In the previous posting, I wrote with some pleasure about the frequency with which some of HealthBridge's papers have been cited. But then shortly after that posting, I came across a reference to our 2004 paper in Public Health Nutrition that completely misrepresented the article! So I reviewed all 51 citations that Google Scholar identifies to see how the paper is actually being cited.

First of all, I learned that Google Scholar does some double counting and there have been 48 citations, of which about half have been in primary scientific literature and the remainder in grey literature and books. A few papers were unavailable or inaccessible, leaving 40 papers for analysis. I looked in each of those 40 papers to see how our paper was cited, and summarized the findings.

*************************************

Appropriate Citations: 33

Citing Major Points of our Paper

· Agriculture interventions more likely to have a positve health impact if they are broad-based and invest in multiple types of "capital" (human, social, physical, financial and environmental). – 10 citations

· Agriculture interventions more likely to have a positve health impact if they include nutrition education - 7 citations - or, more generally, invest in human capital - 1 citation.

Citing Secondary Points of our Paper

· Agriculture interventions often increase food production without improving nutriiton - 3 citations - or had mixed results - 2 citations.

· Agriculture-based interventions impact on nutrition have not been well studied (few studies, weak design) - 3 citations.

· Agriculture interventions more likely to have a positve health impact if they include gender issues - 1 citation.

· Home gardening interventions more often had positive health impacts then other types of interventions - 4 citations

Two other papers used the citations list of our paper to identify relevant studies for a review they were writing.

Inappropriate Citations: 7

Drawing Tangential or Inferred Points from our Paper

· There were five citations which would have been better off using other references. The point that the authors wanted to support was only indirectly or weakly supported in our paper.

Misrepresentation of the Conclusions of our Paper:

· The benefits of investing in the five different types of "capital" is misrepresented in one paper as the benefits of investment in infrastructure - "internet services, computing facilities and other research materials and equipments". This author certainly didn't read our paper, and probably did not even read the abstract.

· The authors state that, while they didn't measure nutrition status, "previous studies indicate that changes in nutritional intake from vegetable gardens in the developing world can have significant impact...". But their interventions was narrowly focused on agriculture production, and therefore unlikely to have a positive impact. They used our paper to support their intervention, when it actually criticizes their type of intervention.

*************************************

That is surprsing to me, and a little humbling, but as I look around, I see it isn’t unusual. An excerpt from an abstract in JAMA is quoted below:

“Fifty randomly selected references from a single monthly issue of The American Journal of Surgery; Surgery, Gynecology and Obstetrics; and Surgery were evaluated for citation and quotation errors. Thirteen major and 41 minor citation errors were found in the three journals. Thirty-seven major quotation errors were identified. The data support the hypothesis that authors do not check their references or may not even read them.” (JAMA. 1990;263:1353-1354)

I believe the overall impact of our paper has been positive. I know individuals who have been positively influenced by it, and the field is steadily and surely tilting in support of what we wrote in the paper. The misrepresentations on the other hand… well, they could have found some other paper to support their claims, and I don’t feel any shared guilt for their error or worry about its consequences. It is interesting to consider if we have made similar errors in our writing (probably) and if the consequences of inappropriate citations are trivial or amusing (as in Steven Jay Gould’s essay The Case of the Creeping Fox Terrier Clone) or important and damaging as in… I can’t think of any inappropriate citations that have had serious scientific consequences, can you?