The document summarizes two studies conducted by the author. The first was writing blog posts for "The Public's Health" where the author learned that factual accuracy is essential for health communication and that the editing process is lengthy and ensures the work represents the publisher. The second was a content analysis of New York Times health articles in March 2014. It found that only 39.6% of articles linked to scientific studies and only 23.8% of those linked to PubMed. Articles about studies were more likely to link to studies.

1. ABSTRACT
Writing about Health for the Public
Contributions to “The Public’s Health” and
a Content Analysis of the New York Times’ Health section
Teagan Keating1, Jonathan Purtle, MPH1
1Drexel University School of Public Health
I worked as a contributing writer for “The Public’s Health,” the public health blog on Philly.com. Through the writing
and editing process, I learned two important lessons about maintaining the reputation of the blog and the
newspaper. First, factual accuracy and precision are essential because there is a wide audience for the finished
product and ensuring that information is correct is paramount to effective health communication. Second, the
editing process can be slow, time-consuming, and sometimes challenging because the finished product represents
not only the writer but everyone associated with it, polishing and re-polishing the post is necessary. I also provide
recommendations for future internships with “The Public’s Health” (TPH).
I conducted a content analysis of the New York Times Health section for March 2014 to explore the practices of
linking to health-related studies indexed in PubMed and in scientific journals. I found that only 39.6% of articles
linked to scientific studies. Although the best practice is to link to PubMed, only 23.8% did so. Of articles about
studies, 79.4% link to studies. There was little difference between linking practices between blog posts and non-blog
posts. Some authors linked more frequently to studies, showing authors have a degree of autonomy in choosing
their own hyperlinks at the NYT.
PARTI
PARTII
IN THE BEGINNING
TPH is the public health blog for philly.com and is the only strictly public health blog associated with a major metropolitan
newspaper in the U.S. I met with Jonathan Purtle and Michael Yudell to kick around ideas about how writing for TPH could
be my master’s project. We came up with a few options but kept the particulars flexible so we could see how the project
unfolded.
Initially, I was drawn to the idea of having a well-read newspaper publish my ideas. Writing for TPH would give me
experience with the writing, editing, and publishing of online content for an audience for which I had not yet written. I also
knew that one of the best ways for me to really learn something is to write about it. I was also getting the sense that I
would be happiest in my career if I was able to talk with the people at the top of those fields and explain what they’re doing
to an audience that wants to know what is happening but does not have the technical training to interpret the findings.
Initially, I had high hopes for my output. In my learning agreement, I declared I would produce three posts per month.
Jonathan encouraged me to change it to two per month, knowing that the editing process would take more time than I
expected.
THE REALITY
I sent a first draft to Jonathan Purtle.
Jonathan sent back his notes.
I sent Jonathan an updated draft.
He sent back more notes.
We did this up to seven times.
Once Jonathan signed off on it,
I sent the draft to Michael Yudell
for his notes.
Once Michael gave his okay,
I sent the the post to
Don Sapatkin.
Then, Don gave me notes and
published the post.
LESSONS LEARNED
Factual accuracy and precision are essential to effective health
communication. This seems simple—find the truth and tell it. I though I
knew how to conduct research and how to properly relay the information
I found. However, I had not written for the public before and I did not
consider the level of scrutiny my writing would be under from both
editors and the public. The kinds of details that, in a piece published in a
less prominent blog, would go unnoticed became sticking points for
moving forward in the writing process. This made the editing process
slow, time-consuming, and sometimes challenging. Because the finished
product represented not only myself but everyone associated with TPH,
polishing and re-polishing each post was necessary.
While I spent much of the year confused and frustrated with the slow
process and the seemingly excessive attention to detail, I realize now
how important this experience is for me professionally. Certainly, the
lesson of letting go of my personal attachment to a particular idea or
phrase will be essential. If I continue to work in communications in the
public health sector, nothing I produce will be entirely my own words—
everything I write will representative of something larger than myself. I
will have to accept that the voice of “the powers that be” will often be
louder than my own, even in work that is published with my name. That
is the nature of producing documents for large organizations with
important missions—there are many stakeholders, and everyone gets a
say.
