More and more foods bear a mishmash of warnings that they might accidentally contain ingredients that could seriously sicken people with food allergies. Yet there are signs that the labels are creating confusion among families that should heed them — even as new testing shows there is a real, if probably small, chance that foods with even the most vaguely worded warnings truly pose a risk.

The disconnect is sparking calls for standards on what are now voluntary warning labels. The Food and Drug Administration plans to seek advice from consumers and food makers, perhaps by year's end, before considering whether to intervene.

Worried the labels may be losing credibility, the industry's Grocery Manufacturers/Food Products Association already is preparing to update its own guidelines on when foods should carry the warnings.

Consumers see the label "on so many products, they say, 'Oh heck, I'm going to ignore it,'" laments Dr. Steve Taylor, a food scientist at the University of Nebraska who co-authored a recent study about the confusion.

For the seriously allergic, "I've characterized it as akin to playing Russian roulette with a really big gun that has 100 chambers and only one bullet. Sooner or later if you eat these products, you're going to eat the wrong one," he said.

About 12 million Americans have some degree of food allergy. Severe food allergies trigger 30,000 emergency room visits a year, and 150 to 200 deaths a year. Food labels help the allergic avoid ingredients that could sicken them.

A law that took effect last year requires foods that intentionally contain highly allergenic ingredients such as peanuts, shellfish or eggs to disclose that in plain language.

The accidental-allergy warnings are different: They're aimed at foods that aren't supposed to contain a particular allergen but might become contaminated with it. They may be made in the same factory, or on the same machines as allergen-containing goods.

In a report to Congress last year, FDA said a quarter of recently inspected food factories had the potential for such a mix-up.

The warnings are voluntary, so different companies use different, sometimes vague, wording. Nor does anyone count how many foods bear them, although all sides agree more are. Enter the new research, in this month's Journal of Allergy and Clinical Immunology.

First, the Food Allergy & Anaphylaxis Network, an influential consumer group, surveyed more than 600 parents of food-allergic children. In 2006, 75 percent said they would never buy a food with an accidental-allergy warning — down from 85 percent when the network posed the same question in 2003.

A warning's wording determined if some parents ignored it: "May contain peanuts" sounds scarier than "packaged in a facility that processes peanuts" — and thus 88 percent said they heeded the first warning while just 64 percent heeded the latter.

That echoes an FDA experiment that found wording matters in persuading allergic consumers to believe the warning.

But are the warnings real? Nebraska's Taylor tested 179 products that bore a variety of accidental-peanut warnings, and found 7 percent did contain peanuts — some of them traces, but some enough to seriously sicken.

That's a small proportion. But Taylor is quick to note he only tested samples from two batches of each product. Test more — say the first batch of oatmeal cookies packaged after the machine was supposedly cleaned of peanut butter cookies — and that number may grow.

Moreover, contrary to some parents' beliefs, peanuts were in some products with every version of the label, including two of 51 foods that bore a "may contain" warning and seven of 68 labeled "made in the same facility."

On the other hand, the allergy network is increasingly concerned that foods that never before carried warnings suddenly are getting them — including puzzling ones, like canned vegetables with nut warnings — a trend perhaps fueling confusion.

"We're seeing every week an increase in the number of these 'may contain' statements on products you wouldn't expect to see them on," says network founder Anne Munoz-Furlong, whose group is pushing FDA and industry for new labeling standards.

"Don't ignore it, because you don't know when it's true," she tells consumers.

The industry is "troubled by what appears to be an increased use of 'may contain' labeling," says Regina Hildwine, labeling policy chief for the grocery manufacturers. "This is not just something that you should put on a package without thinking."

Penny Ackerman of Bethlehem, Pa., is strict about label reading to protect her 3-year-old son, Gregory, who is severely allergic to peanuts and has a milder tree-nut allergy.

But, "we even have to watch labels we didn't used to," she says with frustration.

