Can social isolation fuel epidemics?

From the 20 July 2015 Santa Fe Institute news release


Conventional wisdom has it that the more people stay within their own social groups and avoid others, the less likely it is small disease outbreaks turn into full-blown epidemics. But the conventional wisdom is wrong, according to two SFI researchers, and the consequences could reach far beyond epidemiology.

In a paper published in the July 20 early edition of the Proceedings of the National Academy of Sciences, Laurent Hébert-Dufresne and Benjamin Althouse show that when two separate diseases interact with each other, a population clustered into relatively isolated groups can lead to epidemics that spread like wildfire.

“We thought we understood how clustering works,” Hébert-Dufresne says,”but it behaves exactly opposite to what we thought once interactions are added in. Our intuition was totally wrong.”

At the heart of the new study are two effects that have had a lot of attention in recent years—social clustering and coinfection, in which one disease can change the infection dynamics of another—but haven’t been studied together. That, Hébert-Dufresne and Althouse say, turns out to be a major omission

Ordinarily, the pair say, clustering limits outbreaks. Maybe kids in one preschool get sick, for example, but since those kids don’t see kids from other preschools as often, they’re not likely to spread the disease very far. Coinfection often works the other way. Once someone is sick with, say, pneumococcal pneumonia, they’re more likely than others to come down with the flu, lowering the bar for an epidemic of both diseases.

But put the effects together, Hébert-Dufresne and Althouse discovered, and you get something that is more—and different—than the sum of its parts. While clustering works to prevent single-disease epidemics, interactions between diseases like pneumonia and the flu help keep each other going within a social group long enough that one of them can break out into other clusters, becoming a foothold for the other—or perhaps a spark in a dry forest. Both diseases, Althouse says, “can catch fire.” The end result is a larger, more rapidly developing, epidemic than would otherwise be possible.

That conclusion has immediate consequences for public health officials, whose worst-case scenarios might be different or even tame compared with the outbreaks Hébert-Dufresne and Althouse hypothesize. But there are equally important consequences for network scientists and complex systems researchers, who often think in epidemiological terms. Two ideas, for example, might interact with each other so that both spread more rapidly than they would on their own, just as diseases do.

Now that they’ve realized the importance of such interactions, “we hope to take this work in new and different directions in epidemiology, social science, and the study of dynamic networks,” Althouse says. “There’s great potential.”

Read the early edition paper in the Proceedings of the National Academy of Sciences (July 20, 2015)

Share on Facebook

Teaching Reflection to Doctors to Improve Physician-Patient Interactions

From the 10 July 2015 post at Tufts Now

Physicians in their medical residency training programs often focus on scientific reasoning and research evidence in their efforts to provide medical care. While appropriate, this focus may overshadow subtle and indirect communication that reveals important information about the patient’s experience with their illness that will help the physician provide better care. A new study by researchers at Tufts University School of Medicine and Boston College presents the results of a strategy to train medical residents to reflect on interactions with patients as a way of understanding the meaning of both their patient’s, and their own, communication.

The study directors asked 33 family medicine residents in the Tufts University Family Medicine Residency program at Cambridge Health Alliance to write “open-ended reflections” over the course of one year examining their interactions with patients. The project, which used a qualitative research design, resulted in 756 private reflections that the research team iteratively organized into three principal communication themes: (1) recognizing the interdependence of physician-patient communication (2) attention to the subtleties of patient behavior; and (3) images of growth and awareness about physician-patient communication.

In the report in the Journal of Health Communication published this month, the authors provide sample entries from residents on each of the themes and related sub-themes. On the theme of interdependence of the communication behavior, which included sub-themes on how physicians restrict what patients will tell them; how learning and taking the patient’s perspective can help, and how better communication might promote behavior change, sample entries included...

Share on Facebook

Reduced Mortality Risks and Correlation vs. Causation

From the 22 July 2015 NCCIH research blog item by Director Josephine P. Briggs

Recently, I noticed news articles about a study on chamomile consumption and its potential effect on mortality. Researchers from the University of Texas Medical Branch, Galveston, analyzed data from a population-based study of older Mexican Americans in Southwestern states and reviewed 7-year all-cause and cause-specific mortality. They found a 29 percent decreased risk of death (all-cause mortality) among chamomile users compared with non-users, and concluded, for women at least, that this difference was statistically significant. It’s good news that using this herb is associated with longer life. Nevertheless, let’s remember that, even with statistical corrections, correlations don’t prove causality.

