Industrial payments and publication bias

Academic Influence and Its Relationship to Industry Payments in Orthopaedic Surgery

These authors point out that the Hirsch index (h-index) quantifies research publication productivity for an individual, and has widely been considered a valuable measure of academic influence. The Hirsch index (h-index) mathematically adjusts for the total number of publications and the number of times each publication has been cited. In orthopaedic surgery, higher h-indices have been associated with higher academic rank.

In 2010, the Physician Payments Sunshine Act (PPSA) was introduced as a way to increase transparency regarding U.S. physician-industry relationships. The purpose of this study was to investigate the relationship between industry payments and academic influence as measured by the h-index and number of publications among orthopaedic surgeons for the year 2014. They also examined the relationship of the h-index to National Institutes of Health (NIH) funding.

Of 3,501 surgeons, 78.3% received nonresearch payments, 9.2% received research payments, and 0.9% received NIH support. Nonresearch payments ranged from $6 to $4,538,501, whereas research payments ranged from $16 to $517,007. 

Surgeons receiving NIH or industry research funding had a significantly higher mean h-index and number of publications than those not receiving such funding. Surgeons receiving nonresearch industry payments had a slightly higher mean h-index and number of publications than those not receiving these kinds of payments. Both the h-index and the number of publications had weak positive correlations with industry nonresearch payment amount, industry research payment amount, and total number of industry payments.

Academic surgeons who receive industry research support or NIH funding tend to have higher hindices.

For the overall population of orthopaedic surgery faculty, the h-index correlates poorly with the dollar amount and the total number of industry research payments. Regarding nonresearch industry payments, the h-index also appears to correlate poorly with the number and the dollar amount of payments. 

Comment: While the authors conclude that "These results are encouraging because they suggest that industry bias may play a smaller role in the orthopaedic literature than previously thought," this conclusion is not supported by their data. A high h-index indicates only that the author has a relatively large number of relatively often referenced publications - it does not demonstrate that these publications are unbiased. 

When industry supports research, industry influences what is investigated, how the research is done, and what conclusions of the research are published.  If the results of the research do not satisfy the needs of the supporting company, the company can withdraw funding.  In addition, the powerful effect of non-research funding can be easily imagined when viewing the table above.

====

Use the "Search" box to the right to find other topics of interest to you.

How you can support progress in shoulder surgery

You may be interested in some of our most visited web pages including:shoulder arthritis, total shoulder, ream and run, reverse total shoulder, CTA arthroplasty, and rotator cuff surgery as well as the 'ream and run essentials'

Why does industry give money to surgeons for research?

Increasing Industry Support Is Associated with Higher Research Productivity in Orthopaedic Surgery

These authors point out that orthopaedic surgeons receive a disproportionately small share of funding from the National Institutes of Health, but they receive the largest amount of funding from industry sources. They sought to examine the association between payments from industry partners and research productivity among orthopaedic research authors, as well as to identify predictors of high research productivity.

United States-based physicians who published an article in 2016 in The Journal of Bone & Joint Surgery or The American Journal of Sports Medicine were included in this study. These authors were queried in the Centers for Medicare & Medicaid Services Open Payments database (OPD) to determine the amount of industry payments received, and on Scopus, a bibliometric web site, to assess the quantity (total publication count) and quality (Hirsch index [h-index]) of each author’s research.

Of the 766 included authors, 494 (64.5%) received <$10,000 per year, 162 (21.1%) received between $10,000 and $100,000, and 110 (14.4%) received >$100,000 in total payments. The h-index increased significantly from a mean (and standard deviation) of 13.1 ± 12.9 to 20.9 ± 14.4, and to 32.3 ± 16.7, from the lowest to highest payment cohorts, as did total publication count. 

When authors were stratified by academic position (assistant professor, associate professor, full professor, and nonacademic), those who received more industry payments (>$100,000) had a higher h-index and total publication count at all academic levels relative to lower-earning (<$10,000) authors. 

Independent predictors of a high h-index included industry payments of between $10,000 and $100,000 (odds ratio [OR], 1.63; p = 0.048), payments of >$100,000 (OR, 5.87), associate professorship (OR, 6.53), full professorship (OR, 33.38), and last authorship (OR, 2.22) (p < 0.001 for all comparisons unless otherwise noted).

The authors point out that in addition to direct research funding, industry payments to orthopaedic surgeons include nonresearch financial payments for fellowship funding, consulting, speaking engagements, product development, and royalties. 1.7% of the number of payments directed toward orthopaedic  surgeons were for royalties (i.e. not for research support), but these payments accounted for 69.5% of the total monetary value that orthopaedic surgeons received. 