On a recent grocery trip, Ackerman didn't notice until she got home that the chocolate chips she had always bought with confidence now warned they were made in a factory that uses peanuts. She wonders if the change is because of a new factory, or just that the company hadn't gotten around to labeling until now.

"If you don't see it on the label, is it safe or is it not safe? Because you don't know."

Researchers have a new clue about why a widely used AIDS drug has certain side effects such as mysterious fat deposits, they reported on Monday.

Parallels between the side effects of protease inhibitors -- a critical component of HIV drug cocktails -- and genetic conditions that cause early aging may help explain the often debilitating fat deposits and other results, they said.

Protease inhibitors can cause metabolic problems such as an unhealthy buildup of cholesterol in the blood, high blood pressure, and increased risk of diabetes.

They also trigger a condition called lipodystrophy -- a strange redistribution of body fat that gives some patients wasted cheeks and limbs, and a so-called "buffalo hump" of fat on the back of the neck.

Doctors have long been mystified by how protease inhibitors and other HIV drugs cause such effects, which occur in tens of thousands of drug takers worldwide, said Dr. Charles Flexner at Johns Hopkins University School of Medicine, who was not involved in this study.

In an attempt to shed light on this, a group at the University of California Los Angeles and Purdue University in Indiana treated mouse and human cells with protease inhibitor, and found that they accumulated a particularly clumpy form of a protein called prelamin A.

The drug triggered this by blocking the action of another protein -- called ZMPSTE24 -- that converts prelamin A into its useful form, they reported in the Proceedings of the National Academy of Sciences.

Cells with lower levels of ZMPSTE24 to begin with were particularly affected by the protease inhibitor.

Purdue University's Christine Hrycyna, who worked on the study, said that blocking this protein might contribute to the metabolic side effects of protease inhibitors.

Patients with early aging syndromes, including Hutchinson-Gilford progeria, have symptoms that mimic the side effects, and the same protein is clumped in their cells, said Hrycyna. But how this might affect metabolism is not clear, she said.

"The side effects are probably due not to just one simple thing," Hrycyna said.

"I think this paper may provide new insights into possible mechanisms for some of the side effects of protease inhibitors," added Flexner.

The researchers also tested some of the other drugs that are commonly used in AIDS cocktails, known as highly active antiretroviral therapy or HAART.

But the other drugs did not cause the same protein accumulation, even though they can cause similar side effects in people, the researchers wrote.

"They are probably due to a combination of all these different drugs," Hrycyna added.

The researchers now want to see if their theory holds true in HIV patients, and if versions of protease inhibitors that do not block ZMPSTE24 as much might cause fewer side effects in these patients.

The various anti-inflammatory drugs (NSAIDs), such as ibuprofen and naproxen, which are used to treat arthritis and other painful conditions, differ in the risks of heart attacks and stroke, according to a report in the Annals of the Rheumatic Diseases. Much of the difference seems to depend on how the drug interacts with aspirin, which is commonly given to prevent heart attacks.

"Traditional NSAIDs, and ibuprofen in particular, do have the potential to interact with the effects of aspirin," Dr. Michael E. Farkouh from Mount Sinai School of Medicine, New York told Reuters Health. "For patients at high (risk of heart attack and stroke), this interaction is serious since these patients require aspirin to prevent" these cardiovascular problems.

Farkouh and associates investigated cardiovascular outcomes in high-risk patients with arthritis treated with ibuprofen, naproxen, or lumiracoxib. Their analyses were based on data for over 18,000 participants in the TARGET (Therapeutic Arthritis Research and Gastrointestinal Event Trial) study.

Among patients taking low-dose aspirin, the risk of death, heart attack, or stroke at 1 year was more common with ibuprofen than with lumiracoxib, the authors report, but did not differ significantly between naproxen and lumiracoxib.

Among patients not taking low-dose aspirin, there was no difference in outcomes between ibuprofen and lumiracoxib, the report indicates, but there were fewer cardiovascular events among those taking naproxen rather than lumiracoxib.

"For patients at high cardiovascular risk, naproxen is the drug of choice," Farkouh said. "It has the best overall safety record."

