Posts Tagged ‘research’

Unpublished Studies

Wednesday, January 11th, 2012

A study published in BMJ reports that fewer than half of clinical trials funded by NIH grants resulted in publications within 30 months of completion. Publication of NIH funded trials registered in ClinicalTrialsResearchers at Yale reviewed 635 clinical trials conducted between 2005 and 2008 and found that only 46% appeared in peer-reviewed, MEDLINE-indexed journals by 2011.

Even when the authors extended their search to 51 months after the completion of trials, one-third remained unpublished. They can only speculate on why investigators might not publish their findings. Certainly, some might appear in abstracts or posters that do not appear in electronic indexes.

There’s some comfort in knowing that studies completed later in the sample period were more likely to appear in print. Because public funds supported the trials, it is important that their results, even if negative, be disseminated widely.

Performing Database Searches

Wednesday, August 10th, 2011


Maggio, Lauren A. MS(LIS), MA; Tannery, Nancy H. MLS; Kanter, Steven L. MD. AM Last Page: How to Perform an Effective Database Search. Academic Medicine 2011. 86(8):1057.

Incentives to Publish

Wednesday, August 10th, 2011

South Korean scientists who publish in top-flight journals like Science and Nature receive a $2800 bonus from their government. Turkish scientists who do the same can count on a bonus worth 7.5% of their salary. Other countries reward institutions for publication rates.

Researchers have now looked at whether such incentives have resulted in a greater publication success. Of the 110,870 original research articles submitted to Science over the last 10 years, first authors came from 144 different countries. 7.3% of submissions were accepted.

The study concludes that cash incentives indeed increase journal submissions, but not necessarily acceptance. Rewarding publication with career promotion leads to both greater submission and acceptance rates. Their findings suggest that monetary bonuses are enough to affect the quantity of research, but to improve quality, more symbolic prizes are needed.

first authors from 144 different countries submitted 110,870 original research articles; 7.3% of these submissions were accepted for publication, with first authors from 53 different countries

Human Subjects Revised

Tuesday, July 26th, 2011

The Department of Health and Human Services have released proposed changes to the review of human subjects research. The changes, summarized by the New York Times yesterday, would tighten some controls and loosen others.

On the side of more regulation, it would bring more studies, even those conducted with private funds, under the purview of an Institutional Review Board if the institution accepts any federal money. On the looser side, it would create a category of "excused" research allowing social and behavioral studies to avoid review.

Though the amendments seem thoughtful, they may not address some of the complaints that researchers have with their own IRBs like lack of transparency. It will be helpful if this proposal sparks discussion among university IRBs about how best to communicate their goals to researchers.

Research Funding

Friday, July 15th, 2011

The National Science Foundation has calculated research and development expenditures for science and engineering fields at U.S. universities in FY 2009. The institutions surveyed spend close to $55 billion that year on research.

As in most years, the university spending the most on science and engineering research is The Johns Hopkins University with $1.8 billion. Over 80% of those funds came from the federal government. Boston University ranks #72 on the list with a research budget of $280 million. Over 91% of the funds come from the federal government.

Looking just at expenses for life science research, UC San Francisco tops the list with $930 million spent on research and development. Boston University ranks 75th at $174 million.

Beyond the R01

Thursday, July 14th, 2011

In 2007, two economists were interested in how much an R01 grant from the NIH influences subsequent publication. Presumably, researchers who receive funding would generate more data and produce more academic papers. Quantifying the impact of government funds provides an important argument for continued public investment in science.

The authors analyzed all applications for R01 support (both successful and unsuccessful) between 1980 and 2000. The sample included over 18,000 individual researchers. Surprisingly, the data showed that scientists who received NIH grants produced one additional paper over the subsequent five years compared to those who did not receive the grant. Nor do the citation rates of grantees differ significantly from their unfunded counterparts.

The findings do not necessarily devalue the importance of  NIH-funded research. Biomedical benefits go beyond the publishing of papers. The authors hypothesize that researchers who fail to receive NIH funds usually locate other sources of support or shift their area of interest to a more fundable topic. Resilience is key.

Pfizer in Boston

Thursday, June 9th, 2011

The giant pharmaceutical company Pfizer already has partnerships with academic medical centers in San Francisco and New York City. Now they're adding Boston. According to a press release, the company will house its Center for Therapeutic Innovation in the Longwood area and invest $100 million over five years to bring basic science discoveries to clinical applications.

The semi-autonomous center will be headed by a Pfizer scientist but draw postdocs from academic medical centers. A brochure describes that faculty from academic medical centers will be involved on a steering committee that helps decide projects and monitors conflicts of interest. Important to note in light of a recent U.S. Supreme Court decision involving a Stanford biomedical researcher, Pfizer retains first right to license any clinical probes that come out of the research.

Computer Breach

Monday, April 25th, 2011

For the past 15 years, epidemiologist Bonnie Yankaskas has received federal funding to maintain and analyze a database of mammography results from North Carolina women. Two years ago, her employer, the University of North Carolina School of Medicine, discovered that her server's security had been compromised.

Although Dr. Yankaskas had hired a computer expert to oversee the database, the medical school ultimately found her responsible for the breach as project PI. The school first tried to fire the professor. Then they tried to demote her. In a settlement, Dr. Yankasas will retain her title and salary, but retire at the end of the year.

Even without evidence of harm to the subjects in the database, it is clear that the PI is responsible for training and overseeing the personnel who manage the data. Along with the prestige of conducting important research comes the responsibility for maintaining ethical standards.


Wednesday, April 20th, 2011

Between 1997 and 2009, 1,164 biomedical research articles were retracted. In over half the cases, the cause was scientific misconduct ranging from lack of IRB approval to manipulated data. Though worrisome, these articles represent a small portion of all the literature indexed in PubMed.

More concerning is that, according to a new study, many of these articles continue to be cited well after the retraction is posted. Only 6% of the subsequent citations acknowledge that the original article was flawed. The vast majority of citations occur in literature reviews. Because any search of PubMed would turn up a large "Redacted" watermark on the original article, it could be that authors are not conducting fresh searches to find citations for the literature review section.

Scientific Map

Monday, April 11th, 2011

Inspired by a graphic representation of friendship networks on Facebook, Olivier Beauchesne decided to create a similar map of scientific collaboration. Because he works for a bibliometric analysis company, he had access to a large database of research article citations.

For every publication jointly authored by researchers in different cities, he drew a connection. So, if a faculty member at BU published with a faculty member at the Sorbonne, a line would connect Boston and Paris. To see the complete picture, click here.

Even without an overlay of a geopolitical map, it's clear both how much collaboration takes place and between which countries.