RECOMMENDATIONS FOR NEXT YEAR
If TPH has interns next year, I recommend setting up the internship like
an independent study, particularly if there are a few students working
with TPH. Because online writing, particularly for blogs, is a genre with
which most students will be entirely unfamiliar, taking plenty of time to
learn about the style and mechanics is essential to student success.
Next year, interns should focus on:
u Meeting regularly with TPH leaders
u Learning the theory and basics of online writing
u Workshopping their posts
I hope that next year’s interns can learn from my experience and
understand that working with TPH will be about the process than the
product.
INTRODUCTION
Nearly three-quarters of Americans seek health information online (Fox &
Duggan, 2013). Most of these health information seekers have found that
information they need is behind a paywall, and only 2% choose to pay for
the information, while 83% try to find it somewhere else (Fox & Duggan,
2013). By analyzing at the hyperlinking patterns in the New York
Times’ (NYT) Health section in March 2014, this study explores two
related issues: health information literacy and the importance of linking to
open access studies. Information literacy is a set of skills that individuals
use to determine when more information is needed and how to acquire,
understand, evaluate, and apply that information (Association of College
and Research Libraries, 2000). Health information literacy, then, is the
application of information literacy skills to health information.
At the same time, there are 23.6 million articles indexed by PubMed, a
database of citations and abstracts for over 20 million scientific journal
articles (Dunn, Coiera, & Mandl, 2014). PubMed Central, the free archive
of biomedical research maintained by the National Institutes of Health’s
National Library of Medicine, provides open access to three million
articles (US National Library of Medicine, 2011). Most of the PubMed
Central articles have an entry in the PubMed database (US National
Library of Medicine, 2014). Articles written as a result of NIH funded
research are required to be submitted to PubMed Central.
METHODS
I used content analysis to examine health articles posted online in NYT
during the period of March 1-31, 2014. I chose NYT because it has the
largest number of online subscribers in the United States (Alliance for
Audited Media, 2013) and because it is generally seen as the standard
bearer for news journalism. Using the search function on the NYT home
page, I searched the term “health.” On the results page, I was able to limit
the search by date (March 1-31, 2014) and by section, looking only at
articles that ran in the Health section. I looked at articles by all authors.
Articles were classified in one of sixteen content categories (see Table 1).
Content categories were not mutually exclusive. Articles were classified
by date and author, as well as by type of article (e.g. health study, health
article, recipes for health, etc). I also determined whether the article was
part of an NYT blog.
In order to determine if the article linked to any scientific studies I opened
all hyperlinks in the body of the article. If the hyperlink did lead to a study,
I determined whether the article was indexed in PubMed or in the website
for the journal in which it was published.
Through this simple yet targeted search, I was able to examine health
articles published in March 2014 and determine their subjects, and their
linking patterns.
RESULTS
Articles
that link to
a study
40%
Articles that
do not link to
a study
60%
Proportion of Articles
Linking to a Study, Total
Articles that
do not link to
a study, 21%
Articles that
link to a
study, 79%
Proportion of Articles about
a Study that Link to a Study
1.84 1.84 1.83
0
0.5
1
1.5
2
Mean number of
articles linked to,
overall
Mean number of
articles linked to,
blogs
Mean number of
articles linked to,
non-blogs
Mean Number of Articles
Linked to, by Article Type
Table 1: Articles linking to studies, by content category
Content
Category
Number of
articles that link
to studies
Total number of
articles
Proportion of
articles that link
to studies
Aging 7 12 58.40%
Cancer 1 6 16.70%
Cardiovascular
health
8 9 88.90%
Chronic disease 3 5 60%
Genetics 2 5 40%
Health behavior 15 22 68.20%
Health care 10 15 66.70%
HIV/AIDS 2 6 33.40%
Infectious
disease
2 6 33.40%
Maternal child
health
7 8 87.50%
Obesity 2 2 100%
Other 2 10 20%
Recipes for
health
0 0 0
Reproductive
health
2 5 40%
Social
determinants
1 2 50%
Technology 7 12 58.40%
Table 2: Articles linking to studies, by
author (six most prolific authors)
Author
Articles
that link to
studies
Total
number of
articles
Proportion
of articles
that link to
studies
Martha
Rose
Schulman
0 28 0
Nicholas
Bakalar
11 12 91.70%
Donald G.