SOURCE: Annals of the Rheumatic Diseases, June 2007.

A protein in blood points to rising amounts of a particularly lethal form of body fat around organs, scientists say.

As levels of retinol-binding protein 4 (RBP4) rise, so do levels of "inter-abdominal fat" linked to an increased risk for heart disease and type 2 diabetes, the researchers say.

"Increased inter-abdominal fat is known to be associated with cardiovascular risk," said study co-author Dr. Barbara B. Kahn, chief of the Division of Endocrinology, Diabetes and Metabolism at Beth Israel Deaconess Medical Center, in Boston.

Reporting in the July issue of Cell Metabolism, her team noted that increased "deep belly" fat around organs has long been linked to an increased risk for insulin resistance and type 2 diabetes. "The regulation of this protein may tell us completely new information about what really causes type 2 diabetes," said Kahn, who is also a professor of medicine at Harvard Medical School.

In the study, the researchers found that the amount of RBP4 in the blood accurately reflected the amount of fat surrounding the abdominal organs. That means that "RBP4 might be able to be used as a marker to indicate cardiovascular risk," Kahn said.

She stressed that RBP4 is not a cause of obesity. However, increased levels appear be associated with this particular type of abdominal-fat obesity.

In the study, Kahn and colleagues looked at biopsy samples of abdominal fat from 196 people. They found that more RBP4 is made in visceral (deep belly) fat compared with the subcutaneous fat that lies just beneath the skin. In addition, blood levels of RBP4 are greater in people who are obese. These people have double or triple the amount of RBP4 compared with normal-weight people.

"The gene expression of RBP4 is increased more in visceral adipose [fat] tissue -- the adipose tissue surrounding the internal organs -- than it is in the subcutaneous adipose tissue," Kahn said. So, levels of RBP4 are higher in people who have a so-called "visceral pattern" of obesity compared with people that have a subcutaneous pattern of obesity, she said.

In earlier research, Kahn's team also found that the levels of RBP4 were elevated in people with insulin resistance, people who are obese, and people with type 2 diabetes. This was also the case in healthy people with a family history of diabetes.

According to Kahn, there's ongoing research into drugs that could lower RBP4 levels.

But there's another tried-and-true means of lowering RBP4, she added.

"Levels can also be regulated by physical exercise," Kahn said. In prior research, her team showed that "people who benefited from an exercise program lowered their levels of serum RBP4 when they got more insulin-sensitive," she said.

Kahn had also shown in earlier research that teens who went on a low-carbohydrate diet along with an exercise program lowered their RBP4 levels.

One expert thinks that lowering RBP4 levels might help treat heart disease and type 2 diabetes.

"This study suggests that RBP4 can be a good biomarker to quantify visceral adiposity, which is closely linked to metabolic syndrome," said Dr. Tae-Hwa Chun, from the Department of Internal Medicine, Metabolism, Endocrinology and Diabetes at the University of Michigan. "This article also supports the notion that all the fats are not equal in their functions."

In experiments with mice, RBP4 decreases insulin sensitivity of muscle and liver tissue, which is considered a precursor to diabetes, Chun said.

"It is still not clear whether RBP4 regulates insulin sensitivity by controlling retinoic acid metabolism or by directly acting on muscle or liver cells," Chun said. "The drug Fenretinide, which is shown to lower RBP4 levels, has been already used as a chemotherapeutic agent for cancer. The side effects of the drug, however, need to be carefully weighed against its possible benefit for metabolic diseases."

In another report published in the same issue of the journal, a research team led by Bruce Spiegelman of the Dana-Farber Cancer Institute in Boston identified a gene called PRDM16 that regulates the production of so-called "brown fat" in mice. Brown fat is a type of fat that actually generates heat and counters obesity caused by overeating.

"Brown fat is present in mice and in human infants, where it keeps them warm by dissipating food energy as heat, instead of storing it as 'white' fat," Spiegelman said in a prepared statement. "Human adults don't have much brown fat, but there is some, and from a therapeutic perspective, the question is whether that pathway can be reactivated."