The authors reported that the decreased risk was statistically significant for women after adjusting for age, smoking, chronic conditions, and other known confounding factors. This is the right approach to the data. But it is not enough. Statistically correcting for other factors only captures the impact of measured differences. The women who use chamomile may differ in many ways from those who do not. For example, the use of herbal tea may be a marker of a “healthier” lifestyle. In their paper, the researchers note that “other unmeasured factors, such as frequency and duration of chamomile, level of physical activity, and quality of diet, which were not measured in the survey, could influence the results.”

This study illustrates the challenges of observational (non-experimental) research. This type of research helps us find patterns and signals and can raise new and interesting hypotheses. But let’s remember that there may be a big gap between correlation and causation.



Howrey BT, Peek MK, McKee JM, et al. Chamomile consumption and mortality: a prrospective study of Mexican origin older adults. Gerontologist. April 29, 2015. Epub ahead of print.

Share on Facebook

Finding and Using Health Statistics – A Tutorial

Finding and Using Health Statistics – A Tutorial

FindingStatsFinding and using health statistics has become requisite for a number of careers in the past several decades. It’s also a worthwhile skill for anyone navigating the increasingly complex world of health care and medicine. This free online course from the U.S. National Library of Medicine is divided into three related parts: About Health Statistics, Finding Health Statistics, and Supporting Material. Selecting any of these tabs opens to a table of contents. From there, readers can follow the course page by page. For instance, About Health Statistics begins by reviewing the importance of health stats, moves on to their uses, and then speaks about sources for the gathering of statistics, such as population surveys and registers of diseases. [CNH]

From The Scout Report, Copyright Internet Scout 1994-2015.

Share on Facebook

3 Questions: Carlo Ratti on big data and health predictions [News item] ]

MIT researcher discusses a new study on correlations among medical problems.

From the 7 July 2015 MIT news item

What can big data tell us about the predictability of medical conditions? A new study by MIT researchers published in the journal Scientific Reports  digs into this question by looking at anonymous data from over 500,000 patients. Among the findings is that our electronic medical records contain data that is up to 90 percent predictable — although this level of predictability is only attainable in theory. However, it can guide algorithmic designers and practitioners on what is possible in principle. The co-authors of the paper are Carlo Ratti, director of MIT’s Senseable City Laboratory, and two former computer science researchers at the lab, Dominik Dahlem (who is the lead author) and Diego Maniloff. The data originated with General Electric, which collaborated with Senseable City on a 2011 project on visually plotting health care data. MIT News spoke with Ratti about the new study.

Share on Facebook

Frontiers in Psychiatry [Resource of the Week]


Frontiers in Psychiatry is an academic accomplishment: a high quality, peer-reviewed, open access journal publishing the most outstanding discoveries in the world of psychiatry. Here readers may peruse cutting edge articles in 15 different sections, including Addictive Disorders and Behavioral Dyscontrol, Eating Behavior, Molecular Psychiatry, Neuropharmacology, and others. Over 700 full-length articles make up the well-stocked Archive, including recent publications on a community-based health program for abused children in Brazil, breakthroughs in understanding Tourette Syndrome, and video games for mental health and well-being. Searching for topics of interest is easy and detailed. For instance, “bipolar” returns 22 Articles, 100 Authors, and 11 Research Topics. [CNH]

From The Scout Report, Copyright Internet Scout 1994-2015.

Share on Facebook

Development of a new model to engage patients and clinicians in setting research priorities

From the article at  Journal of Health Services Research & Policy (J HEALTH SERV RES POLICY), 2014 Jan; 19 (1): 12-8. 

Plain English summary
There is some evidence that there is a mismatch between what patients and health professionals want to see researched and the research that is actually done. The James Lind Alliance (JLA) research Priority Setting Partnerships (PSPs) were created to address this mismatch. Between 2007 and 2014, JLA partnerships of patients, carers and health professionals agreed on important treatment research questions (priorities) in a range of health conditions, such as Type 1 diabetes, eczema and stroke. We were interested in how much these JLA PSP priorities were similar to treatments undergoing evaluation and research over the same time span. We identified the treatments described in all the JLA PSP research priority lists and compared these to the treatments described in a group of research studies (randomly selected) registered publically. The priorities identified by JLA PSPs emphasised the importance of non-drug treatment research, compared to the research actually being done over the same time period, which mostly involved evaluations of drugs. These findings suggest that the research community should make greater efforts to address issues of importance to users of research, such as patients and healthcare professionals.

Read the entire article here


Share on Facebook

The problem with P values: defining clinical vs. statistical significance [Reblog]

From the 24 June 2015 Public Library of Science (PLos) blog postcrowdfunding-statistics-2Today we warmly welcome guest writer Sean Sinden to PLOS Public Health Perspectives. His biography is at the end of the post.

