Introduction
The relationship between doctors and the pharmaceutical industry is a subject of considerable public interest and ethical debate. A common question that arises is: Do Doctors Get Paid For Prescribing Drugs? This question taps into concerns about potential conflicts of interest and whether financial incentives could influence a physician’s medical decision-making, ultimately affecting patient care. In the United States, financial interactions between drug companies and physicians are not only common but also substantial. From 2015 to 2017, a significant majority, 67% of US doctors, received payments from the pharmaceutical industry. In numerous specialties, this figure climbed above 80%, and the monetary value of these payments has been on the rise in recent years. In 2018 alone, these payments amounted to a staggering $2.18 billion.
This widespread practice has sparked concerns about the potential for undue influence of commercial interests in healthcare. While some payments might be substantial enough to constitute a significant portion of a physician’s income, even smaller payments can create a sense of obligation, known as reciprocity, potentially nudging doctors towards increased prescribing of certain medications. Previous research has already indicated a link between physician prescribing habits and various forms of interaction with the pharmaceutical industry, such as receiving product information, direct engagement with sales representatives, or accepting free drug samples. These interactions often involve financial payments, but the lack of transparency around these payments has historically made it challenging to fully understand their impact.
However, the landscape changed with the introduction of the Open Payments system in 2013, mandated by the “Sunshine Act.” This system mandates the public disclosure of financial transfers exceeding $10 from drug and medical device manufacturers to US physicians and other healthcare providers. Open Payments has opened the door to large-scale investigations into the financial dimension of physician-industry collaborations. Since its inception, numerous studies have leveraged Open Payments data to explore whether industry payments affect physician prescribing patterns. Given the recent surge in research in this area, a comprehensive overview of this body of work is essential to understand the full scope of the issue.
These studies have examined industry payments across diverse medical specialties and drug categories, yielding varied results. Consequently, a clear consensus on the overall relationship between industry payments and physician prescribing, and whether this relationship is causal, has remained elusive. To address this gap, a systematic review was conducted to investigate the association between industry payments and physician prescribing practices. This review aimed to determine: 1) the nature of the association between payments and prescribing across different medical fields, and 2) whether there is sufficient evidence to conclude that these payments cause doctors to alter their prescribing behaviors.
Methods: Investigating the Payment-Prescribing Link
To conduct this comprehensive review, a rigorous methodology was employed to ensure the findings were robust and reliable.
Data Sources and Search Strategy
A comprehensive search strategy was implemented across five major databases: Medline (Ovid), Embase.com, the Cochrane Library (Wiley), Web of Science Core Collection (Clarivate), and EconLit (EBSCO). The search strategy was carefully designed using keywords and subject headings related to both prescribing practices and pharmaceutical industry relationships, such as “prescriptions,” “practice patterns,” “conflict of interest,” and “Open Payments.” Filters were used to exclude animal-only studies, and no language or date restrictions were applied to maximize the search’s breadth. Duplicate studies were removed, and databases were initially searched in September 2019, with an update in September 2020. Additionally, a separate search for existing reviews on the topic was conducted in MEDLINE, and Scopus was used to identify further relevant studies from the references of the included articles.
Study Selection and Eligibility Criteria
The retrieved studies underwent a rigorous selection process. Titles and abstracts were independently screened by two reviewers, and disagreements were resolved through group consensus. Studies deemed potentially relevant were then subjected to full-text review by two independent reviewers. Studies were included if they met specific criteria:
- Availability of full text.
- Empirical, peer-reviewed experimental or observational studies (excluding guidelines, opinion pieces, and reviews).
- Focus on physicians (including other independent clinical practitioners).
- Investigation of financial payments from pharmaceutical companies as the exposure.
- Examination of pharmaceutical product prescribing as the outcome.
Data Extraction and Quality Assessment
A standardized template was used to extract key information from the eligible studies, including study characteristics, analytic design (variables, statistical tests), results, and risk of bias. Data extraction was performed independently by one reviewer and verified by a second reviewer, with disagreements resolved through group discussion. The Risk of Bias in Non-Randomized Studies of Interventions (ROBINS-I) tool was utilized to assess the risk of bias for each analysis. This tool evaluates bias across several domains, assigning an overall risk of bias level (unclear, low, moderate, serious, or critical). Analyses with critical risk of bias were excluded from the evidence synthesis due to concerns about their reliability.