The researchers hope their discovery will lead to new ways of treating obesity in humans.

Despite many hospitals' reluctance to make the investment, installing an electronic medical records system pays for itself in less than two years, a new study finds.

Such a system was put in place at the University of Rochester Medical Center, N.Y., and recouped its initial cost within 16 months, say the authors of a study in the July issue of the Journal of the American College of Surgeons.

That experience runs counter to the worries of many health care providers, who are often reluctant to invest in such tracking systems.

"Health care providers most frequently cite cost as a primary obstacle to adopting an electronic medical records system. And, until this point, evidence supporting a positive return on investment for electronic health records technologies has been largely anecdotal," study co-author Dr. David A. Krusch, of the department of surgery at the University of Rochester, said in a prepared statement.

He and his colleagues analyzed the return on investment of an electronic health records system used in five ambulatory offices representing 28 health care providers. The study compared the costs of a number of tasks -- such as pulling patient charts, creating new charts, filing time, support staff salary, and data transcription -- in the third quarter of 2005 (after the system was installed) to costs in the third quarter of 2003, when the tasks were still being done manually.

The new system reduced costs by almost $394,000 per year, the study found, and nearly two-thirds of those savings came from a drastic reduction in the amount of time required to manually pull patient charts. The electronic medical records system cost $484,577 to install and operate in the first year, which means the medical center recouped its investment within 16 months.

After the first year, it cost about $114,000 a year to operate the system. That means an annual savings of more than $279,500 for the medical center, which works out to close to $10,000 per health care provider using the system, the researchers said.

Placenta previa occurs when, during pregnancy, the placenta drops lower in the uterus and covers the opening to the cervix. The placenta may also separate from the uterine wall as the cervix begins to dilate.

Placenta previa affects about one in 200 pregnant women in the third trimester, the American Pregnancy Association says.

The association lists these common risk factors for placenta previa:

* Having had one or more previous deliveries.
* Having had a previous Caesarean section.
* Having had a multiple birth (twins or triplets).
* Having had prior surgery on the uterus.

Fewer high school students are having sex these days, and more are using condoms. The teen birth rate has hit a record low.

More young people are finishing high school, too, and more little kids are being read to, according to the latest government snapshot on the well-being of the nation's children. It's good news on a number of key wellness indicators, experts said of the report being released Friday.

"The implications for the population are quite positive in terms of their health and their well-being," said Edward Sondik, director of the National Center for Health Statistics. "The lower figure on teens having sex means the risk of sexually transmitted diseases is lower."

In 2005, 47 percent of high school students — 6.7 million — reported having had sexual intercourse, down from 54 percent in 1991. The rate of those who reported having had sex has remained the same since 2003.

Of those who had sex during a three-month period in 2005, 63 percent — about 9 million — used condoms. That's up from 46 percent in 1991.

The teen birth rate, the report said, was 21 per 1,000 young women ages 15-17 in 2005 — an all-time low. It was down from 39 births per 1,000 teens in 1991.

"This is very good news," said Sondik. "Young teen mothers and their babies are at a greater risk of both immediate and long-term difficulties."

The birth rate in the 15-19 age group was 40 per 1,000 in 2005, also down sharply from the previous decade.

Education campaigns that started years ago are having a significant effect, said James Wagoner, president of Advocates for Youth, a Washington-based nonprofit group that focuses on prevention of teen pregnancy and sexually transmitted diseases.

"I think the HIV/AIDS epidemic and the efforts in the '80s and '90s had a lot to do with that," Wagoner said of the improved numbers on teen sex, condoms and adolescent births.

"We need to encourage young teens to delay sexual initiation and we need to make sure they get all the information they need about condoms and birth control," he said.

The report was compiled from statistics and studies at 22 federal agencies, and covered 38 key indicators, including infant mortality, academic achievement rates and the number of children living in poverty.

