Medicaid Spending on Drugs Marketed Without US Food and Drug Administration Approval in 2020

Abstract

This cross-sectional study examines the proportion of Medicaid prescription and spending on drugs marketed in the United States without Food and Drug Administration approval.

Although the US Food and Drug Administration (FDA) reviews the safety and effectiveness of new drugs before they are marketed, some drugs marketed before 1938 continue to be sold without formal regulatory approval. In 2006, the FDA launched the Unapproved Drugs Initiative (UDI) to provide incentives for manufacturers of unapproved drug products to voluntarily submit safety and effectiveness data in return for a period of market exclusivity.¹ The UDI has been controversial because, in some cases, it was disruptive to competition, leading to higher prices and drug shortages.^2,3,4 Citing these concerns, the Department of Health and Human Services (HHS) announced an end to the program in November 2020. However, in May 2021, the FDA withdrew the HHS notice and reinstated the program.⁵ We studied use and spending on drugs marketed in the US without FDA approval among Medicaid patients in 2020.

Methods

In this cross-sectional study, we identified unapproved drugs based on manufacturer-reported data to the Medicaid Drug Rebate Program and used National Drug Codes to link this information to 2020 Medicaid State Drug Utilization Data (eMethods in the Supplement). For each unapproved drug, we searched for similar FDA-approved drugs and over-the-counter products with the same active ingredient, formulation, and dose (eFigure in the Supplement).

We performed descriptive analyses and compared the mean spending per unit of unapproved drug products and corresponding similar approved versions with 2-tailed Wilcoxon signed rank tests at a significance of P < .05 using R version 3.6 (R Project for Statistical Computing). This study was not submitted for institutional review board review because it used public, nonidentifiable data that do not constitute human subject research according to 45 CFR 46.102. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Results

Of the 31 293 National Drug Codes with any reported Medicaid use in 2020, 327 (1.0%) were unapproved drugs, representing 134 products with a distinct active ingredient, formulation, and dose. Most unapproved drugs (87 [64.9%]) were marketed by a single manufacturer; 117 (87.3%) were distributed by 3 or fewer manufacturers (Table 1).

Table 1. Characteristics of Unapproved Drugs in 2020.

Characteristic	Unique products, No. (%) (N = 134)
Formulation
Oral	48 (35.8)
Injected	43 (32.1)
Topical	36 (26.9)
Othera	7 (5.2)
Marketing status
Approval withdrawn following FDA drug efficacy study implementation review	6 (4.5)
Unapproved drugs	128 (95.5)
No. of manufacturers
1	87 (64.9)
2	22 (16.4)
3	8 (6.0)
4	9 (6.7)
≥5 or more	8 (6.0)
Medicaid spending, median (IQR), $	114 716 (21 502-416 316)
Prescriptions, median (IQR), No.	1040 (198-4631)
Similar products availableb
FDA approvedc	21 (15.7)
Most spending (>75%) on approved version	9 (42.9)
Similar spending (25-75%) between versions	5 (23.8)
Most spending (>75%) on unapproved version	7 (33.3)
Over-the-counterd	10 (7.5)

Abbreviation: FDA, US Food and Drug Administration.

^a Includes ophthalmic, inhaled, and rectal routes of administration.

^b Defined as products having the same active ingredient, route of administration, and dose.

^c FDA approved versions were identified from the Medicaid data and stratified based on whether a majority of Medicaid spending in 2020 was on the approved or unapproved version. The 21 products with approved versions were sodium chloride intravenous solution; atropine sulfate syringe; atropine sulfate vial; calcium chloride syringe; chlordiazepoxide/clidinium bromide capsule; citric acid/sodium citrate solution; cysteine vial; dextrose in water syringe; heparin intravenous solution; hydromorphone vial; potassium bicarbonate/citric acid effervescent tablet; potassium citrate/citric acid solution; potassium phosphate, monobasic-dibasic vial; sodium bicarbonate vial (2 different doses); sodium chloride vial, sodium phosphate, monobasic-dibasic vial; sodium/potassium/potassium citrate/sodium citrate/citric acid solution; tetracaine ophthalmic drops; water for injection, sterile vial; and zinc sulfate vial.