Data Synthesis and Analysis
Due to the heterogeneity in analytic approaches and study characteristics, a quantitative meta-analysis was not feasible. Therefore, a qualitative synthesis of the results was conducted. For each analysis, the association between payments and prescribing was categorized as positive (payments associated with increased prescribing), inverse (payments associated with reduced prescribing), or null (no significant association). Studies were categorized based on whether all analyses showed a positive association, no positive association, or a mix of positive and null findings. Temporal and dose-response relationships were also assessed for each analysis.
Results: Consistent Link Between Payments and Increased Prescribing
The systematic review process yielded a substantial body of evidence, indicating a consistent association between industry payments and physician prescribing behavior.
Study Selection and Risk of Bias
The initial database searches identified 3,460 unique studies. After screening and full-text review, 37 studies met the inclusion criteria. However, one study with critical risk of bias was excluded, leaving 36 studies for the final evidence synthesis. The risk of bias assessment revealed that 15 studies had moderate risk of bias, 21 had serious risk of bias, and, as mentioned, one had critical risk and was excluded. The most common source of bias was confounding, inherent in observational studies, and potential inaccuracies in Open Payments data. Despite these limitations, many studies demonstrated low risk of bias in other areas, such as missing data and outcome measurement, due to the use of large government datasets.
Characteristics of Included Studies
The majority of the included studies (35 out of 36) were published from 2016 onwards, reflecting the recent availability of Open Payments data. Medicare data was the most frequently used source for measuring physician prescribing (34 studies), and Open Payments was the predominant source for industry payment data (32 studies). Most studies were conducted in the US (all but one) and were nationwide in scope (32 studies).
Figure 1: This PRISMA flow diagram illustrates the study selection process, detailing the number of studies identified, screened, and included in the systematic review. It provides a clear visual representation of how the final set of 36 studies was derived from the initial search results.
The drugs studied varied, with 11 studies analyzing multiple drug classes, and others focusing on specific categories like opioids (7 studies), antineoplastics (3 studies), and anti-VEGF agents (3 studies). Twenty-two studies analyzed all physicians in aggregate, while others focused on specific specialties such as primary care, hematology-oncology, and urology. The types of industry payments assessed also varied, with most studies examining all general payments, and some focusing on subsets or research payments.
Table 1: Study Characteristics
Table 1: This table summarizes the key characteristics of the 36 included studies, providing a detailed overview of publication year, route and class of drugs studied, data sources for payments and prescribing, types of prescribing outcomes assessed, physician specialties analyzed, categories of industry payments evaluated, and assessment of temporal and dose-response relationships. It offers a structured summary of the diverse features of the research included in this review.
Consistent Positive Association
Across the 36 studies, a striking consistency emerged: 30 studies found only positive associations between industry payments and prescribing across all their analyses. The remaining 6 studies showed mixed results, with some analyses indicating a positive association and others showing no association. Importantly, no study found only null or negative associations. This pattern held even when considering the risk of bias, with a majority of both moderate and serious ROB studies still identifying a positive association.
Of the 101 individual analyses conducted within these 36 studies, an overwhelming 89 (88%) identified a positive association between industry payments and increased prescribing. Only 12 analyses found no significant association, and none found a negative association.
Types of Payments and Prescribing Outcomes
Studies consistently reported that food and beverage payments were the most frequent type of payment, while payments for consulting and honoraria, though less frequent, were of higher monetary value. Most studies (30) analyzed all general payments in aggregate. Interestingly, 5 out of 6 studies that compared food & beverage payments to other payment types found a stronger association between food & beverage payments and prescribing.
The most common prescribing outcome measured was prescribing volume for the drug of interest (15 studies). Other outcomes included prescribing costs, the fraction of prescribing for the drug of interest, and the fraction of prescribing for branded drugs.
Evidence for Temporal and Dose-Response Relationships
Nine studies assessed the temporal relationship between payments and prescribing. Six of these found that payments were associated with increased prescribing in subsequent years. Three studies using time series analysis reported immediate increases in prescribing following industry payments. Furthermore, 25 studies investigated dose-response relationships, and all of them found evidence that higher payment amounts or frequency were associated with greater changes in prescribing.
Discussion: Implications of Industry Payments on Prescribing Practices
This systematic review provides compelling evidence for a consistent association between financial payments from the pharmaceutical industry and physician prescribing practices. The findings indicate that doctors who receive industry payments are more likely to prescribe the drugs of the paying companies, leading to increased prescribing costs and a preference for brand-name drugs over generics. This association spans various physician specialties and drug classes, suggesting a broad influence of financial incentives on medical decision-making.