Other highlights:

_The percentage of children covered by health insurance decreased slightly. In 2005, 89 percent of children had health insurance coverage at some point during the year, down from 90 percent the previous year.

_The percentage of low birthweight infants (born weighing less than 5 pounds, 8 ounces) increased. It was 8.2 percent in 2005, up from 8.1 percent in 2004.

_More youngsters are getting reading time. Sixty percent of children ages 3-5 (and not in kindergarten) were read to daily by a family member in 2005, up from 53 percent in 1993.

_The percentage of children who had at least one parent working year round and full-time increased to 78.3 percent in 2005, up from 77.6 percent the previous year.

_More young people are completing high school. In 2005, 88 percent of young adults had finished high school — up from 84 percent in 1980.

The report was released by the Federal Interagency Forum on Child and Family Statistics — a consortium of federal agencies that includes the National Institute of Child Health and Human Development, the Census Bureau and the Administration for Children and Families.

In the month after a surprising analysis revealed possible heart risks from the blockbuster diabetes drug Avandia, reports of side effects to federal regulators tripled.

The sudden spike is a sign that doctors probably were unaware of the drug's possible role in their patients' heart problems and therefore may not have reported many such cases in the past, several experts said.

It also shows the flaws of the safety tracking system and suggests that a better one might have detected a potential problem before the drug had been on the market for eight years.

Avandia is used to control blood sugar, helping more than 6 million people worldwide manage Type 2 diabetes, the kind that is linked to obesity. These people already are at higher risk for heart attacks, so news that the drug might raise this risk by 43 percent was especially disturbing.

In the 35 days after May 21, when the New England Journal of Medicine published the analysis on the Internet, reports of heart attacks, deaths and hospitalizations leaped. The sharp rise in reports of heart problems appears in data obtained by The Associated Press through a Freedom of Information Act request to the federal Food and Drug Administration.

Only five heart attacks were reported in the 35 days before the study, compared with 90 in the same period afterward. Heart-related hospitalizations went from 11 to 126. The reports involve rosiglitazone, sold as Avandia and Avandamet.

Reporting a drug's side effects is voluntary, and only a crude indication rather than a scientific measure of how many problems patients are actually having. The FDA relies on this unenforced system once a drug is on the market. Critics say it leads to haphazard oversight in which problems can be missed because doctors don't connect the dots between a drug and symptoms they see in an individual patient.

With Avandia, the published analysis likely led to more cases being reported, said Vanderbilt University diabetes specialist Dr. Alvin C. Powers.

"Now, patients and their doctors are much more aware of the possible link between Avandia and cardiovascular disease. This is good — this is going to help us going forward to determine whether or not this drug is safe," he said.

The drug's manufacturer, British-based GlaxoSmithKline PLC, insists that the drug is safe and effective.

"This is a very well-known phenomenon," where news reports lead to increased reporting, said company spokeswoman Mary Anne Rhyne. "It's good that there's awareness of the reporting system, but drawing conclusions on such data is inappropriate."

The FDA plans hearings on safety concerns about the drug on July 30. In the meantime, diabetes experts have advised users of the medication to talk to their doctors and not to immediately discontinue it.

The side effects reported range from as minor as a blister to as serious as sudden cardiac death. Most of the reports the AP reviewed seemed to involve serious side effects, and rosiglitazone was listed by the FDA as the "primary suspect" rather than other medicines the patient may have been taking.

There was a total of only 50 adverse event reports in January and 73 in February. From April 16 to May 21, when the study was published, 121 events were reported, including 11 deaths. In the 35 days after the study, 357 events were reported, including 38 deaths.

"You really can't infer anything about incidence rates from that," because the spike in reports is likely due to the "publicity effect" of the study, said Dr. David Graham, an FDA drug safety expert.

Dr. David Nathan, chief of diabetes care at Massachusetts General Hospital, agreed, saying it was "not conceivable" that only five people among the 1 million Americans taking Avandia had heart attacks in the month preceding the May 21 study, as the FDA reports suggest.

