A bug of SAS?

A bug of SAS?

• Numbers great and small: the problem of big datum
• Numerical accuracy in SAS software
• TS-654 Numeric Precision 101

We may have a problem on create A1c categories by using NHANES III data, which is a discrepancy between multilevel categories and two-level categories, then I figured out 'ROUND()' function can fix this kind discrepancy. However, this issue is still haunting our a lot. After I looked into more, I found the SAS did not pick up 5.2 into the '1' group, when I use 'GHP >= 5.2' (see codes and output below).

There is no similar problem with NHANES 99-04. The differences between NHANES III and NHANES 99 are: 1) NHANES III dataset is a SAS version 6 dataset; 2) GHP of NHANES III has been formatted as F6.1.

     

LIBNAME NHANES3 V6 'Q:\epistat\datasets\NHANES\ORIGINAL\NHANES3\';

DATA N3;

  SET NHANES3.LABNEW (KEEP=GHP);       * This is a SAS v6 dataset;

    IF . LT GHP LT 7777;               * GHP is in a F6.1 format;

    GHP2=ROUND(GHP,0.1);               * What is 'ROUND()' doing here?;

        IF GHP >= 5.2 THEN GHPGRP1=1;  ELSE GHPGRP1=2;

        IF GHP > 5.19 THEN GHPGRP2=1;  ELSE GHPGRP2=2;

    IF GHP2>=5.2 THEN GHPGRP3=1;  ELSE GHPGRP3=2;

        LABEL GHPGRP1='ORIGINAL GHP VALUE, CUTPOINT 5.2'

              GHPGRP2='ORIGINAL GHP VALUE, CUTPOINT 5.19'

                  GHPGRP3='ROUNDED GHP VALUE, CUTPOINT 5.2';

RUN;

PROC FREQ DATA=N3;

  TABLES GHPGRP1 GHPGRP2 GHPGRP3 GHPGRP1*GHPGRP3; RUN;

============ OUTPUT ============

                                                                                             379

The FREQ Procedure

              ORIGINAL GHP VALUE, CUTPOINT 5.2

                                    Cumulative    Cumulative

GHPGRP1    Frequency     Percent     Frequency      Percent

------------------------------------------------------------

      1       11535       49.14         11535        49.14

      2       11941       50.86         23476       100.00

 

              ORIGINAL GHP VALUE, CUTPOINT 5.19

                                    Cumulative    Cumulative

GHPGRP2    Frequency     Percent     Frequency      Percent

------------------------------------------------------------

      1       13463       57.35         13463        57.35

      2       10013       42.65         23476       100.00

 

               ROUNDED GHP VALUE, CUTPOINT 5.2

                                    Cumulative    Cumulative

GHPGRP3    Frequency     Percent     Frequency      Percent

------------------------------------------------------------

      1       13463       57.35         13463        57.35

      2       10013       42.65         23476       100.00

 

Table of GHPGRP1 by GHPGRP3

GHPGRP1(ORIGINAL GHP VALUE, CUTPOINT 5.2)

          GHPGRP3(ROUNDED GHP VALUE, CUTPOINT 5.2)

Frequency|

Percent  |

Row Pct  |

Col Pct  |       1|       2|  Total

---------+--------+--------+

       1 |  11535 |      0 |  11535

         |  49.14 |   0.00 |  49.14

         | 100.00 |   0.00 |

         |  85.68 |   0.00 |

---------+--------+--------+

       2 |   1928 |  10013 |  11941

         |   8.21 |  42.65 |  50.86

         |  16.15 |  83.85 |

         |  14.32 | 100.00 |

---------+--------+--------+

Total       13463    10013    23476

            57.35    42.65   100.00
=========================================

Thank you all. J and I discussed this underlying issue yesterday as well. I could not find any exact 5.2 of GHP value. I don't think NCHS inputted this kind of GHP value.

=========================================

I tried changing the following statement,  and GHPGRP2 was assigned a value of 2 in record 39.  Removing one of the decimal places resulted in a value of 1.

IF GHP >= 5.1999999999999999 THEN GHPGRP2=1;  ELSE GHPGRP2=2;

I tried changing the format and unformatting GHP, but I could only get it to display 5.2 in record 39.  Apparently, the value is not exactly 5.2, but the precision is so deep that it cannot be displayed.