McNeil
3 9 33.40%
Gretchen
Reynolds
6 7 85.70%
Anahad
O’Connor
3 5 60%
Paula Span 3 5 60%
DISCUSSIONIf readers want to evaluate scientific evidence for the claims made
in 60% of NYT Health section articles, the readers have to seek that
information out for themselves. This creates another step in the
process of finding, reading, and evaluating health research
outcomes to make informed decisions and ideas about personal
and public health and creates a barrier to health information literacy.
Furthermore, the overall proportion of articles linking to PubMed is
very low (23.81%). For news stories, the best practice is to link to
PubMed. Linking to PubMed means the reader can: a) click on the
embedded link and view the article for free if it is indexed in
PubMed Central, or b) access it via subscription or pay for the
individual article. Linking directly to the journal website is less
desirable because there will likely be no link to the article in
PubMed Central, even if it was funded by federal dollars and is in
fact indexed there. This means that readers may end up either
paying for an article that is available free elsewhere, or trying to find
the same information from another, potentially less reliable source.
Of the 46 total authors, fewer than half (46%) linked to any studies.
This may be due to the difficulty of adding hyperlinks to print stories
published online later or may be a stylistic or research choice on the
authors’ point. Interestingly, there is no difference in the mean
number of links to studies between blog posts and non-blog posts. It
is possible that some authors do not consider hyperlinks to studies
to be an important resource to readers. NYT’s policy on this issue is
not public.
REFERENCES
Alliance for Audited Media (2013). Top 25 U.S. newspapers for March 2013. Retrieved from
http://www.auditedmedia.com/news/research-and-data/top-25-us-newspapers-for-march-2013.aspx.
Association of College and Research Libraries (2000). Information literacy competency standards for higher
education. Retrieved from http://www.ala.org/acrl/sites/ala.org.acrl/files/content/standards/standards.pdf
Dunn, A. G., Coiera, E., & Mandl, K. D. (2014). Is Biblioleaks inevitable? Journal of Medical
Internet Research 16(4). DOI: 10.2196/jmir.3331
Fox, S. & Duggan, M. (2013). Health online 2013. Pew Research Center’s Internet & American
Life Project. Retrieved from http://www.pewinternet.org/files/old-media//Files/Reports/PIP_HealthOnline.pdf
Program Evaluation and Methodology Division (1996). Content analysis – a methodology for
structuring and analyzing written material. General Accounting Office. Retrieved from http://www.gpo.gov/fdsys/pkg/
GAOREPORTS-PEMD-10-3-1/html/GAOREPORTS-PEMD-10-3-1.htm.
National Institutes of Health (2014). NIH Public Access Policy Details. Retrieved from
http://publicaccess.nih.gov/policy.htm.
US National Library of Medicine (2011). PubMed Central Overview. Retrieved from
http://www.ncbi.nlm.nih.gov/pmc/about/intro/.
US National Library of Medicine (2014). PubMed Central FAQs. Retrieved from
http://www.ncbi.nlm.nih.gov/pmc/about/faq/#q2.
Special thanks to Jonathan Purtle, my academic advisor and project preceptor,
for his tireless assistance conceptualizing, revising, and polishing blog posts
and the research study. Thank you to Michael Yudell and Don Sapatkin for
helping me keep my writing readable and precise.
Graphic of the editing process created by Nathan Kuruna.