Survey included 407 patients.
eAppendix 1. Consent form in English
eAppendix 2. Consent form in Spanish
eAppendix 3. Questionnaire in English
eAppendix 4. Questionnaire in Spanish
Customize your JAMA Network experience by selecting one or more topics from the list below.
Stein GE, Kamler JJ, Chang JS. Ophthalmology Patient Perceptions of Open Payments Information. JAMA Ophthalmol. 2018;136(12):1375–1381. doi:10.1001/jamaophthalmol.2018.4167
What do patients think about physicians receiving financial remuneration from the medical industry?
In this survey of 407 participants, 212 (53.5%) wanted to know if their physician receives payments of any kind from the medical industry, and 161 (41.9%) disapproved of their physician receiving $100 or more in payments of any kind.
While not necessarily generalizable to other patient groups, these data suggest that patients would like to know if their physicians receive payments from the medical industry, and many of them would disapprove of physicians receiving payments.
Since 2014, medical industry payment data to physicians have been public via the Open Payments database. Patient opinions regarding these data help us to understand concerns and policymakers to improve reporting mechanisms for such payments.
To assess patient perceptions of Open Payments information.
Design, Setting, and Participants
This cross-sectional, questionnaire-based study was conducted in 3 ophthalmology clinic sites of an academic institution in Manhattan, New York City, New York. All patients older than 18 years who were waiting for appointments were eligible. Data were collected from January to June 2016 and analyzed from June to September 2016.
Participants answered 27 questions about the Open Payments database in English or Spanish. Demographic information was also collected.
Main Outcomes and Measures
Key questionnaire results included patient awareness of the Open Payments database and perceptions of physicians’ financial relationships. Cronbach α validation of the survey was performed, and Poisson multivariable regression analysis was performed to evaluate the association between patient characteristics and responses.
A total of 407 individuals participated. The mean (SD) age of study participants was 58.8 (17.9) years, and 220 (54.2%) were women. Of these, 30 (7.3% [95% CI, 5.1%-19.4%]) were aware of the Open Payments database, and 109 (26.8% [95% CI, 24.8%-34.0%]) planned to access it. More than half (n = 212; 53.5% [95% CI, 48.6%-58.5%]) wanted to know if their physician receives industry payments. Regarding payments of any kind valuing $100, 161 (41.9% [95% CI, 37.0%-46.9%]) disapproved. Similarly, 178 (45.8% [95% CI, 40.8%-50.7%]) disapproved of $500 payments, and 221 (57.0% [95% CI, 52.0%-61.9%]) disapproved of $25 000 payments. Poisson multivariable regression analysis demonstrated that participants who took the survey in Spanish were 38% more likely to approve of physicians receiving payments than were those who took the survey in English (incidence rate ratio [IRR], 1.38 [95% CI, 1.19-1.59]; P < .001). For every 1 year of age, the likelihood of approval for a physician receiving payments decreased by 1% (IRR, 0.995 [95% CI, 0.99-1.00]; P = .007). Participants with graduate degrees were 20% less likely to approve of physicians receiving payments, compared with those with less than a high school degree (IRR, 0.80 [95% CI, 0.66-0.97]; P = .02).
Conclusions and Relevance
If the survey is validated, and if these results are generalizable outside of the 3 academic centers in ophthalmology surveyed, the findings suggest that many patients disapprove of physicians receiving payments from industry. However, few patients had accessed the Open Payments database or planned to access it. Further investigation is required to determine if these results can be generalized for other settings.
The Open Payments database was established in 2014 by the Center for Medicare and Medicaid Services to provide greater transparency of physician and hospital relationships with industry.1 Any payments or transfer of value to physicians and hospitals from industry are reported and available online to the public. Payments are divided into multiple categories, such as consulting, royalties, speaker honoraria, food, travel, gifts and charitable contributions, and research support. In the first 5 months in which Open Payments data were collected (August 1, 2013, to December 31, 2013), 55 996 individual payments were made to 9855 ophthalmologists.2
It is unclear if patients are interested in receiving these data and how they view, interpret, and value this information. Recent studies suggested that even small industry payments are correlated with increased usage of associated medications, even if causation is still unclear.3-5 In this study, we sought to determine how patients in an ophthalmology practice felt about professional relationships between physicians and the drug and medical device industries. Further understanding of what patients find to be important may aid in improving the data reporting system and may make physicians aware of how this information is perceived by the public.