=========================================

Thank you B and D. We are pinpointed the issue. I re-run my codes and get outputs below. Usually SAS gives a little bit more from what we see. However, this time (NHANES III) SAS gives a little bit less from what we see. Keep tune and be aware. Using round() if your want fix this issue now.

LIBNAME NHANES3 V6 'Q:\epistat\datasets\NHANES\ORIGINAL\NHANES3\';

DATA N3;

  SET NHANES3.LABNEW (KEEP=GHP);       * This is a SAS v6 dataset;

    IF ROUND(GHP,.1) EQ 5.2;

        DIFF_GHP_FROM_5POINT2=GHP-5.2;

        GHP_GE_5POINT2=(GHP GE 5.2);

        GHP_GE_5POINT19=(GHP GE 5.19);

RUN;

TITLE 'OUTPUT OF NHANES III';

PROC PRINT DATA=N3 (OBS=5); FORMAT GHP DIFF_GHP_FROM_5POINT2 F32.31; RUN;

data two;

   input a b @@;

   c=b*0.1;

   ca_diff=c-a;

   c_ge_point3=(a ge 0.3);

   c_le_point3=(c le 0.3);

cards;

0.1 1 0.2 2 0.3 3 0.4 4 0.5 5

;

run;

title 'output of testing dataset';

proc print data=two; format c ca_diff f32.31; run;

title;run;

 

OUTPUT OF NHANES III

                                                                              GHP_GE_    GHP_GE_

 Obs                                GHP              DIFF_GHP_FROM_5POINT2   5POINT2   5POINT19

   1   5.200000000000000000000000000000   -.000000000000000888178419700125      0          1

   2   5.200000000000000000000000000000   -.000000000000000888178419700125      0          1

   3   5.200000000000000000000000000000   -.000000000000000888178419700125      0          1

   4   5.200000000000000000000000000000   -.000000000000000888178419700125      0          1

   5   5.200000000000000000000000000000   -.000000000000000888178419700125      0          1

output of testing dataset

                                                                                  c_ge_   c_le_

Obs   a   b                                 c                           ca_diff  point3  point3

 1   0.1  1  .1000000000000000000000000000000  .0000000000000000000000000000000     0       1

 2   0.2  2  .2000000000000000000000000000000  .0000000000000000000000000000000     0       1

 3   0.3  3  .3000000000000000000000000000000  .0000000000000000277555756156289     1       0

 4   0.4  4  .4000000000000000000000000000000  .0000000000000000000000000000000     1       0

 5   0.5  5  .5000000000000000000000000000000  .0000000000000000277555756156289     1       0

 

Free articles on PERIPHERAL ARTERIAL DISEASE on Ann Intern Med

In January 2007 Annals launched a new monthly section:

IN THE CLINIC

Our third issue on
PERIPHERAL ARTERIAL DISEASE
is available at:

www.annals.org/cgi/content/abstract/146/5/ITC3-1

Find Out More at:

www.annals.org/intheclinic

FW: Philippine Dept of Health Endorses Medicinal Plant Bitter Gourd for Diabetes

 

 

We have treated DM patients using bitter melon for many years in China. There are effects but may not be claimed as cured, I think.

---

Ampalaya can cure diabetes, says DoH

Tuesday, March 6, 2007

The efficacy of ampalaya or bitter gourd in treating diabetes was officially announced by the Department of Health (DoH) yesterday. It cited a 10-year study that found out that the vegetable can effectively regulate blood sugar in the same way as a regular anti-diabetes drug. Results of the study conducted by the Philippine Council for Health Research and Development (PCHRD) elevated the ampalaya from a mere nutritional supplement to a real medicine. The study has been certified by the Philippine Institute of Traditional and Alternative Health Care (PITAHC). "We compared ampalaya leaves with an anti-diabetes drug, and we found out that ampalaya has the same effect on the patient. It means the action of ampalaya on blood sugar is equivalent to the action of the medicine," Dr. Cirilo Galindez, PITAHC director general, said. "According to the study, it even has more blood sugar lowering effect," DoH Undersecretary Jade del Mundo said. The study revealed that a 100 milligram per kilo dose per day is comparable to 2.5 milligrams of the anti-diabetes drug Glibenclamide taken twice per day, Del Mundo said. In 2003, ampalaya was omitted from the list of government-recognized medicinal plants, and was demoted from being a scientifically validated medicinal plant to folkloric. Former Health Secretary Dr. Jaime Galvez Tan said this happened because previous scientific studies were not enough to support the medicinal purposes of ampalaya. The DoH has issued an order reinstating ampalaya as an effective medicine for Type 2 diabetes or diabetes mellitus. Other plants recognized by the government as medicinal and marketed locally include lagundi (for fever, asthma, and headache), sambong (for gaseous stomach, fever, headache, and aromatic bath), acapulco (as wound wash and for itch), yerba buena (for cough, toothache, dizziness, fainting, hysteria, and arthritis), bayabas (guava), and "tsaang gubat."