"It just heightens the concern about the poor reporting we have," said Nathan, who has received speaker fees from Glaxo and other drug companies. Powers and Graham have no financial ties to any diabetes drug makers.

The issue has roiled the medical community and sparked congressional probes into whether the FDA is properly investigating safety issues. The FDA issued a "safety alert" about the drug only after the May 21 study came out, even though Glaxo had informed the agency of its own analysis of heart risks nearly a year beforehand and possibly as early as 2005.

Avandia's label warns about possible heart failure and other heart problems when taken with insulin. The drug also raises LDL, or bad cholesterol, and can cause fluid retention and weight gain.

China, reeling from a series of health scandals, told food and drug companies on Friday to put safety first and warned the growing problem could threaten social stability in the world's most populous nation.

The revelations have grabbed global attention after patients in Panama died from poisonous ingredients in medicine and pets died in the United States from substandard feed, while tainted Chinese toothpaste was found in Central America and elsewhere.

"Companies are the people with the first line of responsibility for food and drug safety, and must strengthen management, uphold the law in their operations, honestly follow regulations and guarantee safety," food and drug watchdog chief Shao Mingli said.

"Food and drugs are special products, and manufacturers and sellers cannot only go after economic gain," Shao told a work meeting, according to a statement on the watchdog's Web site (www.sda.gov.cn).

"Ensuring food and drink safety for the public is our unshirkable responsibility," he added.

This week, domestic media reported on dumplings found to have been stuffed with cardboard scraps and bogus rabies vaccines.

With the Beijing Olympics just over one year away, officials also said a specially developed food quality control system would be employed at 39 Olympic test events starting August 8.

Public fears about food safety grew in 2004, when at least 13 babies died of malnutrition in Anhui province, in eastern China, after they were fed fake milk powder with no nutritional value.

These incidents could provoke unrest, Shao said, echoing a warning given by a colleague last weekend.

"Food and drug safety is critical to the people's health and lives, and is critical for social stability and harmony," said Shao, whose predecessor, Zheng Xiaoyu, was executed on Tuesday for corruption.

"This issue can easily morph into a much larger one and directly affect the image of local governments and the state, affect social stability and harmony as well as socio-economic development," Shao added.

The growing list of food safety scandals has prompted the army to step up checks on where its rations come from, a state newspaper said on Friday.

The People's Liberation Army -- whose 2.3 million servicemen make it the world's largest military -- will only be allowed to order food from suppliers who pass local government hygiene and safety tests, the official China Daily said.

"To strengthen food safety is to guarantee the PLA's combat capacity," it quoted Zhou Pengjun, an official at the General Logistics Department, as saying.

Canteens have to install more dishwashers, food will be sampled regularly and most staff will not be allowed to share food from common dishes as is the usual custom in China, the report said.

The army has also pledged to share information with civilian food and safety officials, the newspaper added.

More than 120 people who ate from the same booth at the Taste of Chicago food festival last week became ill, at least nine of them with salmonella poisoning and 10 who were hospitalized, the health department said Thursday.

The Chicago Department of Health was investigating the cases and confirmed that nine were caused by salmonella bacteria. The number could increase because lab results are pending in some of the cases.

The 126 people all ate at the Pars Cove Persian Cuisine booth, which served cucumber salad over hummus, grilled lamb and beef, pomegranate barbecued chicken and baklava.

It was the first confirmed outbreak of a food-borne illness associated with the Taste of Chicago in at least 20 years, the health department said in a release. The 10-day outdoor festival, which ended Sunday, brings out food from more than 70 restaurants and concerts.

Mike Bambouyani, manager of Pars Cove, said the restaurant is cooperating with health officials to determine the cause of the outbreak.

Officials inspected the restaurant and were interviewing and testing restaurant workers who handle food. The bacteria could have been transmitted by food handlers or could have come from a food ingredient, officials said.

Salmonella can cause serious and sometimes fatal infections, especially in young children, frail or elderly people and others with weakened immune systems. Symptoms include fever, diarrhea, nausea, vomiting and abdominal pain.