The practice of null hypothesis testing has traditionally been used to interpret the results of studies in a wide variety of scientific fields. Briefly, significance testing involves the calculation of an outcome statistic, known as the P value. The P value represents the probability of finding a difference, by chance, between two sets of values larger than that which was observed, assuming no difference between the two sets of values. Conventionally, if that probability is less than 0.05 the outcome is deemed “statistically significant”. If this sounds confusing, it’s because it is!

P values are commonly misinterpreted and misused to answer research questions, but in actuality they fail to provide much information to the reader (1). This method of statistical analysis has been met with criticism throughout its history and increasingly so in the past few decades. The shortcomings of significance testing have been identified in both the academic and non-academic literature. I encourage anyone interested in the mechanistic limitations of significance testing to seek out such resources.

The P value doesn’t begin to explain the importance of a study’s outcome or the amount of the effect observed, though many researchers mistakenly believe it to.

Share on Facebook

Mulford Renovations Start Monday, June 22

This summer, the Mulford Library is undergoing renovations (and we’re not talking about the work on the Student Services area on the first floor!). Renovations to the library part of the Mulford Library Building will start on Monday, June 22, to start the construction part of renovations. Contractors expect a 6-week work period (though the end of July). New furniture will arrive later this summer or early fall.

Renovations are based on student recommendations and needs. They include:

  •  reconfiguring small group study space on the fifth floor, eliminating the central study area and making two of the side rooms larger. These rooms will get soundproofing, new furniture, new electrical, and whiteboards.
  • creating a new small group study space outside of room 417, by removing the two center stacks to make space for small tables; it will also get new carpeting and lighting to match the lounge outside of room 411.
  • adding electrical outlets, new furniture, and new lighting on the main stair landing between the third and fourth floors
  • adding electrical outlets to the sixth floor southwest, so that all study carrels in that area have electricity
  • addition of two standing desks on the fifth floor
  • replacing window treatments in 408
  • new furniture in room 520
  • new conference room in 517
  • replacement of soft seating
  • additional whiteboards in study rooms and study carrels

We apologized in advance for any disruptions due to noise and construction. Any questions can be directed to Jolene Miller,

Share on Facebook

A sight for sore eyes: Visually training medical students to better identify melanomas [News release]

From the 27 May 2015 University of Alberta news release
UAlberta researchers hope by improving training methods for health professionals, patients will enjoy better health outcomes.  By Ross Neitz.

Each year, thousands of Canadians are given the news: they have skin cancer. It is the most common form of cancer in Canada and around the world, but if detected early, survival rates are extremely high. According to Liam Rourke, early detection doesn’t happen nearly as often as it could.“The difficulty is that people have a really hard time detecting skin cancer melanomas early,” says Rourke, an associate professor in the Faculty of Medicine & Dentistry’s Department of Medicine at the University of Alberta. “One of the reasons is that it’s an exceptionally difficult task because the melanomas look exactly like freckles and moles and other common benign lesions that everybody has.“The ability to detect these things early is pivotal and so we wanted to look at how people are trained to detect these things.”Rourke is the lead author of a study in JAMA Dermatology that examines best practices in teaching medical students and health professionals how to detect, categorize and identify skin lesions. By conducting a meta-analysis other studies in the field, his team found that many traditional methods of teaching focus on what some would argue are the least important aspects of the job.“The conventional sense of how this should be taught is by giving people factual knowledge about skin and its normal and abnormal development. We were surprised to see that there was very little effort to stress the visual part of this task. There was very little organized and systematic effort to train people’s visual systems to discriminate between two things that look a lot alike,” says Rourke.The team of researchers concluded there’s much room for improvement in how health professionals are trained. Based on the results of the study, Rourke’s team is developing perceptual training modules they hope to test through further studies.According to Rourke, an initial pilot study has already been undertaken that has shown promising early results. In it, a small group of undergraduate students with no medical knowledge, training or vocabulary were given iPads containing a module showing different types of skin lesions. Over the course of about two hours the students were presented with several examples of melanomas and non-melanomas. At first the examples were labeled as such, followed by the pictures again being shown without labelling. The students were then asked to identify the examples correctly. As the students progressed through the module, the examples became more varied and complex.“We saw an enormous effect,” says Rourke. “Within the two to three hours of training they reached a level of expertise that was close to the level of a dermatology resident.”He adds, “The training translates well to other groups because it doesn’t rely on learning a lot of anatomy or physiology or biomedical knowledge about skin lesions. The task is simply to identify skin cancers and differentiate them from benign lesions.”


Related resources

  • Visual Dx
    The VisualDx decision support and reference tool for physicians includes 1,300+ diagnoses and 29,000+ medical images to aid diagnosis and therapeutic decisions.
  • Dynamed (Dermatological Disorders)


Share on Facebook