Since I don't think that MSNBC is a reliable source of data <rolling eyes>, I researched the actual study.



http://www.ajcn.org/content/90/3/505.full.pdf


In middle-aged adults, the relation between body mass index (BMI; in kg/m2) and overall mortality is U-shaped, with increased risk in the lowest and highest percentiles. That is, very skinny and very fat people are at higher risk of death, but normal and overweight people are at lower risk of death.

In contrast, studies in elderly persons suggest that the risk of mortality decreases with increasing body weight. In the elderly, being overweight or even obese does not increase morbidity or mortality, but being underweight correlates with increased morbidity (e.g. bedsores) and mortality (death).

http://jama.ama-assn.org/content/298/17/2028.short

Cause-specific relative risks of mortality from the National Health and Nutrition Examination Survey 1971-1994, with mortality follow-up through 2000 (571, 042 person-years of follow-up) were combined with data on BMI from 1999-2002 with underlying cause of death information for 2.3 million adults 25 years and older from 2004 vital statistics data for the United States…. [This is a gigantic data set, so the results are statistically very reliable. – W]

Based on total follow-up, underweight (BMI < 18) was associated with significantly increased mortality from noncancer, non-CVD (cardiovascular disease) causes (23,455 excess deaths) but not associated with cancer or CVD mortality. Overweight (BMI 25-30) was associated with significantly decreased mortality from noncancer, non-CVD causes (69,299 fewer deaths) but not associated with cancer or CVD mortality.

[Obesity (BMI >30) was associated with elevated deaths, but I won’t discuss that here because we are focused on underweight. – W]

http://jama.ama-assn.org/content/293/15/1861.full

Excess Deaths Associated With Underweight, Overweight, and Obesity
1. Katherine M. Flegal, PhD;
2. Barry I. Graubard, PhD;
3. David F. Williamson, PhD;
4. Mitchell H. Gail, MD, PhD
AMA. 2005;293(15):1861-1867. doi: 10.1001/jama.293.15.1861

Objective To estimate deaths associated with underweight (body mass index [BMI] <18.5), overweight (BMI 25 to <30), and obesity (BMI =30) in the United States in 2000.

Results

Relative to the normal weight category (BMI 18.5 to <25), obesity (BMI =30) was associated with 111,909 excess deaths and underweight with 33, 746 excess deaths.

Overweight (BMI between 25 and 30) was not associated with excess mortality (86,094 fewer deaths). [end quotes]


Low BMI is correlated with higher risk of osteoporosis even when Low BMI is defined as 22 and not as 18 or less. A combination of lower nutritional intake and less physical stress on the bones seems to be the reason. Even if you lift weights from time to time, a heavier person carries more weight around ALL the time. If you are light weight, it’s essential to stress the bones regularly and eat plenty of calcium, magnesium and Vitamin D to maintain bone strength. Osteoporosis can lead to morbidity and mortality because a broken hip can lead to a quick decline in the elderly.

http://www.ncbi.nlm.nih.gov/pubmed/11256896
http://ehealthmd.com/content/are-you-risk-developing-osteopo...

When a person becomes critically ill, reserves of fat may provide a nutritional buffer during recovery. A person (especially an elderly person) who does not have nutritional reserves may die during a critical illness (e.g. pneumonia) while a person who has reserves may survive and recover.

As we all know, statistics are valid for entire populations, because there is large variation between individuals. This is why it doesn't make sense to say, "My mom is thin, but she's OK." Also, the studies cover mortality, so if Mom is still alive, she hasn't died yet, so it's impossible to know how she would survive a critical illness.

Each person is different due to genetics and epigenetics. About 20% of the population has the APOE epsilon HDL haplotype, which leads to cardiovascular disease and Alzheimer's if meat is eaten and the person is sedentary.

My father had diabetes. I noticed that my triglycerides were rising into the danger zone. By following a low-carb diet, I was able to reduce my triglyceride level from 175 to 99 without medication (also I lost 15 pounds). My husband needs carbs to function -- his genetics are different.

Bottom line: Maintaining a BMI between 22-30 is the best for long life. Each person should consider their personal and family genetics when deciding precisely how to meet this goal with diet and exercise.