This was a cross-sectional analysis of patients receiving ophthalmic care at Columbia University Medical Center in New York City, New York. Institutional review board approval was obtained for the facility where the study was conducted. Participants were given information about the study and gave verbal consent prior to survey completion. The study was compliant with the Health Insurance Portability and Accessibility Act, and the research adhered to the tenets of the Declaration of Helsinki.6
Prospective participants were patients of the ophthalmology service at Columbia University Medical Center. Patients were approached in the waiting rooms of 2 different floors within the Eye Institute at the main campus in Washington Heights and at a satellite clinic in Midtown Manhattan. Eligible participants had to be at least 18 years of age. All adult patients who expressed understanding of informed consent and stated that they had not previously participated were eligible. Most participants were recruited and enrolled by one physician (G.E.S.). The physician for a particular participant was not tracked to instill participant confidence that answers would be anonymous.
The questionnaire was developed with the goal of understanding patient perception while limiting response bias and ensuring anonymity. The authors and other physicians in the academic department composed the questions.
Responses were anonymous. Patients were asked if they had previously completed the questionnaire to prevent multiple responses. There was no compensation for completing the survey.
Participants were given background information about the study, and verbal informed consent was obtained. To limit external influence, participants who had questions about individual survey items after starting the survey were instructed to answer the question as best as they thought they could. The informed consent agreement is provided in eAppendix 1 and eAppendix 2 of the Supplement.
Participants had the option of completing the consent and questionnaire in English or Spanish. Spanish-language documents were approved by the Spanish Translation Center at Columbia University Medical Center. Most participants completed the questionnaire in the clinic, while a few took it home and returned it at their next visit; this rate was not measured.
Demographic information made up the first 5 questions. Participants were asked about age, sex, highest level of education achieved, health insurance, and subjective sense of vision. Regarding vision, participants were asked if they had no visual impairment, had visual impairment but were able to perform routine activities independently, or had visual impairment that necessitated assistance for daily tasks.
Patients were asked if they were aware of the Open Payments database or Sunshine Act, and if they had tried to or planned to access the database. Participants were also asked if they were physicians themselves.
A series of questions asked patients if they agreed, disagreed, or had no opinion about wanting to know if their physician or his or her staff received gifts, meals, or payments of any kind from a drug or medical device company. A second series of questions asked if patients approved, disapproved, or had no opinion about their physician accepting meals, travel reimbursement, gifts, books, royalties, or other payments from a drug or medical device company. Several of these questions were qualified by the physician receiving payment to learn about the product, explain a product to others, or give an expert opinion. Last, a series of questions were aimed at identifying the quantity of payments that would be acceptable to patients. Patients were asked if they approved, disapproved, or had no opinion about their physician receiving payments of $100, $500, $1000, $10 000, $50 000, or $100 000. English and Spanish versions of the questionnaire are in eAppendix 3 and eAppendix 4 in the Supplement.
Power analysis was performed and suggested that a sample size of 400 was enough to demonstrate statistical significance with a 5% margin of error and a 95% confidence level. Reliability and internal consistency are key steps for the validation of questionnaires aimed to assess patient-related outcomes, as recommended by the US Food and Drug Administration (FDA).7 A separate validation should be performed for questionnaires in different languages. Therefore, to assess the reliably and internal consistency of our newly developed questionnaire, Cronbach α was calculated for the English and Spanish versions of the survey. General and FDA guidelines recommend a Cronbach α of .70 to .95.8
Answers were placed into categories of approval, disapproval, or no opinion. A composite score of approval for each participant was calculated. These data were analyzed with Poisson regression analysis to determine if certain background variables were associated with the likelihood of approval of receiving payments from medical industry. Univariable Poisson linear regression analysis was used to look for an association between background data and likelihood of approval for each variable. Next, multivariable Poisson regression analysis was performed to search relationships between background criteria and likelihood of approval. Predictors with P < .25 from the univariable analysis were tested in the multivariable analysis. For each background variable, an incidence rate ratio (IRR) as well as the P value and 95% confidence interval were calculated. The IRR can be interpreted as the likelihood of approval of payments to physicians compared with the reference variable.