Updated guidelines advise focusing on women's lifetime heart risk

http://www.americanheart.org/presenter.jhtml?identifier=3045524

Highlights of the changes include:

• Recommended lifestyle changes to help manage blood pressure include weight control, increased physical activity, alcohol moderation, sodium restriction, and an emphasis on eating fresh fruits, vegetables and low-fat dairy products.
• Besides advising women to quit smoking, the 2007 guidelines recommend counseling, nicotine replacement or other forms of smoking cessation therapy.
• Physical activity recommendations for women who need to lose weight or sustain weight loss have been added – minimum of 60–-90 minutes of moderate-intensity activity (e.g., brisk walking) on most, and preferably all, days of the week.
• The guidelines now encourage all women to reduce saturated fats intake to less than 7 percent of calories if possible.
• Specific guidance on omega-3 fatty acid intake and supplementation recommends eating oily fish at least twice a week, and consider taking a capsule supplement of 850–1000 mg of EPA (eicosapentaenoic acid) and DHA (docosahexaenoic acid) in women with heart disease, two to four grams for women with high triglycerides.
• Hormone replacement therapy and selective estrogen receptor modulators (SERMs) are not recommended to prevent heart disease in women.
• Antioxidant supplements (such as vitamin E, C and beta-carotene) should not be used for primary or secondary prevention of CVD.
• Folic acid should not be used to prevent CVD – a change from the 2004 guidelines that did recommend it be considered for use in certain high-risk women.
• Routine low dose aspirin therapy may be considered in women age 65 or older regardless of CVD risk status, if benefits are likely to outweigh other risks.   (Previous guidelines did not recommend aspirin in lower risk or healthy women.)
• The upper dosage of aspirin for high-risk women increases to 325 mg per day rather than 162 mg.   This brings the women’s guidelines up to date with other recently published guidelines.

Consider reducing LDL cholesterol to less than 70 mg/dL in very high-risk women with heart disease (which may require a combination of cholesterol-lowering drugs).

FTP site for big files

<<www.transferbigfiles.com.url>> TransferBigFiles.com is an nice FTP
site to upload and download files.

Shortcut to: http://www.transferbigfiles.com/

Potential to create a genetic test to predict diabetes

This is A breakthrough gene  (see report below) .

 

According to this short report, ' ...explain up to 70% of the genetic background of type 2 diabetes...' and '...SLC30A8, which is involved in regulating insulin secretion....', do these mean that insulin resistance is less important than insulin secretion among persons with type 2 DM?

 

This is A big leap for a gene study team, A baby step for predicting type 2 DM by using genes only. Hope I am wrong.

 

---

Subject: Potential to create a genetic test to predict diabetes

Breakthrough gene find may halt spread of adult diabetes

STEWART PATERSON

February 12 2007

An international team of scientists has made a breakthrough which could help halt the growth of diabetes.

Researchers from Britain and Canada have identified the gene that causes diabetes, which means potential sufferers could be tested to assess their risk of developing the disease.

It is hoped the discovery will also help develop new treatments for adult onset, or type 2 diabetes, which is one of the most common illnesses among middle aged and older people in the UK.

In Scotland, around 150,000 people suffer type 2 diabetes while another 60,000 are estimated to have the disease but are unaware. The number of people with the illness is expected to increase by half over the next decade.

The research, published online in the science magazine, Nature, is the first time the genetic make-up of any disease has been mapped in such detail.

If adult diabetes is not properly managed, patients can suffer severe complications, such as blindness, amputations and kidney disease, and are at far higher risk of developing heart disease or suffering a stroke.