^d Over-the-counter versions were identified by searching GoodRx.com and a major retail pharmacy website, allowing for minor but non–clinically significant deviations in dose between the unapproved prescription and over-the-counter versions (eg, phenazopyridine 95-mg vs 100-mg tablets). The 10 products with over-the-counter versions were phenazopyridine tablets (2 different doses), ammonia inhaled ampules, prenatal vitamins, selenium sulfide shampoo (2 different doses), and urea cream (3 different doses) or lotion.

Overall, 21 unapproved drugs (15.7%) had similar products approved by the FDA; in 9 cases, more than 75% of Medicaid spending was on the approved version, and in 7 cases, more than 75% of Medicaid spending was on the unapproved version. There was no difference in price between the 21 pairs of approved and unapproved versions of similar products. Ten unapproved products (7.5%) had similar over-the-counter versions.

Unapproved drugs had 1.4 million prescriptions in Medicaid in 2020 (median [IQR] 1040 [198-4631]), representing 0.2% of the 678 million Medicaid prescriptions. Medicaid spending for these drugs totaled $62.5 million (median [IQR], $114 716 [$21 502-$416 316]), or 0.2% of $29.6 billion net Medicaid prescription drug spending in fiscal year 2019.⁶ Three unapproved drugs (ie, phenobarbital, porcine thyroid hormone, and phenazopyridine) accounted for 964 600 of 1.4 million (68.9%) prescriptions and $29.3 million of $62.5 million (46.8%) spending (Table 2).

Table 2. Unapproved Drugs With Highest Medicaid Spending in 2020.

