will not receive the savings. Nonetheless, from a healthcare sector only perspective, OBBT provides high value.

Our analysis has several limitations. Data on OBBT cost and effectiveness is limited, and cost and effectiveness may differ in different settings. However, we made conservative estimates for baseline cost and performed an analysis with 50% increase in cost, showing that OBBT expansion is cost-effective even under pessimistic assumptions about OBBT cost and effectiveness. Additionally, we assumed that the costs of expanding access to OBBT are linear; that is, the same cost is incurred for each patient who receives OBBT. If there are economies of scale, OBBT will be more cost-effective than we have estimated.

Our analysis also highlights high-priority areas for future research. Additional empiric evaluations of health and economic outcomes associated with OBBT, including analyses in diverse healthcare systems and conducted in diverse populations, would substantially improve the evidence for decision making around OBBT. Information about retention and return to use rates would be particularly useful. Studies on approaches for incentivizing provision of OBBT in primary care would helpful, as implementation may be challenging even if regulatory barriers have been removed. Finally, studies of alternative strategies for provision of OBBT would be useful in establishing best practices in terms of quality and efficiency.

One of the many tragedies of the U.S. opioid crisis is that the majority of the millions of individuals who experience OUD^{31, 32} do not receive evidence-based, life-saving treatment. Recent regulatory reforms make it somewhat easier for physicians to prescribe buprenorphine for OUD¹¹, but more dramatic steps will be needed to significantly reduce the damage of the opioid crisis.³³ Our analysis shows that expanding provision of OBBT, such as was done in France during that nation’s epidemic, would be cost-effective from a healthcare sector perspective, and cost-saving from a broader societal perspective. Importantly, such expansion could prevent significant morbidity and mortality by improving access to MOUD for many individuals who currently do not have access.^{34, 35}