The researchers at McGill University, Montreal, and Imperial College London, believe their findings explain up to 70% of the genetic background of type 2 diabetes.

Our research shows that this technology can generate big leaps forward Professor David Balding, Imperial College

In addition, one of the genetic mutations they detected may further explain the causes behind the disease, potentially leading to a new therapy.

Lead researcher Dr Constantin Polychronakos, of McGill University, said: "The rapidly increasing prevalence of type 2 diabetes is believed to be due to environmental factors, such as increased availability of food and decreased opportunity and motivation for physical activity, acting on genetically susceptible individuals."

The study revealed people with the disease have a mutation in a particular zinc transporter known as SLC30A8, which is involved in regulating insulin secretion.

Type 2 diabetes is caused by a deficiency in insulin and the researchers believe it may be possible to treat it by fixing this transporter.

Professor Philippe Froguel, of Imperial College London, said: "The two major reasons why people develop type 2 diabetes are obesity and a family link. Our new findings mean we can create a good genetic test to predict people's risk of developing this type of diabetes.

"If we can tell someone their genetics mean they are pre-disposed towards type 2 diabetes, they will be much more motivated to change things such as their diet to reduce their chances of developing the disorder."

The scientists reached their conclusions after comparing the genetic makeup of 700 people with type 2 diabetes and a family history of the condition, with 700 controls. They looked at mutations in the building blocks, called nucleotides, which make up DNA.

There are mutations in about one in every 600 nucleotides and the scientists examined more than 392,000 of these to find the ones specific to type 2 diabetes.

Professor David Balding, epidemiologist at Imperial College, said: "Our research shows this technology can generate big leaps forward. The task now is to study the genes identified in our work more intensively, to understand more fully the disease processes involved, devise therapies for those affected and to try to prevent future cases."

© All rights reserved. Reproduction in whole or in part without permission is prohibited.

Obesity worse than inactivity

There are overwhelming evidences that both obesity and physical inactivity related to higher risk of diabetes. However, comparison between two is not fair to physical activity. Physical inactivity is in behavior domain, and obesity is in sort of bio-marker domain. Obesity is a mediator and an offspring of behavior including physical inactivity , over-eating, genetic factors , etc.. 

---

 

	Powered by

	SAVE THIS | EMAIL THIS | Close

Obesity poses larger diabetes risk than inactivity NEW YORK (Reuters) -- Although obesity and lack of physical activity both raise the risk of type 2 diabetes in women, obesity appears to be the more important factor, researchers report in the journal Diabetes Care. Dr. Frank Hu of the Harvard School of Public Health, Boston, Massachusetts, and colleagues note that the relative contribution of obesity and inactivity to the risk of developing type 2 diabetes remains controversial. To investigate further, the researchers monitored 68,907 women taking part in the Nurses' Health Study, a large ongoing study that is evaluating women's health over time. The women in the current trial had no history of diabetes, cardiovascular disease or cancer at study entry. During 16 years of follow-up, there were 4,030 incident cases of type 2 diabetes. After allowing for age, smoking, and other diabetes-associated factors, the risk of type 2 diabetes increased progressively with increasing body mass index (BMI - the ratio of height to weight often used to determine whether someone is overweight or too thin). The risk also increased with waist circumference, and decreased with physical activity levels. Using women who had a healthy weight (BMI of less than 25) and were physically active as the reference group, the relative risks of type 2 diabetes were 16.75 in women with a BMI of 30 or more and were inactive. The corresponding risk in obese women who were active was 10.74. In women who were lean but inactive, the relative risk was 2.08. Although both variables were significant predictors of type 2 diabetes, the researchers found that the association for waist circumference was substantially stronger than that for physical inactivity. They researchers conclude that "the magnitude of risk contributed by obesity is much greater than that imparted by lack of physical activity," and therefore "weight loss and maintenance of healthy weight should be emphasized as an eventual goal to prevent the onset of type 2 diabetes." Copyright 2007 Reuters. All rights reserved.This material may not be published, broadcast, rewritten, or redistributed.