Comment: This article indicates that orthopaedic companies support publications by "thought leaders". It does not discuss why industry might do this and what effect this research support may have on the type of research or the conclusions of this research conducted by these investigators.  For example, is industry more likely to support (a) investigation of a non-operative approach to managing chronic cuff tears or (b) investigation of a new cuff repair technique using multiple suture anchors made by the supporting company? As another example, is industry likely to continue much desired research and personal support for a surgeon (a) if the surgeon's research reveals a relatively high complication rate for a new prosthesis or (b) if the surgeon's research reveals that osseous ingrowth by CT scan is better for this new component? What effect might this have on the direction of the surgeon's research?

We recognize the value of industrial support for orthopaedic research. We also recognize that

We're talking real money here (see this link): data from 2016

and see this link (link), where two of the top three recipients are orthopaedic surgeons

As discussed in Demographics of Disclosure of Conflicts of Interest at the AAOS, industrial payments to physicians may also affect the work presented at national meetings.

This article might well have included a discussion of these important aspects of the industry-investigator relationship. 

=====

The reader may also be interested in these posts:

Healing through joint replacement

Supporting progress in shoulder surgery

Consultation for those who live a distance away from Seattle.

Click here to see the new Shoulder Arthritis Book.

Click here to see the new Rotator Cuff Book

Information about shoulder exercises can be found at this link.

Use the "Search" box to the right to find other topics of interest to you.

You may be interested in some of our most visited web pages including:shoulder arthritis, total shoulder, ream and run, reverse total shoulder, CTA arthroplasty, and rotator cuff surgery as well as the 'ream and run essentials'

See from which cities our patients come.

See the countries from which our readers come on this post.

Agenda for shoulder research

We had the opportunity to review the abstracts from the recent meeting of the American Shoulder and Elbow Surgeons. Since this was a closed meeting at which unpublished work is presented, it would be premature to discuss the papers presented. However, a careful review of these papers strongly suggested a research agenda for all interested in improving the care we offer individuals disabled by shoulder arthritis and rotator cuff pathology. We hope that the list below will stimulate meaningful investigation.

* What are the best predictors of clinical success in the non-operative and surgical management of a chronic symptomatic rotator cuff defect: patient factors (age, BMI, activity, employment status, insurance coverage, smoking status, emotional health, physical health, ASA classification, expectations of the outcome) shoulder factors (tear size, chronicity, comfort, function, strength, fatty infiltration, prior surgery), treatment factors (exercises, injections, non-repair surgery, repair surgery, repair method, post surgery rehabilitation) and resulting rotator cuff integrity.

*What are the best predictors of clinical success in surgical management of glenohumeral arthritis: patient factors (age, BMI, activity, employment status, insurance coverage, smoking status, emotional health, physical health, ASA classification, expectations of the outcome) shoulder factors (diagnosis, radiographic anatomy, rotator cuff status, chronicity, comfort, function, strength, prior surgery), and treatment factors (prosthesis type, surgical technique, post surgery rehabilitation).

*Why do we not have a shoulder arthroplasty registry in the United States that would enable us to analyze the outcomes of this procedure for all surgeons and all patients, rather than having our knowledge confined to what we can glean from case reports from individual centers that represent a highly selected and non-representative subset of the national experience? Why do we not require that implant manufacturers support such an effort through a mandatory levy as they do in some other countries?

*In that each year sees a new ‘crop’ of ‘innovative’ new shoulder implants, how can we assess the value added (=incremental benefit to the patient/incremental cost) of new implants?

*What is the optimal strategy for detecting bacteria in the wounds of shoulders coming for revision arthroplasty, recognizing that Propionibacterium and coagulase negative staph do not incite the inflammatory response traditionally associated with perprosthetic hip and knee arthroplasty infections? What are are the pathogenetic and therapeutic implications of positive deep cultures for these organisms?

----

If you have suggestions for topics you'd like us to address in this blog, please send an email to shoulderarthritis@uw.edu.

Use the "Search the Blog" box to the right to find other topics of interest to you. 

You may be interested in some of our most visited web pages including:shoulder arthritis, total shoulder, ream and run, reverse total shoulder, CTA arthroplasty, and rotator cuff surgery.

See the countries from which our readers come on this post.

Ream and Run for Shoulder Arthritis - patients under 55 years - research foundation 7

It is well known that patients with shoulder arthritis who are 55 years of age or younger have special challenges related to their activity levels, their expected longevity, and their more complex forms of arthritis. Shoulder fellows Saltzman and Mercer investigated our results with the ream and run procedure in patients in this younger age group. Among 65 shoulders, nine required revision surgery. These had had an average of 3 surgeries on their shoulder prior to the ream and run, in contrast to the 56 unrevised shoulders which had had an average of 1 prior surgery. For the 56 unrevised surgery the Simple Shoulder Test functions that could be performed improved from a mean of 4 before surgery to a mean of 9.5 at an average of 43 months after surgery (p<.001). The results are shown in more detail below.

