Pitfalls and Safeguards in Industry-Funded Research.

Background: Physicians should follow ethical principles in their relationships with industry and be mindful that such relationships-if they are perceived as conflicts of interest-can undermine trust in the patient-physician relationship. Methods: By identifying potential pitfalls and safeguards that can help prevent problems, this article focuses on ensuring that physician-industry relationships do not result in ethical transgressions or cause damage to doctor-patient relationships. Results: Patient trust in physicians can be undermined by the perception that a physician-investigator is operating in the best interest of the research rather than the best interest of the patient. Payments from the pharmaceutical industry to physician-investigators are transparent because of the Sunshine Act, and patients can easily determine if their personal physicians have received money from industry. Research subsidies from industry should represent fair market value for the work performed. Postmarketing trials with the primary goal of increasing familiarity with a drug and prescribing rates should be avoided. Medical societies play an important role in establishing standards for professional conduct. Conclusion: Ethically sound actions in physician relationships with industry should be guided by professional standards, medical society guidelines, and local institutional policies.

An Analysis of Informed Consent Form Readability of Oncology Research Protocols.

Twenty-two percent of adults in the United States have only basic health literacy skills. We used a multiple linear regression model to identify associations between readability of informed consent documents with study sponsor, study phase, and approval year using a sample of 143 oncology studies at Ochsner Medical Center. The M ± SD Flesh-Kincaid Reading Grade Level (RGL) was 10.33 ± 0.85 and Flesh Reading Ease (FRE) was 52.89 ± 5.49. National Cancer Institute studies had a significantly lower mean RGL and FRE as compared with other sponsors (RGL 9.85 ± 0.66 vs. 10.72 ± 0.79; p value < .0001). Mean RGL did not differ by study phase. Future research should include assessment and improvement of the readability of informed consent documents.

Do N-of-1 Trials Need IRB Review?

There is no standard policy regarding the regulatory or institutional approval of N-of-1 trials in the United States. The objective of this study was to examine whether institutional review boards (IRBs) accredited by the Association for the Accreditation of Human Research Protection Programs (AAHRPP) consider N-of-1 trials as meeting the definition of human subjects research (45CFR46.102) and requiring IRB approval. A questionnaire was distributed via email to 170 AAHRPP-accredited IRBs in the United States. Responses were analyzed using statistical and qualitative methods. Nineteen of 59 respondents reported viewing N-of-1 trials as research. Twelve respondents reported having a policy regarding N-of-1 trials, and in all cases, such policies did not consider N-of-1 trials as meeting the definition of research. This topic deserves wider examination in the IRB literature and community to inform policies and guidance as N-of-1 trials become more common in the pursuit of personalized, precision medicine.

DNR, DNAR, or AND? Is Language Important?

The American Heart Association in 2005 moved from the traditional do not resuscitate (DNR) terminology to do not attempt resuscitation (DNAR). DNAR reduces the implication that resuscitation is likely and creates a better emotional environment to explain what the order means. Allow natural death (AND) is the name recommended in some settings to make the meaning even clearer. Most hospitals still use the obsolete DNR term. Medical staffs should consider moving to DNAR and in some settings to AND. Language is important.

Protecting human research subjects: the past defines the future.

The creation of Institutional Review Boards to assure the protection of research subjects came out of terrible research abuses that resulted in the Belmont Report and federal regulations establishing rules for federally funded research and its independent review. The Common Rule became widely accepted as the way to oversee human research that is funded by federal agencies, or used in FDA submissions. The Office of Human Research Protections, now under the Secretary of DHHS, created Federalwide Assurances with groups that receive federal funding and others, the vast majority of which have agreed to apply the same ethical rules to all research regardless of funding source. There are controversies over the best methods to protect human research subjects, confusion about how to handle some of the gray areas, increased regulatory burdens, and debates about the adequacy of the IRB system. New exciting directions have evolved and overall, research subjects appear better protected than ever.

Data mining a diabetic data warehouse.

Diabetes is a major health problem in the United States. There is a long history of diabetic registries and databases with systematically collected patient information. We examine one such diabetic data warehouse, showing a method of applying data mining techniques, and some of the data issues, analysis problems, and results. The diabetic data warehouse is from a large integrated health care system in the New Orleans area with 30,383 diabetic patients. Methods for translating a complex relational database with time series and sequencing information to a flat file suitable for data mining are challenging. We discuss two variables in detail, a comorbidity index and the HgbA1c, a measure of glycemic control related to outcomes. We used the classification tree approach in Classification and Regression Trees (CART) with a binary target variable of HgbA1c >9.5 and 10 predictors: age, sex, emergency department visits, office visits, comorbidity index, dyslipidemia, hypertension, cardiovascular disease, retinopathy, end-stage renal disease. Unexpectedly, the most important variable associated with bad glycemic control is younger age, not the comorbiditity index or whether patients have related diseases. If we want to target diabetics with bad HgbA1c values, the odds of finding them is 3.2 times as high in those <65 years of age than those older. Data mining can discover novel associations that are useful to clinicians and administrators [corrected].