Physicians’ Perspectives on FDA Approval Standards and Off-label Drug Marketing

Abstract

This nationwide survey of US primary care internists and medical specialists evaluates physicians’ attitudes about the US Food and Drug Administration’s drug approval standards and off-label drug marketing.

Recently, 2 fundamental aspects of the US Food and Drug Administration’s (FDA’s) pharmaceutical market oversight have become controversial. First, the FDA created and increasingly uses expedited development and approval pathways, emphasizing greater reliance on postmarketing drug data.¹ Some argue that approval standards are becoming inappropriately low and postapproval evaluation too lax.²

Second, a court decision and state and federal legislation have chipped away at rules blocking advertising of drugs for non–FDA-approved indications (“off-label promotion”), which leads to patient harm by permitting manufacturers to encourage use of inadequately vetted medications. We surveyed a national sample of primary care internists and medical specialists to characterize physicians’ attitudes on these topics.

Methods

We contacted a random sample of 500 clinically active internists, 500 endocrinologists, and 500 cardiologists, all from the American Board of Internal Medicine Diplomate list. From May to December 2017, potential responders were provided online and paper opportunities to respond ($50 honorarium); 84 of these had out-of-date contact information. The project was approved by the Committee for the Protection of Human Subjects at Dartmouth College, with informed consent implied from completing and returning the survey.

We created and pretested a novel 25-question survey assessing perspectives about off-label drug promotion, FDA approval standards, and connections between drug evaluation and pricing. The questions were part of a broader survey about communicating evidence supporting prescription drug use.³ All respondents provided demographic information, primary practice area, and years since medical school or residency graduation.

Results

Among 686 respondents (48% response rate), 385 (59%) were men, 368 (56%) white, 223 (34%) Asian, 18 (3%) African American, and 28 (4%) Hispanic. The mean (SD) age of respondents was 44 (8) years, and they were 18 (8) years removed from medical school.

Most viewed the FDA’s drug approval regulations favorably, with 80% (n = 540) agreeing that its approval process helps “protect the public from ineffective or dangerous drugs.” A majority of respondents (78%; n = 522) preferred that drugs be tested in 2 prospective randomized trials, with the remainder (21%; n = 143) recommending only 1 such trial (very few [1%, n = 8] favored uncontrolled studies and patient experience) (Table 1). At the same time, the majority (65%; n = 438) were satisfied with the FDA’s bar for approving new drugs; only a minority characterized the current FDA standards as “too high” (24%; n = 165) or “much too high” (3%; n = 20).

Table 1. Attitudes of Physicians^a About Preapproval Testing of Investigational Drugs.

Statement of Attitude	No. (%; 95% CI) (N = 686)
Drugs should be approved if efficacy is demonstrated
In at least 2 randomized trials to ensure replication	522 (78; 74-81)
In a single randomized trial	143 (21; 18-24)
In uncontrolled studies and patient experience	8 (1; 0-2)
Before a new drug goes on the market, the company should have to
Prove the drug is safe	26 (4; 2-5)
Prove the drug is effective	10 (1; 1-2)
Both	636 (94; 93-96)
Neither	2 (0.3; 0-1)
I think drugs should be approved if they are shown to improve
Patient outcomes	256 (38; 34-42)
Surrogate measures alone	26 (4; 2-5)
Both patient outcomes and surrogate measures	390 (58; 54-62)
Some have proposed that drug companies should be allowed to market their drugs without first proving that they work in prospective clinical trials. What do you think of this proposal?
Great idea	7 (1; 0-2)
Good idea	24 (4; 2-5)
Bad idea	249 (37; 33-41)
Terrible idea	394 (58; 55-62)
Mostly, FDA regulations
Keep too many lifesaving drugs from reaching the market quickly	132 (20; 17-23)
Protect the public from ineffective or dangerous drugsb	540 (80; 77-83)
In general, do you think the FDA’s bar for approving new drugs in recent years isc
Much too high	20 (3; 2-4)
Too high	165 (24; 21-28)
About right	438 (65; 61-69)
Too low	49 (7; 5-9)
Much too low	2 (0.3; 0-1)
When the FDA approves a new drug, generally, I trust that the benefits outweigh the risks for my patients with the approved indications
Yes	605 (91; 88-93)
No	62 (9; 7-11)
When the FDA approves a new drug, I often review the data on which approval was based myself to confirm the drug’s appropriateness for my patientsd
Yes	466 (70; 66-73)
No	202 (30; 27-34)
Some approved drugs have only been shown to improve surrogate measures alone (eg, lower hemoglobin A1c), while other approved drugs have been shown to improve patient outcomes (eg, reduce the risk of cardiovascular events). All else being equal, prices should be
Higher for drugs improving surrogate measures	19 (3; 2-4)
Higher for drugs improving patient outcomes	183 (27; 24-31)
The same	467 (70; 66-73)

Abbreviations: ED, emergency department; FDA, Food and Drug Administration; ICU, intensive care unit.

^aPhysicians must have previously reported being clinically active (ie, ≥40% of time and ≥21 h/mo in patient care) in nonacute settings (ie, ≤50% of time in the ICU, ED, or cardiac catheterization laboratory).

^bWomen were more likely than men to think that FDA rules protected the public (85%, n = 271 vs 78%, n = 383; P = .02).

^c35% of cardiology specialists (n = 182) thought that current FDA standards were higher than they should be, compared with 21% of internists (n = 193) and 27% of endocrinologists (n = 233) (P = .03).

