Monday, April 29, 2013

Diabetes Screening


Share | The Greeks, Asclepions of Hippocrates's time, screened for diabetes by taste. An alternative method was accomplished by poring the patient's urine on the ground and observing whether or not the urine attracted ants. The taste-test was said to have greater sensitivity while the ante-test offered greater specificity. It might be interesting to compare the sensitivity and specificity of the ante-test and the taste-test with today's fasting blood sugar and one hour glucose challenge test.

Sunday, April 28, 2013

Medical Education, medical information


Share |
Educators do not seem much interested in the clinical viewpoint of medical education, so here it is anyway. Medical education equates to the delivery of medical information -- much more than it is possible to teach.  Medical schools should be obligated to provide current medical information from the first year of medical school and throughout the life of the graduating physician --- Current forms of accessing medical information, CME or GME are completely inadequate and obsolete, locked up in copyright, cost and limited by the distribution of the printed text.
Currently there is no real-time source for the total sum of medical knowledge and the leading edge of information growth. Physicians and students alike need such a source that is up-to-date, real time, 24/7. That source should contain it all, terminology and diagnostic criteria as currently understood by the medical school and the specialties.[1] Every medical school should provide total medical terminology, information and current knowledge for every one of their student, residents, graduate and physicians under their employ at no cost by secure 24/7 online access.

Given such a source of current medical information, indexing the relational database could offer an ongoing differential list of diagnostic possibilities for each-and-every sign, symptom and laboratory finding. Despite advances in medical science, missed diagnosis continues to plague the profession. A correct and timely diagnosis seems more likely with instant access to lists of all the possibilities, criteria for diagnosis and brief summary. Statistical analysis of those lists should be ongoing and real-time offering simple probabilities for both single positives and combinations. This statistical process will be essential for assimilating genomic data and applying it to personal medical care.

Information technology offers an unlimited repository of knowledge accessible through a relational database. The computer never forgets. It lends itself to statistical analysis, but it does not think.  Thinking is the job of the student and the physician. That critical clinical thinking and the basic sciences remain the educational challenges of the medical school. Medical information is so vast and so rapidly changing, however, that it has long since grown beyond the capacity of any one physician to learn and forget much less to remember.

Today in the real world the physician is met with time constraints, productivity demands and repetition, all of which discourage discovery and lead to missed diagnosis. Diseases and treatments fall into familiar patterns. A hundred and fifty, or so, conditions fall easily into a recurring pattern of diagnoses. A one-page encounter form can cover the needed ICD codes required for insurance. That recurring pattern, however, erodes away the physicians heard earned clinical acumen. We often overlook rare disease possibilities and there are so many of them with new discoveries all the time. Additionally, increasing numbers of Immigrants bring in problems common to their home country but only now cropping up here in the US.
Physicians keep up to date with expensive seminars and long hours reading expensive journals at home but nonetheless slowly fall behind. The older a physician grows, the greater the clinical judgment but the more he or she forgets. The content shrinks.

Traveling to a conference or accessing journals presents problems of time and money for a rural physician. A limitless source of medical information at the fingertips would help. The generalist must fill the gaps between specialties and cover a broad spectrum of medical challenges. A total medical information system would help met that need. Proprietary systems like Epocrates provide some of the needed medical information. None of the proprietary systems, however, list everything, nor do they have the ability to stay current. Furthermore, marketing and often greed motivate the content; they emphasize drugs and treatment more than diagnosis. Medical schools need to provide an umbilical cord for the student and to the lifetime of the graduate physician regardless of the specialty. We have an obligation in medical education to lead the way and to a pursuit of excellence.


[1] A word of caution, often overlooked by non-clinical educators, information should be limited to medical students and graduate MDs --- those with the ability, education and dedication to care for patients. Providing that information to various assistants and alternative providers will cut off the supply and even the existence of Primary Care Physicians. Another mandate that should go without saying but today is often ignored; physicians must all freely exchange information techniques and knowledge between one another.
 

Friday, April 26, 2013

Gestational Diabetes Screening, who’s “Best Evidence”


Share |
The National Institutes of Health consensus panel has determined that the one-step screening test for gestational diabetes results in greater numbers of positive outcomes than the traditional two-step method. The one-step requires a 3-hour glucose tolerance test. The traditional method specifies screening with a non-fasting glucose challenge testing at one hour. Only the positives then go on to the 3-hour GTT. The one-step approach results in 15% to 20% positives, whereas the traditional two-step protocol results in only 5% to 6% positives.[1] The panel concluded that the current data do not indicate whether the one-step protocol improves outcome or diagnostic accuracy.

From the data, one might wonder if the non-fasting screen misses significant numbers resulting in fewer diagnoses of gestational diabetes – a matter of sensitivity. Alternatively, the 3-hour test may lack specificity resulting in false positives. Looking further in the literature,[2] O’Shea and O’Connor in Ireland argue in favor of an HbA1c in the second trimester.    The American College of Obstetrics and Gynecology recommends screening by history, risk factors or the non-fasting 50g glucose challenge.[3] The US Preventive Services Task Force argues that there is insufficient evidence for or against anti-natal screening for diabetes. Mayo Clinic states that history and risk factor screening may be sufficient for women under 25.[4]
Obviously, the increasing obesity among young women poses a threat. Hospitals have broadly adopted the one-step approach. One wonders if the hospitals favor the more expensive option with higher numbers of positive results for business reasons. You might conversely ask if the US Preventive Services Task Force leans in the opposite direction for reasons of cost containment.
There is nonetheless a consensus that the definitive test for gestational diabetes is the 3-hour GTT with proven correlation to outcome. Is this not an argument in favor of professional judgment in selecting when to do the GTT rather than elusive best evidence? --- Who's best evidence and who’s viewpoint?

 




[1] http://tinyurl/c642s5e
[2] www.ncbi.nlm.nih.gov/pubmed/22838107
[3]http://www.acog.org/Resources%20And%20Publications/Committee%20Opinions/Committee%20on%20Obstetric%20Practice/Screening%20and%20Diagnosis%20of%20Gestational%20Diabetes%20Mellitus.aspx
[4] http://www.mayoclinic.com/health/gestational-diabetes/DS00316/DSECTION=tests-and-diagnosis