--

Use the "Search the Blog" box to the right to find other topics of interest to you.

You may be interested in some of our most visited web pages including: shoulder arthritis, total shoulder, ream and run, reverse total shoulder, CTA arthroplasty, and rotator cuff surgery.

See the countries from which our readers come on this post.

Ream and Run for Shoulder Arthritis - comparison to total shoulder - research foundation 6

We are often asked to compare the ream and run procedure to the total shoulder in the treatment of shoulder arthritis. Shoulder fellows Clinton, Franta and Lenters matched 35 consecutive patients for whom we had performed the ream and run to a similar group of patients for whom we had performed a total shoulder; they presented the results in an article comparing these two procedures. The characteristics of the two groups are shown in the table below.

The time course for recovery is shown in the table and figure below.

We concluded that in the hands of a surgeon experienced in the procedure, the ream and run procedure can offer similar functional recovery to that achieved in similar patients having total shoulder arthroplasty.

--

Use the "Search the Blog" box to the right to find other topics of interest to you.

You may be interested in some of our most visited web pages including: shoulder arthritis, total shoulder, ream and run, reverse total shoulder, CTA arthroplasty, and rotator cuff surgery.

See the countries from which our readers come on this post.

Ream and Run for Shoulder Arthritis - is wear an issue? - research foundation 5

We know that shoulder arthritis affects both sides of the joint as shown below.

 This alters the normal load distribution (below left) to one of load concentration and progressive wear (below right)

Simply replacing the humeral head (ball of the shoulder joint) alone, does not address the glenoid wear and can be associated with progressive glenoid wear.

The ream and run procedure restores the glenoid surface to a concentric concavity.

We wanted to know if there was wear of the glenoid bone after this procedure. Shoulder fellows Mercer and Saltzman developed a method for documenting the position of the humeral head center relative to the scapula on standardized plain radiographs that can answer this question. In this method, standardized templates (shown below) are placed over the x-rays to track the position of the center of the head of the humerus.

They used this method in assessing glenoid wear at a minimum of two years after hemiarthroplasty with concentric glenoid reaming. They found that the average wear rate was minimal: the movement of the head center towards the scapula was less than 0.4 mm per year.

--

Use the "Search the Blog" box to the right to find other topics of interest to you.

You may be interested in some of our most visited web pages including: shoulder arthritis, total shoulder, ream and run, reverse total shoulder, CTA arthroplasty, and rotator cuff surgery.

See the countries from which our readers come on this post.

Ream and Run for Shoulder Arthritis - clinical series - research foundation 4

Our first major publication on the ream and run procedure was co authored by shoulder fellows Lynch, Franta, and Lenters along with resident Montgomery; a link to this article, "Self Assessed Outcome Two to Four Years after Shoulder Hemiarthroplasty with Concentric Glenoid Reaming" can be found here. Our patients assessed their own comfort and function before and sequentially after the ream and run using the Simple Shoulder Test. The patients ranged in age from 35 to 80 years of age.  Their scores improved from 4.7 to 9.4 out of a possible 12. These results were very similar to the results from our series of total shoulder joint replacements as shown in Table II of the article, shown below.

Using the ream and run, we were able to able treat shoulders with severe posterior erosion of the glenoid bone as shown below

and to center the prosthetic humeral head on a reconfigured and regenerated joint surface as shown below.

--

Use the "Search the Blog" box to the right to find other topics of interest to you.

You may be interested in some of our most visited web pages including: shoulder arthritis, total shoulder, ream and run, reverse total shoulder, CTA arthroplasty, and rotator cuff surgery.

See the countries from which our readers come on this post.

Ream and Run for Shoulder Arthritis - initial clinical results - research foundation 3

Based on the encouraging laboratory studies demonstrated in the previous posts we began offering the ream and run procedure to selected and informed patients with arthritis of the shoulder who wished to avoid the risk of glenoid component failure and the activity limitations recommended for total shoulder replacement. In this procedure, we expose the arthritic glenoid (shown below),

remove the damaged cartilage, and then ream the bone to a spherical concavity (shown below) before replacing the humeral side of the joint using the technique explained in our April 6 post.

We were truly excited to see that in many cases, after completely removing the cartilage from the surface of the glenoid, new soft tissue regenerated to cover the reamed bony surface as shown by the dark space between the metal ball and the bone of the glenoid shown on x-rays taken a year after surgery. In the x-rays shown below, this regenerative zone is marked with a red "R".