^dTrust was more common among younger physicians (mean [SD] age, 44.0 [8.1] years) than among older ones (47.6 [8.7] years) (P = .002).

Most (60%; n = 407) thought that the FDA should “definitely not” or “probably not” allow off-label promotion to physicians and that it would be a “bad idea” or “terrible idea” to allow such promotion by sales representatives in physicians’ offices (71%; n = 479) or in medical journals (68%; n = 458) (Table 2). Many predicted that off-label promotion would increase prescriptions for drugs without meaningful benefits (61%; n = 416) and for diseases not previously considered medical problems (84%; n = 568). Physicians were much more likely to believe that off-label promotion would worsen (rather than improve) clinical decisions (42%; n = 285 vs 30%; n = 198).

Table 2. Attitudes of Physicians^a About Off-label Promotion.

Statement of Attitude	No. (%; 95% CI) (N = 686)
The FDA should allow drug companies to promote off-label uses of drugs to
Physicians
Definitely no	157 (23; 20-26)
Probably no	250 (37; 33-40)
Probably yes	232 (34; 31-38)
Definitely yes	42 (6; 4-8)
Patients
Definitely no	512 (75; 72-78)
Probably no	122 (18; 15-21)
Probably yes	42 (6; 4-8)
Definitely yes	5 (1; 0-1)
The FDA allowing off-label promotion by
Sales representatives in physicians’ offices
Terrible idea	161 (24; 21-27)
Bad idea	318 (47; 43-51)
Good idea	176 (26; 23-29)
Great idea	24 (4; 2-5)
Drug ads in medical journals
Terrible idea	183 (27; 24-30)
Bad idea	275 (41; 37-44)
Good idea	204 (30; 27-34)
Great idea	17 (3; 1-4)
Direct-to-consumer ads
Terrible idea	489 (72; 69-75)
Bad idea	154 (23; 20-26)
Good idea	33 (5; 3-6)
Great idea	3 (<1; 0-1)
If the FDA changes its rules and allows off-label promotion of drugs to physicians, what will happen to prescriptions for
Drugs with meaningful benefits for patients
Decrease	42 (6; 4-8)
No change	299 (44; 40-48)
Increase	335 (50; 46-53)
Drugs with no meaningful benefits for patients
Decrease	63 (9; 7-11)
No change	198 (29; 26-33)
Increase	416 (61; 58-65)
Safer drugs
Decrease	86 (13; 10-15)
No change	375 (56; 52-59)
Increase	214 (32; 28-35)
More dangerous drugs
Decrease	123 (18; 15-21)
No change	228 (34; 30-37)
Increase	325 (48; 44-52)
Drugs for serious conditions with limited treatment options
Decrease	25 (4; 2-5)
No change	146 (22; 18-25)
Increase	505 (75; 71-78)
Drugs for conditions not conventionally considered medical problems
Decrease	13 (2; 1-3)
No change	94 (14; 11-16)
Increase	568 (84; 81-87)
How will drug company off-label promotion affect
Patient health
Much worse	27 (4; 3-5)
Worse	230 (34; 30-38)
No change	249 (37; 33-41)
Better	163 (24; 21-27)
Much better	6 (1; 0-2)
Physicians’ clinical decisions
Much worse	22 (3; 2-5)
Worse	263 (39; 35-43)
No change	193 (29; 25-32)
Better	186 (28; 24-31)
Much better	12 (2; 1-3)
Drug company profits
Much worse	7 (1; 0-2)
Worse	6 (1; 0-2)
No change	29 (4; 3-6)
Better	339 (50; 46-54)
Much better	295 (44; 40-47)
Controlling high drug prices
Much worse	165 (24; 21-28)
Worse	252 (37; 34-41)
No change	185 (27; 24-31)
Better	57 (8; 6-11)
Much better	17 (3; 1-4)
Should drug companies charge less for drugs prescribed for non-FDA-approved indications than for drugs prescribed for FDA-approved indications?
Definitely yes	103 (15; 13-18)
Probably yes	200 (30; 26-33)
Probably no	250 (37; 34-41)
Definitely no	117 (17; 15-20)

Abbreviations: ED, emergency department; FDA, Food and Drug Administration; ICU, intensive care unit.

^aPhysicians must have previously reported being clinically active (ie, ≥40% of their time and ≥21 h/mo in patient care activities) in nonacute settings (ie, ≤50% of their time in the ICU, ED, or cardiac catheterization laboratory).

More than half of physicians (54%; n = 367) felt that manufacturers need not charge less for drugs used off-label, while 70% (n = 467) supported similar pricing whether FDA approval was based on surrogate or patient outcomes.

Discussion

Most respondents favored a relatively high bar for new drug approval and preferred that the FDA maintain its current restrictions on off-label marketing. Since the FDA approves about one-third of new drugs on the basis of a single randomized trial and often relies on results from nonrandomized studies,⁴ we thus observed a disconnect between many physicians’ perceptions of the drug approval process and its current reality. Physicians’ views on the dangers of widespread off-label marketing undercut judges’ rationales for permitting off-label marketing; for example, the majority decision in 1 notable case did not recognize that off-label marketing statements, even if not fraudulent, may still fall well short of accurate, unbiased information.⁵

Our results are limited by social desirability and also nonresponse; however, the close match to known physician statistics⁶ is reassuring. The views of practicing clinicians should inform evolving FDA policies about drug approval standards and off-label promotion.