All tests were 2-sided, and P < .05 was considered significant. Statistical analysis was performed with Stata version 14 (StataCorp LP). Data was collected from January to June 2016, and the analysis was performed from June to September 2016.
A total of 407 patients participated in the study. The mean (SD) age was 58.8 (17.9) years. More women (n = 220 [54.2%]) than men (n = 186 [45.8%]) participated in the study. Approximately two-thirds of the participants completed the questionnaire in English (n = 280 [68.8%]) rather than in Spanish (n = 127 [31.2%]).
Participants had a range of education: 54 (13.7%) reported having completed some high school education or less, 124 (31.6%) reported having completed high school, 104 (26.5%) reported having completed college, and 111 (28.2%) reported having completed a graduate degree. Thirty percent of participants (n = 82 [29.7%]) who took the survey in English completed high school or less, compared with 96 participants (82.1%) who took the survey in Spanish, whereas 194 participants (70.3%) who took the survey in English completed college or more, compared with 21 participants who took the survey in Spanish (17.9%). A χ2 test demonstrated that the differences in education level between participants in the 2 language groups were significant (χ23 = 125.9; P < .001).
Only 7 participants (1.7%) had no health insurance, 160 (39.3%) had Medicaid, 134 (33.3%) had Medicare, and 101 (25.1%) had primary commercial insurance. Participants had varying quality of vision: 87 (21.9%) identified themselves as having no visual impairment, 272 (68.3%) reported having some visual impairment, and 39 (9.8%) reported having significant visual impairment necessitating assistance for routine activities (Table 1).
A minority of 30 participants (7.7% [95% CI, 5.1%-10.4%]) were aware of the Open Payments database, and 35 participants (9.0% [95% CI, 6.2%-11.9%]) were aware of the Sunshine Act. Very few participants (n = 12; 3.1% [95% CI, 1.4%-4.9%]) had tried to access the Open Payments database, but 109 (29.4% [95% CI, 24.8%-34.0%]) stated that they planned to access the database in the future. A few participants (n = 11; 2.8% [95% CI, 1.2%-4.4%]) of participants identified themselves as physicians (Table 2).
A total of 160 participants (42.2% [95% CI, 37.2%-47.2%]) stated that they did not care to find out if their physician had received industry payments. For those who wanted to know, 60 participants (15.8% [95% CI, 12.2%-19.5%]) stated that an online database alone was enough, and 58 participants (15.3% [95% CI, 11.7%-18.9%]) wanted some additional form of notification (mail, email, or notification at the time of a clinical visit). One hundred one participants (26.6% [95% CI, 22.2%-31.1%]) felt that their physician should tell them at the visit in addition to having other methods of finding the information.
More than half of all participants (n = 212; 53.5% [95% CI, 48.6%-54.5%]) agreed with the statement that they would like to know if their physician receives payments of any value. Roughly one-third wanted to know that their physician accepted a meal for himself or herself (n = 128; 32.2% [95% CI, 27.6-36.8]) or for their office staff (n = 138; 34.4% [95% CI, 29.8-39.1]). Slightly more than half of all participants wanted to know if their physician received compensation for consulting or lecturing (n = 219; 55.2% [95% CI, 50.3-60.1]; Table 3).
Patients expressed a range of opinions about whether they approved, disapproved, or had no opinion about certain payments to physicians that are reported in the database (Table 3). Notably, 198 participants (50.9% [95% CI, 45.9%-55.9%]) disapproved of their physician receiving gifts from a company (with no qualifying reason given) and 109 participants (28.2% [95% CI, 23.7%-32.7%]) disapproved of their physician receiving a meal (with no qualifying reason given). Only 70 participants (17.8% [95% CI, 14.0%-21.6%]) disapproved of their physician receiving an educational course or textbook from medical industry.
When the value of the transfer to physicians was presented as an amount, disapproval increased with increasing value of the transfer. Nearly half of the participants (n = 161; 41.9% [95% CI, 37.0%-46.9%]) disapproved of their physician receiving $100 per year in payments of any kind from the medical industry, and this increased to 221 participants (57.0%; 95% CI, 52.0%-61.9%) for payments of $25 000 per year (Figure).
Cronbach α, a measurement of the reliability of an instrument, was calculated for both language versions of the questionnaire. In this study, it can be interpreted as the reliability of the questionnaire at measuring patient approval or disapproval of data associated with the Open Payments Database. Cronbach α was .89 for the English version and .95 for the Spanish version of the questionnaire, suggesting excellent reliability.