Drug name and formulationa	Manufacturers, No.	Medicaid prescriptions in 2020, No. in thousands (N = 1444 thousand)	Medicaid spending in 2020, in millions of US $ (%)b (N = $62.5 million)
1. Phenobarbital
Overall	10	265.0 (18.3)	11.2 (18.0)
Oral tablet	8	185.9 (12.9)	5.7 (9.1)
Oral elixir	5	75.3 (5.2)	4.8 (7.7)
Injected vial	2	3.9 (0.3)	0.7 (1.1)
2. Pork thyroid, oral tablet	2	254.2 (17.6)	10.1 (16.2)
3. Phenazopyridine, oral tablet	7	476.7 (33.0)	7.9 (12.6)
4. Hyoscyamine
Overall	12	181.8 (12.6)	4.0 (6.4)
Oral tablet, sublingual	7	82.9 (5.7)	1.7 (2.8)
Oral tablet	7	75.5 (5.2)	1.7 (2.7)
Oral tablet, rapid dissolve	7	17.0 (1.2)	0.4 (0.7)
Oral elixir	3	3.9 (0.3)	0.1 (0.1)
Oral drops	3	1.3 (0.1)	0.1 (0.1)
Oral tablet, extended release	1	1.0 (0.1)	<0.1 (0.1)
Injected, ampule	1	0.2 (<0.1)	<0.1 (<0.1)
5. Atropine sulfatec
Overall	4	25.7 (1.8)	2.9 (4.6)
Injected, vial	2	23.9 (1.7)	2.7 (4.3)
Injected, syringe	1	1.8 (0.1)	0.2 (0.3)
Topical ointment	1	0.1 (<0.1)	<0.1 (<0.1)
6. Lidocaine, topical cream	3	43.9 (3.0)	2.6 (4.1)
7. Potassium phosphate, injected vial	2	12.6 (0.9)	2.5 (4.1)
8. Ethyl alcohol, injected viald	2	10.0 (0.7)	2.4 (3.8)
9. Sodium phosphate, injected vial	1	10.7 (0.7)	2.1 (3.3)
10. Sulfacetamide sulfur
Overall	5	13.3 (0.9)	1.6 (2.6)
Topical cleanser	5	11.3 (0.8)	1.2 (2.0)
Topical suspension	1	1.8 (0.1)	0.4 (0.6)
Topical cream	1	0.1 (<0.1)	<0.1 (<0.1)
Topical medication pad	1	<0.1 (<0.1)	<0.1 (<0.1)
11. Zinc, copper, manganese, chromium and selenium, injected vial	1	9.9 (0.7)	1.5 (2.4)
12. Zinc sulfate, injected vialc	1	6.0 (0.4)	1.5 (2.4)
13. Selenium sulfide, topical shampoo	5	13.4 (0.9)	1.4 (2.2)
14. Hydrocortisone, lidocaine, aloe vera
Overall	2	2.3 (0.2)	1.2 (1.9)
Rectal gel with applicator	1	1.1 (0.1)	0.8 (1.2)
Rectal kit	2	1.2 (0.1)	0.4 (0.7)
15. Urea
Overall	2	10.9 (0.8)	1.0 (1.6)
Topical cream	2	9.1 (0.6)	0.8 (1.3)
Topical lotion	1	1.8 (0.1)	0.2 (0.3)
16. Lidocaine, hydrocortisone
Overall	2	4.1 (0.3)	0.8 (1.3)
Rectal kit	2	1.4 (0.1)	0.7 (1.1)
Rectal cream	2	2.1 (0.1)	0.1 (0.2)
Rectal cream with applicator	1	0.6 (<0.1)	0.1 (0.1)
17. Opium, oral tincture	2	1.8 (0.1)	0.7 (1.1)
18. Zinc, copper, manganese, and chromium, injected vial	1	3.9 (0.3)	0.7 (1.1)
19. Epinephrine
Overall	2	5.3 (0.4)	0.6 (1.0)
Injected, syringe	2	5.2 (0.4)	0.6 (1.0)
Injected, ampule	1	0.1 (<0.1)	<0.1 (<0.1)
20. Salicylic acid
Overall	2	3.6 (0.2)	0.6 (1.0)
Topical liquid film	1	2.5 (0.2)	0.3 (0.5)
Topical gel	1	0.8 (0.1)	0.2 (0.3)
Topical shampoo	1	0.2 (<0.1)	0.1 (0.1)
Topical foam	1	0.1 (<0.1)	<0.1 (<0.1)

^a Because some manufacturers made more than 1 formulation of a drug, the overall number of unique manufacturers may be fewer than the sum of each individual formulation.

^b Spending is the total amount reimbursed in 2020, not accounting for rebates, which for these noninnovator products would have been 13% of average manufacturer price.

^c At least 1 version of this product has been approved by the Food and Drug Administration.

^d On March 26, 2020, the Food and Drug Administration requested that manufacturers cease sales of unapproved versions of dehydrated alcohol products under the Unapproved Drugs Initiative.

Discussion

In 2020, unapproved drugs accounted for only 0.2% of Medicaid prescriptions and 0.2% of estimated net prescription drug spending in the program. By comparison, in 2006, when the FDA launched the UDI, thousands of unapproved drugs were marketed, representing approximately 2% of US prescriptions.² Our analysis was limited because it was restricted to Medicaid, which represents a small proportion of the retail prescription drug market; Medicaid data include only outpatient use of drugs, which may exclude or underestimate spending for drugs primarily used in hospitalized patients.

Our findings suggest that offering a period of market exclusivity under the UDI is no longer an effective means to encourage manufacturers to voluntarily submit safety and effectiveness data. The reason is that most remaining unapproved drugs are made by one or few manufacturers and already have minimal market competition. A better approach would be for the FDA to adopt a more targeted regulatory framework, proactively determining which remaining unapproved drugs lack existing evidence of safety and effectiveness and coordinating with the National Institutes of Health to fund any necessary new research.

Supplement.

eMethods. Supplemental Methods

eFigure. Unapproved Drug Products Marketed in 2020