--

Use the "Search the Blog" box to the right to find other topics of interest to you.

You may be interested in some of our most visited web pages including: shoulder arthritis, total shoulder, ream and run, reverse total shoulder, CTA arthroplasty, and rotator cuff surgery.

See the countries from which our readers come on this post.

Ream and Run for Shoulder Arthritis - healing of the reamed bone - research foundation 2

Our May 16 post illustrated some long-forgotten work on the hip, showing that after reaming the bone of the hip socket, the surface of the bone could heal over with a smooth layer of soft tissue. My colleagues John Clark and John Sidles along with Kristi Gibbs and Tony Norman performed an in vivo lab study of the healing response of the canine socket in response to glenoid reaming followed by a period of articulation with a metal humeral implant. The full text of the article can be seen here. The normal canine socket in cross section is shown below. Note the reddish staining cartilage on the surface and the blue bone beneath.

The surface of the glenoid was then reamed using a specially designed reamer, shown below.

The smooth reamed surface is shown below.

A cross section after reaming is shown below. Note that the glenoid has a new contour different than that of the normal socket and that all of the cartilage has been removed. The many small bone plates (trabeculae) that previously supported the surface cartilage have been fractured by the reaming. We suspect that these many small fractures along with the accompanying bleeding bring stem cells and growth factors to the reamed surface to instigate the healing response.

The humeral head (ball side of the shoulder's ball and socket joint), was replaced with a special humeral prosthesis.

This is shown in place on a post-operative x-ray.

After 10 weeks of use, the cross-section of the glenoid shows substantial, but incomplete healing of the reamed surface by red-staining soft tissue.

On closer look, buds of regenerating soft tissue (red) can be seen pushing up from the reamed bony surface (blue-green) indicated by the arrow. The black dots in the soft tissue represent living cells. It is of note that this regenerative surface is powerful, pushing up against the resistance of the metal ball articulating with the reamed glenoid surface. This reminds me of the power of tree roots growing beneath the sidewalk.

At 26 weeks, a cross section of the glenoid shows complete healing with red soft tissue covering the reamed glenoid surface.

The regenerated soft tissue is firmly bonded to the underlying bone by arching strands of fibrous tissue. This type of fixation could not be accomplished with a plastic socket or by interposing a graft of capsule or meniscus between the humeral head and glenoid. Note also that the red soft tissue contains living cartilagenous cells, confirming that fibrocartilage has regenerated on the surface.

While this surface covering is not the same as the hyaline cartilage that covers normal joints, it appears to provide a smooth and durable covering for bone articulating with a metal ball. In contrast to an artificial surface that deteriorates over time, this living, smooth joint surface firmly bonded to the underlying bone has the potential to remodel continuously with use.

--

Use the "Search the Blog" box to the right to find other topics of interest to you.

You may be interested in some of our most visited web pages including: shoulder arthritis, total shoulder, ream and run, reverse total shoulder, CTA arthroplasty, and rotator cuff surgery.

See the countries from which our readers come on this post.

Ream and Run for Shoulder Arthritis - shaping the arthritic socket - research foundation 1

As the idea of the ream and run procedure progressed towards clinical application, shoulder fellows Weldon and Boorman carried out some important laboratory research on Optimizing the Glenoid Contribution to the Stability of a Humeral Hemiarthroplasty. In this study on cadaver shoulders they found that the glenoid contribution to shoulder stability (as measured by the balance stability angle) was decreased by the removal of cartilage and labrum and was restored by spherical reaming to a level similar to resurfacing the glenoid with a polyethylene component. They studied the stability of four different glenoid states:

Interestingly, they found that removal of the cartilage from the glenoid surface dramatically diminished the stability provided by the glenoid socket - especially in the posterior inferior direction (compare 'native' to 'denuded' on the figure below - the 225 degree direction is posterior inferior). This helps understand why the humeral head often starts to slip posteriorly in arthritis as shown in previous posts.

When the face of the glenoid is spherically reamed, as we do in performing a ream and run, the stability is restored - even beyond that of the normal glenoid (see 'reamed 22.5' and 'reamed 25.0' on the figure above. As the table below shows, the stability of the reamed glenoid is comparable to that of a polyethylene glenoid component.

--

Use the "Search the Blog" box to the right to find other topics of interest to you.

You may be interested in some of our most visited web pages including: shoulder arthritis, total shoulder, ream and run, reverse total shoulder, CTA arthroplasty, and rotator cuff surgery.

See the countries from which our readers come on this post.