A univariate model was constructed with Poisson regression models and IRRs were calculated. This can be interpreted as the likelihood of an event occurring. An IRR greater than 1 indicates that the group is more likely to approve of physicians receiving payments and an IRR less than 1 indicates that the group is more likely to disapprove of physicians receiving payments.
In the univariate model, the association between background variables and IRRs was tested without controlling for any factors. Participants who took the survey in Spanish were 34% more likely to approve of physicians receiving payments (IRR, 1.34 [95% CI, 1.21-1.49]; P < .001). Women were 12% more likely to approve of physicians receiving payments, although this was barely statistically significant (IRR, 1.12 [95% CI, 1.00-1.26]; P = .05). Each year of age was associated with a 1% less likelihood of approval (IRR, 0.997 [95% CI, 0.99-1.00]; P = .04). Those who completed college and graduate school were 22% (IRR, 0.79 [95% CI, 0.67-0.91]; P = .002) and 28% (IRR, 0.719 [95% CI, 0.61-0.85]; P < .001) less likely to approve of physicians receiving payments than were those with less than a high school education, respectively. Those who were aware of the database, and who had already tried to search for the database, were also more likely to disapprove of physicians receiving payments from pharmaceutical and medical device industries.
In the final multivariable regression analysis, we included only those variables with P < .25 from the univariate model. We found that those who took the survey in Spanish were 38% more likely to approve of physicians receiving payments (IRR, 1.38 [95% CI, 1.19-1.59]; P < .001). In this model, there was no longer a statistically significant association with sex. Again in this model, participants were 1% less likely to approve of physicians receiving payments from industry for each year of age (IRR, 0.995 [95% CI 0.99-1.00]; P = .01). Those who completed graduate school were 20% less likely to approve than those who completed less than college (IRR, 0.80 [95% CI, 0.66-0.97]; P = .02). Those who stated that they were aware of the database were 16% less likely to approve of physicians receiving payments (IRR, 0.84 [95% CI, 0.70-1.00]; P = .047; Table 4).
While the association between marketing and physician behavior had been previously reported, Open Payments data have been used in combination with other Centers for Medicare and Medicaid Services data to evaluate physician behavior. Several recent reports have concluded that there is an association between pharmaceutical payments and the selection of anti–vascular endothelial growth factor agents.3,5 This form of industry influence has also been reported in other specialties of medicine.4 This study revealed that the surveyed patients were concerned about their physicians receiving industry payments. Consistent with this set of beliefs, patient disapproval increased as the total value of the payment increased. Other studies also concluded that a significant percentage of patients would like to know about their physicians receiving payments.9 Our results demonstrated that disapproval of this behavior was positively associated with education and age, similar to previous findings.10,11 This may be particularly impactful for ophthalmologists, who see a large share of patients older than 65 years.
We included subjective sense of vision as a background characteristic, because we hypothesized that patients with greater vision loss might be less concerned about involvement with medical industry. Our results showed no association between subjective sense of vision and patient opinions.
The findings may be applicable to all areas of medicine, because ophthalmology patients likely interact with other physicians. As both a medical and surgical subspecialty, ophthalmology is in a unique position, because potential conflicts of interest may occur with brand-name and high-cost medications as well as use of surgical devices and techniques. Therefore, ophthalmologists may be under higher scrutiny by patients regarding these financial relationships.
While many patients were concerned about remuneration from the medical industry, other findings suggested the influence of the database is limited at the present. Our results showed a lack of awareness of Open Payments data. These results are consistent with a survey study by Pham-Kanter et al,12 examining if patients were treated by health care professionals who had received payments in the Open Payments database. In their study, 12% of respondents, compared with 8% in our survey, were aware of the database. Generally, in both studies, patients were of similar education background and were primarily urban-dwelling. In this study, we had a slightly older population, with higher rates of Medicare and Medicaid coverage.
In our data, patients disapproved of the physician and staff receiving payments at similar rates of disapproval to the same behavior conducted by the physician alone. This is important because Open Payments may undervalue meals provided to staff; perhaps this omission should be corrected in the future.13 Fewer respondents, though still a large percentage, disapproved when payments were tied to a specific task, such as lecturing, consulting, education, or development of intellectual property.