Stephen's Guide to the Logical Fallacies

Stephen's Guide to the Logical Fallacies
by Stephen Downes

Table of Contents

Welcome

How To Use This Guide

Table of Contents

Fallacies of Distraction

• False Dilemma: two choices are given when in fact there are three options
• From Ignorance: because something is not known to be true, it is assumed to be false
• Slippery Slope: a series of increasingly unacceptable consequences is drawn
• Complex Question: two unrelated points are conjoined as a single proposition

Appeals to Motives in Place of Support

• Appeal to Force: the reader is persuaded to agree by force
• Appeal to Pity: the reader is persuaded to agree by sympathy
• Consequences: the reader is warned of unacceptable consequences
• Prejudicial Language: value or moral goodness is attached to believing the author
• Popularity: a proposition is argued to be true because it is widely held to be true

Changing the Subject

• Attacking the Person:
  1. the person's character is attacked
  2. the person's circumstances are noted
  3. the person does not practise what is preached
• Appeal to Authority:
  1. the authority is not an expert in the field
  2. experts in the field disagree
  3. the authority was joking, drunk, or in some other way not being serious
• Anonymous Authority: the authority in question is not named
• Style Over Substance: the manner in which an argument (or arguer) is presented is felt to affect the truth of the conclusion

Inductive Fallacies

• Hasty Generalization: the sample is too small to support an inductive generalization about a population
• Unrepresentative Sample: the sample is unrepresentative of the sample as a whole
• False Analogy: the two objects or events being compared are relevantly dissimilar
• Slothful Induction: the conclusion of a strong inductive argument is denied despite the evidence to the contrary
• Fallacy of Exclusion: evidence which would change the outcome of an inductive argument is excluded from consideration

Fallacies Involving Statistical Syllogisms

• Accident: a generalization is applied when circumstances suggest that there should be an exception
• Converse Accident : an exception is applied in circumstances where a generalization should apply

Causal Fallacies

• Post Hoc: because one thing follows another, it is held to cause the other
• Joint effect: one thing is held to cause another when in fact they are both the joint effects of an underlying cause
• Insignificant: one thing is held to cause another, and it does, but it is insignificant compared to other causes of the effect
• Wrong Direction: the direction between cause and effect is reversed
• Complex Cause: the cause identified is only a part of the entire cause of the effect

Missing the Point

• Begging the Question: the truth of the conclusion is assumed by the premises
• Irrelevant Conclusion: an argument in defense of one conclusion instead proves a different conclusion
• Straw Man: the author attacks an argument different from (and weaker than) the opposition's best argument

Fallacies of Ambiguity

• Equivocation: the same term is used with two different meanings
• Amphiboly: the structure of a sentence allows two different interpretations
• Accent: the emphasis on a word or phrase suggests a meaning contrary to what the sentence actually says

Category Errors

• Composition: because the attributes of the parts of a whole have a certain property, it is argued that the whole has that property
• Division: because the whole has a certain property, it is argued that the parts have that property

Non Sequitur

• Affirming the Consequent: any argument of the form: If A then B, B, therefore A
• Denying the Antecedent: any argument of the form: If A then B, Not A, thus Not B
• Inconsistency: asserting that contrary or contradictory statements are both true

Syllogistic Errors

• Fallacy of Four Terms: a syllogism has four terms
• Undistributed Middle: two separate categories are said to be connected because they share a common property
• Illicit Major: the predicate of the conclusion talks about all of something, but the premises only mention some cases of the term in the predicate
• Illicit Minor: the subject of the conclusion talks about all of something, but the premises only mention some cases of the term in the subject
• Fallacy of Exclusive Premises: a syllogism has two negative premises
• Fallacy of Drawing an Affirmative Conclusion From a Negative Premise: as the name implies
• Existential Fallacy: a particular conclusion is drawn from universal premises

Fallacies of Explanation

• Subverted Support (The phenomenon being explained doesn't exist)
• Non-support (Evidence for the phenomenon being explained is biased)
• Untestability (The theory which explains cannot be tested)
• Limited Scope (The theory which explains can only explain one thing)
• Limited Depth (The theory which explains does not appeal to underlying causes)

Fallacies of Definition

• Too Broad (The definition includes items which should not be included)
• Too Narrow (The definition does not include all the items which shouls be included)
• Failure to Elucidate (The definition is more difficult to understand than the word or concept being defined)
• Circular Definition (The definition includes the term being defined as a part of the definition)
• Conflicting Conditions (The definition is self-contradictory)

Logic Resources

References

For Educators

Copyright Notice

What's New

Awards

Author

• Stephen Downes' Homepage
• Email Stephen