Potential Mechanisms and Causality
The observed association raises crucial questions about the underlying mechanisms. While it’s plausible that payments influence prescribing by creating an incentive to favor specific drugs, it’s also possible that drug companies target payments towards physicians who already prescribe their products frequently. However, the evidence from temporal analyses, particularly time series studies showing immediate prescribing increases after payments, strongly suggests a causal relationship: payments appear to drive changes in prescribing behavior. This is further supported by the consistent dose-response relationships observed, indicating that the magnitude of payments is linked to the extent of prescribing changes.
Figure 2: This figure visually summarizes the study results, highlighting the consistent positive association between industry payments and prescribing. Panel A shows the distribution of study results, categorized as “all positive” or “mixed,” while Panel B presents the results of individual analyses, classified as “positive,” “inverse,” or “null,” and further characterized by whether they assessed temporal and/or dose-response relationships. This figure provides a clear visual synthesis of the review’s main findings.
Impact on Prescribing Quality and Patient Care
The influence of industry payments on prescribing raises concerns about the quality of patient care. While in some cases, increased use of a beneficial drug might be a positive outcome, the overall distribution of industry payments suggests this is unlikely to lead to widespread improvements in patient outcomes. Pharmaceutical companies tend to focus promotional efforts on drugs that may be less effective or offer limited therapeutic advantages. This may be because well-established, highly effective drugs require less promotion, while those with marginal benefits need more aggressive marketing to gain market share.
Empirical evidence suggests that industry promotion can negatively impact prescribing quality. Studies have linked physician-industry interactions to inappropriate and lower-quality prescribing. Specifically, research indicates that industry payments are associated with increased prescribing of low-value drugs, including less effective options or more expensive alternatives to generics. Examples include increased prescribing of expensive brand-name drugs when cheaper generics are available and the promotion of newer, potentially less safe drugs over established, safer alternatives. These findings highlight the potential for industry payments to negatively affect prescribing quality and potentially harm patients by promoting the use of less optimal or more costly medications.
Policy Implications and the Need for Reform
Despite long-standing concerns about industry influence, payments to physicians remain prevalent. Current regulations, like the Sunshine Act, primarily focus on transparency through payment disclosure, without directly restricting these payments. Physician opposition has been a significant barrier to more substantial reform, often rooted in the belief that accepting payments does not influence their practice. However, the findings of this review challenge this belief, suggesting that industry payments do impact prescribing decisions.
To safeguard patient care and ensure independent medical decision-making, a reevaluation of the acceptability of industry payments to physicians is necessary. The medical community needs to move beyond historical opposition to reform and advocate for policies that minimize or eliminate financial conflicts of interest in prescribing practices.
Limitations
This review, like any research synthesis, has limitations. The included studies were primarily observational and relied on Open Payments data, which may contain inaccuracies. The observational nature of the studies means causality cannot be definitively proven, although the temporal evidence strongly suggests a causal link. Furthermore, most studies focused on Medicare data, potentially limiting generalizability to other patient populations. Publication bias, while less likely in this field, cannot be entirely ruled out. The majority of studies had serious risk of bias, although many of these risks were inherent to the study design and data sources.
Conclusion: Reclaiming Independent Medical Decisions
This systematic review provides robust evidence that financial payments from the pharmaceutical industry are consistently associated with increased physician prescribing of the paying companies’ drugs. This association is evident across diverse medical specialties, drug classes, and prescribing decisions. Crucially, temporal and dose-response evidence strongly suggests a causal relationship, indicating that these payments influence prescribing behavior. Furthermore, there is evidence that industry payments contribute to the increased use of lower-value drugs, potentially compromising prescribing quality and patient outcomes.
These findings underscore the need for a critical re-evaluation of the ethical implications of industry payments to physicians. To ensure that patient well-being remains the paramount concern, and to foster truly independent medical decision-making, the medical community should advocate for substantial reforms aimed at reducing or eliminating the influence of industry payments on prescribing practices. Transparency alone is insufficient; proactive measures are needed to mitigate the potential for financial incentives to compromise the integrity of medical care.
References
[List of references from the original article would be included here, maintaining the original numbering and links if possible, or re-formatted to a standard citation style.]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplement
NIHMS1722666-supplement-Supplement.docx (325.7KB, docx)