Patients who took the survey in Spanish were more likely to approve of payments than their counterparts who took the survey in English. This finding was significant after controlling for education and may reflect a cultural difference in physician perception. A high correlation between language and education in our respondents made it difficult to truly separate these 2 variables, however, because of the significant association between participant education level and the language chosen for survey completion.
A measurement error may affect any survey, because answers may not reflect what participants actually believe or would do outside of the survey.14 The responses in this study could be influenced by participants learning about Open Payments in the survey questions or during the informed consent. We tried to limit error by explaining the Open Payments Database in a brief and concrete way, so that respondents could understand the purpose of the study and agree to participant. Patient recruitment in a systematic fashion was reinforced because 1 physician did nearly all of the enrollment. Our analysis of the survey, demonstrated by the high Cronbach α, showed that it was a reliable instrument.
The study population was participants in New York City during 2016, and different results could have been observed in another place or time. Compared with the US population, our study sample had a high graduate degree completion rate (28.2%). Opinions of individuals who were not involved in ophthalmic care were excluded from this survey population. Survey respondents may have been slightly younger and healthier than a general ophthalmology practice distribution, possibly because older and sicker patients declining survey participation. More respondents (40%) had Medicaid coverage than the national mean, which was 19.4% in 2016.15
Political affiliations were not investigated in this survey. Because New York City may be more liberal and/or Democratic than other areas of the country, this may affect the generalizability of the results. However, to our knowledge, patient opinions on this topic and correlation with political party affiliation have not been measured. Also, party affiliation may not dictate opinions on single issues. Finally, the practice catchment area includes the New York, New Jersey, and Connecticut metropolitan area, which is ethnically and socioeconomically diverse. Generalizability could certainly be improved by including participants from a wider area, but it seems reasonable to conclude that a nontrivial percentage of patients has concerns about financial relationships between physicians and industry.
Industry interactions are part of medical care and important to advancing the ophthalmologic field. However, some argue that public interests may be undermined if they are not aligned with the financial interests of the medical industry.16 Disclosures aim to increase awareness of pharmaceutical industry involvement. Critics of disclosure argue that the effect of disclosure is limited and that more needs to be done.16 Ironically, disclosure may actually engender trust, because the physician is perceived as being very honest about his relations with industry.17
The results from this study show that this patient cohort felt that physicians should generally avoid financial interactions with industry. However, it appears that many did not access the Open Payments database, and it is unclear if or when they will. Incorporating patient concerns may guide the government to more effectively report this information. It behooves ophthalmologists to understand how industry interactions are reported and how this information affects the relationships of physicians with patients.
Corresponding Author: Jonathan S. Chang, MD, Department of Ophthalmology and Visual Sciences, School of Medicine and Public Health, University of Wisconsin, 2870 University Ave, Ste 206, Madison, WI 53705 (firstname.lastname@example.org).
Accepted for Publication: July 13, 2018.
Published Online: September 13, 2018. doi:10.1001/jamaophthalmol.2018.4167
Author Contributions: Drs Stein and Chang had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Stein, Chang.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Stein.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Stein.
Administrative, technical, or material support: All authors.
Conflict of Interest Disclosures: All authors have completed and submitted ICMJE Form and Disclosure of Potential Conflicts of Interest. No disclosures were reported.
Funding/Support: This study was supported by the Gerstner Family Foundation, an unrestricted grant to Columbia University by Research to Prevent Blindness, an unrestricted grant to the Department of Ophthalmology and Visual Sciences to the University of Wisconsin Madison Department of Ophthalmology and Visual Sciences, and a National Eye Institute Vision Research Core grant (P30 EY016665, the University of Wisconsin Madison Department of Ophthalmology and Visual Sciences).
Role of the Funder/Sponsor: The sponsors had no role in the design and conduct of the study; collection, management, analysis and interpretation of the data; preparation, review or approval of the manuscript; and decision to submit the manuscript for publication.
Additional Contributions: We thank C. Gustavo de Moraes, MD, Department of Ophthalmology, Columbia University Medical Center, for assistance with statistical analysis, and Michael Deng, BA, Department of Ophthalmology, Columbia University Medical Center, for administrative support. They were not compensated for their contributions.
Create a personal account or sign in to: