Investigating, localizing salmonella outbreak

As some of you may have noticed, there’s an egg recall going on. It all began when the CDC’s PulseNet monitoring program noticed a fourfold jump in the number of salmonella cases being reported, which spurred investigations around the country. This jump is evident in the graph below. Don’t be fooled by the dropoff at the end, it has more to do with the reporting process than with an actual decrease in the number of salmonella cases (which clearly isn’t happening).

Image courtesy of the Centers for Disease Control and Prevention

Image courtesy of the Centers for Disease Control and Prevention

Health officials then traced it all back to a man outlets love to describe as a sort of rogue Iowa egg magnate and his Wright Country Eggs (satellite view?).

As we stand now, the tainted eggs could have been distributed through any number of channels, but constitute a tiny fraction of the national egg supply.

For reporters digging into this national recall story, or looking to localize it to their coverage area, AHCJ has a strong archive of foodborne illness resources.

Start with a classic, the AHCJ article “Fatal Food: A study of illness outbreaks ,” in which Thomas Hargrove details SHNS’ massive investigation into the nation’s food safety monitoring system. Not only is Hargrove’s how-to instructive, his actual findings are useful examinations of state and local food safety systems around the country.

For your own investigation, look at Mining NLM databases: PubMed, Medline and more and the rich set of resources in the sidebar to Hargrove’s story.

If you’re looking for solid numbers and the most up-to-date national context, see Covering Health’s recent post on the CDC’s lates foodborne illness data, as well as our examination of 2009 foodborne illness rates.

Other relevant Covering Health posts include:

Schneider: FDA lacks resources to keep food safe
CDC assembles rogues gallery of food bugs
Private food auditors didn’t stop outbreaks
Lax oversight, complex supply chains aid outbreaks

CDC releases 2007 foodborne illness numbers

Aug. 12th, 2010 by Andrew Van Dam · 2 Comments
Filed under: Health data, Hot Health Headline 

In the latest Morbidity and Mortality Weekly Report, the CDC has released the 2007 numbers on foodborne illness in the United States. Norovirus (39 percent) was the most common culprit, followed by Salmonella (27 percent). In terms of illnesses caused, poultry led the way, followed by beef and leafy greens. In the majority of the 1,097 reported outbreaks of foodborne illness, no agent was identified – a fact the CDC attributes to the small scale of many of those outbreaks. Here’s a breakdown of what investigators managed to find:

food2

Those looking to dig a little bit deeper into the numbers should consult this four-page PDF, which breaks it all down by contaminant, food, number of outbreaks and number of illnesses caused.

Resources for covering food safety

Tip Sheets

Websites

Related

CDC: Obesity rises unabated, no state meets goals

obesityAccording to Vital Signs, a new part of the CDC journal Morbidity and Mortality Weekly Report (more on that in a minute), the number of adults whose self-reported numbers indicated obesity rose 1.1 percent between 2007 and 2009. Nationally, 26.7 percent of adults were obese in 2009, a number that’s even higher for non-Hispanic blacks (36.8 percent), Hispanics (30.7 percent) and folks who didn’t graduate from high school (32.9 percent.) [PDF transcript of today's briefing.]

States ranged from Mississippi (34.4 percent) to Colorado (18.6 percent) and none met the federal Healthy People 2010 obesity target of 15 percent.

Healthy People 2010 was started by the HHS in 2000 as an effort to improve public health and eliminate disparities across the country. Obesity percentages were a key benchmarks, as were tobacco use, access to health care, mental health, environmental quality and immunization.

The objectives for Healthy People 2020 are being considered now. They would keep the obesity-related goals from the 2010 effort and augment them with nutrition-based standards.

The CDC says Vital Signs, which will be published on the first Tuesday of each month, “is designed to provide the latest data and information on key health indicators – cancer prevention, obesity, tobacco use, alcohol use, access to health care, HIV/AIDS, motor vehicle passenger safety, health care-association infections, cardiovascular health, teen pregnancy, infant mortality, asthma and food safety.”

FOI request for H1N1 documents still pending

Eleven months after she filed a Freedom of Information Act request with the Centers for Disease Control and Prevention for documents related to H1N1, CBS News’ Sharyl Attkisson reports that she has finally received a response from the U.S. Department of Health and Human Services.

It states that it has a certain document responsive to my FOI request. The letter says HHS will try really hard to provide the document soon, but that the folks there are very busy so it may take awhile. It says that if I want expedited processing - something I had already applied for a year ago - I should let them know. A phone number was provided in case I had any questions.

Of course the investigative report I was working on is long over, as the bureaucrats must have known it would be by now.

Attkisson says she called the number provided in the letter and left a message several weeks ago and has yet to hear back. Her initial request was prompted by the CDC’s decision to stop testing and tracking H1N1, something her sources told her was hasty and more about influencing the public’s perception of the illness than it was about public health.

Earlier: Freedom of Information: Stalled at CDC and D.C. Government (Oct. 27, 2009)

Related

CDC says monitoring system finds no ill from spill

The CDC has two major monitoring programs active in the Gulf of Mexico during the spill: The National Poison Data System and Biosense.

The National Poison Data System tracks calls to American poison centers. As of July 12, it had tracked 1,221 calls regarding the spill, 722 of which regarded exposure to spill-related toxins such as oil, dispersant or food contaminants. The other 499 calls came from folks seeking information about the health effects of the spill. The majority of the calls have come from the gulf states, but some originated from as far away as California, Michigan and Massachusetts.

Biosense is a public health tool that tracks real-time changes in a population’s health status. Among other things, it tracks more than 80 health facilities on the Gulf Coast and provides states affected by the spill with daily updates. According to the latest available data, it has “found no trends in the number of illnesses and injuries that would require further public health investigation.”

In addition to focusing resources of these two national programs, the CDC has collected state public health monitoring resources from Louisiana, Mississippi, Alabama and Florida.

Rafael Olmeda of the South Florida Sun-Sentinel points out that the CDC has posted “Gulf Oil Spill Information for Pregnant Women,” which generally advises everyone to stay away from oil spill affected areas.

Related

Raw, warm vegetables breed illness in salsa, guac

New research implicates guacamole or salsa in 3.9 percent of restaurant-related outbreaks of foodborne illness between 1998 and 2008, more than double the rates of previous measurement periods. Both sauces often combine raw ingredients – tomatoes, peppers and cilantro – that have each been blamed for past outbreaks, the CDC release said.

salsaPhoto by anitasarkeesian via Flickr

Improper storage and temperature were blamed for 30 percent of the outbreaks, and another 20 percent were caused by worker-related contamination. The outbreaks are common enough that the government even gives them their own acronym (SGA!), an honor that’s admittedly not particularly rare in the world of federal bureaucracy.

CDC began conducting surveillance for foodborne disease outbreaks began in 1973, yet no salsa- or guacamole-associated (SGA) outbreaks were reported before 1984. Restaurants and delis were the settings for 84 percent of the 136 SGA outbreaks. SGA outbreaks accounted for 1.5 percent of all food establishment outbreaks from 1984 to 1997. This figure more than doubled to 3.9 percent during the ten-year period from 1998 to 2008.

According to the release, the primary weapon against such outbreaks is simply the awareness that vegetables are a threat.

“Possible reasons salsa and guacamole can pose a risk for foodborne illness is that they may not be refrigerated appropriately and are often made in large batches so even a small amount of contamination can affect many customers,” (Magdalena Kendall, Oak Ridge Institute for Science and Education researcher) says. “Awareness that salsa and guacamole can transmit foodborne illness, particularly in restaurants, is key to preventing future outbreaks.”

How the CDC used consumer mapping tech in Haiti

Writing for Government Health IT, Carolyn Duffy Marsan explores the CDC’s reliance on free and open source mapping software and web applications in the aftermath of the Haiti earthquake. It seems at first almost incongruous, a colossal federal agency working to save lives in the wake of a great disaster using the same Google Maps tools we’d use to plot the shortest route to Costco, but, as Marsan explains, there are fantastic reasons to rely upon free consumer technology in such a situation.

The most obvious, of course, is accessibility. Even in the decentralized chaos that reigned in the hours following the quake, thousands of people on the ground had access to Google Maps, or at least to messaging services that would connect them to people with access to the maps. Likewise, Google Earth’s satellite imagery helped officials in the United States rapidly locate health services, as well as water sources and open areas where displaced locals were likely to congregate, and thus to coordinate first responders and organize on-site data, even from their remote location.

googlemaps

Another advantage, officials told Marsan, was that the data accumulated by American and international agencies in such formats was highly transferable and simple to combine, mash up and eventually hand over to the Haitian government. Furthermore, this data portability has allowed the myriad agencies and individuals involved in the recovery efforts to quickly adopt slightly more specialized tools such as wiki-like OpenStreetMap and the well-known crisis crowdsourcer Ushahidi.

Ushahidi takes data supplied via just about any available format and turns it into something useful through a blend of high technology and human effort.

Ushahidi maps were also used to display field hospitals, pharmacies and medical supplies. Now the group is working with a team of doctors in Boston to create the ability to track the spread of infectious diseases. Ushahidi Haiti has a network of 300 volunteers that help create maps from the thousands of text messages and emails received from the field. Sewell says the data provided by the diaspora has been more accurate than the media in some cases.

In a mapless world like post-quake Haiti, consumer mapping tools helped document the new geography and coordinate efforts, both through OpenStreetMaps, which supplied the first usable maps of the affected areas, and through mobile devices.

“You can’t say, ‘Let’s meet at a particular street,’ because the streets have been destroyed,” (Antonio Zugaldia, an information officer with the Pan American Health Organization’s Emergency Operations Center in Washington DC) said. “Instead, you offer a real-time location using GPS on your BlackBerry and you can share that with your colleagues with Google Latitude. You can then easily divide your team into sub-teams and keep track of them. It’s a very simple tool that provided an excellent service.”

ER visits caused by nonmedical use of opioids double in 5 years

Jun. 18th, 2010 by Andrew Van Dam · Leave a Comment
Filed under: Health data, Hot Health Headline, Tools 

The latest Morbidity and Mortality Weekly Report from the CDC focuses on the rapid increase of emergency department visits caused by the abuse and misuse of prescription painkillers. The report is based on a review of the five most recent years of data from the Drug Abuse Warning Network.

DAWN’s national estimates are based on a 220-hospital sample. According to DAWN, “nonmedical use” means “taking a higher-than-recommended dose, taking a drug prescribed for another person, drug-facilitated assault, or documented misuse or abuse, all of which must be documented in the medical record.”

The big takeaway?

… the estimated number of ED visits for nonmedical use of opioid analgesics increased 111% during 2004-2008 (from 144,600 to 305,900 visits) and increased 29% during 2007–2008. The highest numbers of ED visits were recorded for oxycodone, hydrocodone, and methadone, all of which showed statistically significant increases during the 5-year period.

It’s a number-heavy report, so I’ve put together a quick overview with the help of the DAWN and MMWR reviews, as well as this DAWN report. You’ll find it below.

drugabuse

CDC used flawed data on lead in drinking water

The Washington Post’s Carol Leonnig reports that an investigation by the House Committee on Science and Technology’s Subcommittee on Investigations and Oversight has confirmed what The Washington Post first reported last year, namely:

leadPhoto by blandm via Flickr

The nation’s premier public health agency knowingly used flawed data to claim that high lead levels in the District’s drinking water did not pose a health risk to the public… And, investigators determined, the agency has not publicized more thorough internal research showing that the problem harmed children across the city and continues to endanger thousands of D.C. residents.” Those who need a refresher on the issue can refer to the Post’s timeline and story archive.

The larger issue here is that the committee and the Government Accountability Office are looking into how the CDC’s National Center for Environmental Health/Agency for Toxic Substances and Disease Registry evaluates public health issues.

Subcommittee Chairman Brad Miller (D-NC) had some harsh words for the department:

“We need more honesty and transparency and less attitude from these offices. When you work at a public health science agency and the words most frequently used are ‘haphazard,’ ‘hit-or-miss’ and ‘ad hoc,’ maybe you should pause and reflect.”

CDC: 12 percent of pools closed after inspection

The CDC’s focus on “recreational water venues,” which include swimming pools, interactive fountains, kiddie/wading pools, water slides and and therapy pools,  is understandable given that about 314 million Americans visit them each year.

swimming-pools
Photo by Tom@HK via Flickr

From 1997 to 2006, the CDC says, infections caused by norovirus, shigella and other bacteria caused almost a quarter of pool-related gastroenteritis outbreaks, despite the fact that they could have been kept at bay with proper maintenance and inspections.

To better understand those lapses, the CDC examined 121,020 pool inspections volunteered by 15 jurisdictions. The majority came from the states of Florida (52,752) and South Carolina (22,111). The numbers tell a pretty compelling story on their own:

Pool inspections are a common, yet overlooked source of community health data, and this CDC release should be easy to localize, especially if you live in one of the 15 areas that volunteered the data used in the report. See the full list after the jump.
Read more

Next Page »