+
Related Flashcards
Related Topics
Cards In This Set
| Front | Back |
|
The upper limit of normal of the fasting blood glucose for diagnosing diabetes mellitus has been lowered from 140 mg/dL to 126 mg/dL. What effect does this change have on diagnosing diabetes mellitus?
|
*Increases the predictive value of a negative test result(Increases NPV), as we made the test more sensitive you will have less FN (You cantch more people with disease)and more TN(you also increased number of FP so PPV and specificity go down)*formula for calculating the NPVis TN/TN + FN. Tests with 100% sensitivity always have a 100% NPV; therefore, a negative test result excludes disease.The less the FP rate, the greater the specificity of a test. A test with no FPs has 100% specificity. The predictive value of a positive test result (PV+) is the likelihood that a positive test result is a TP rather than a FP; therefore, the formula for PPV is TP/TP + FP. Tests with 100% specificity always have a PV+ of 100%; therefore, they are most useful for confirming disease.
|
|
*Study is designed to evaluate the feasibility of treatment X in children with chronic headaches.dozens of children with chronic headaches are recruited for the study,In addition to their usual therapy, all children are treated with X three times a week for few months...Type of study?Susceptible to what type of bias?
|
*This type of PROSPECTIVE Study is called CASE SERIES-Note that in this medical research you track group of people with similar exposure(here Treatment X) to establish correlation between exposure and outcome, HOWEVER note that it Has No Control-group(Hence has Low Internal Validity>As you selectively pick particular group based on their exposure, Case series studies are especially susceptible to SELECTION Bias.*AT LEAST you should know that case series/case report studies are type of DESCRIPTIONAL studies at the levels of Individuals/small groups and that one of the example of case report studies was study of Young individual males with PCP that eventually lead to discovery of AIDS.If you don't get anything here remember to do this on examp: Ask yourself does the study compare AT LEAST 2 DIFFERENT groups(Or more)? if the answer is NO then you can pick case series/case reports(These 2 are kinda same case series basically will pick Men who use IV drugs as a single group and report it, while case reports will report them individually, point is you do NOT compare them to people who are Not IV drug users, you just check what is going on in those cases).
|
|
*you organize subjects into smokers and nonsmokers and note who gets COPD.*Uncontrolled hypercholesterolemia and Controlled<you note who gets MI.type of study?What info/question answer it gives you?
|
*Prospective Cohort study(you want to know who will get the disease), if you divided participants into exposed vs non-exposed group but checked who has already developed the disease than it would've been "Historical" Cohort study.(Like check who developed cancer:Radiologists vs Other physicians, exposure was radiation)In other words Cohort study gives you information about Relative Risk(RR)=Probability of event occurring in exposed/Probability of event occurring in Non-exposed group-Basically tells your HOW MUCH MORE likely those who were exposed to Risk will develop/developed the disease.
|
|
*You look at radiologists vs other physicians and note who gets cancer, in the future...vs*You look at physicians who got cancer and check how many of them were radiologistsvs*You look at radiologists and other physicians and note who has developed the cancer.
|
*PROSPECTIVE cohort you compared 2 groups and calculated relative risk(how much more likely those who were exposed to radiation would develop cancer-Relative risk)vs*Case-Control study-you checked physicians with cancer and then checked how many of them have been exposed to radiation(By being radiologist), so here you basically calculated Odds Ration(OR)-what were the odds for physician who has cancer to have been exposed to radiation(Radiologist)when compared to physicians who got cancer but weren't exposed to radiation(weren't radiologists),so here you could say that those physicians who had cancer were more likely to have been radiologists.vsHistorical cohort study- you checked if the exposure to radiation(being radiologist) affected the likelihood of getting the cancer, by basically comparing probability of radiologists developing cancer to probability of Non-radiologists developing cancer, and this is historical because you note Who already has developed the cancer(Not will develop the cancer), this ratio of probabilities gives us Relative Risk of radiologists to get cancer when compared to other physicians(Contrast this with odds ratio which would tell you what were the odds that physician who has developed the cancer would have been radiologist)
|
|
Your group went out in streets and asked people 2 questions:1)Do you eat Junk food?2)Do you have Abdominal pain?what type of study does this resemble?*You established that 100 out of 100 people who ate Junk food also had complains of abdominal pain....<Can you say that Junk food Causes Abdominal pain?
|
*You basically assessed what happens at this moment not years after or before the start of study but right now how many people who eat junk food also complain of abdominal pain?That Best fits Cross-sectional study description...even if all 100 out of 100 who ate junk food also complained of abdominal pain you ABSOLUTELY Can NOT say that Junk food causes abdominal pain(Just as you can't say that abdominal pain leads to people eating junk food, it can be either one of these 2statements), all you can say is that there is a CORRELATION between these 2 but you can NOT establish Causality<Highly tested characteristic of cross-sectional study.
|
|
If the alpha is set at 0.01 instead of 0.05.what will happen to likelihood of making type 1 error?
|
*Type 1 error-is basically false positives(Stating that there is an effect or difference when none exists=null hypothesis incorrectly rejected in favor of alternative hypothesis), and P-value corresponds to this chance(alpha), obv. lower the value(P<0.01 instead of 0.05) the higher is your confidence in reporting the significant findings.
|
|
A test is performed on 5 normal people who have the following test results: 49 mg/dL, 54 mg/dL, 51 mg/dL, 56 mg/dL, and 60 mg/dL. The standard deviation of the test is 4 mg/dL. What is the reference interval (normal range) for the test?
|
*First find Mean=49+51+54+56+60=260/5=52.*Normal range is +-2SD, hence 52-4x2=44,52+4x2=60.<So normal range here is 44-60
|
|
*New blood test to detect prostate cancer is evaluated in 400 male volunteers. A needle biopsy of the prostate gland is done on all men with serum prostate-specific antigen concentrations greater than 5 ng/mL (N<4). One hundred men undergo biopsy procedures; 90 are found to have prostate cancer, and five are found to have chronic prostatitis.*What you need to evaluate sensitivity?
|
*Prostate biopsies of men with test results equal to or below 5 ng/mL.(To evaluate False Negatives)*Remember sensitivity equals TP/TP+FN, we know TP (90), so now we need to find out FN(So on test <5 means that they are negative, but if biopsy shows that they indeed have cancer=False negative)
|
|
*1,2,3,4,5,6,7*Scientist stole 7th result of experiment.How would Median,Mean,Mode will change>?
|
*Well first Mean was 1+2+3+4+5+6+7/7=(4), now mean is 1+2+3+4+5+6/6=3.5*Median was 4 (just pick the number in middle of first line), Median for second case would be mean of middle 2 as you can't pick 1 middle number, so here it would be 3,5.*Mode won't be affected here as it is simply the value that occurs most often in data(So for example in our cases if 3 for example was repeated 2 times-more than any other number, it would be the mode, so even in that case mode wouldn't be changed as still the most repeated value is 3).*So in the end Mean and Median would decrease however Mode won't change.
|
|
*from 1000 which receive placebo, 30 of them die*From another 1000 which receive drug X, 20 people die*This observations are made over 3 years.*Per year the drug cost Per patient is 500 bucks for each patient.*How much money should be spent to save 1 patient?
|
*Basically drug saves 10 more people when compared to placebo.*So to save 1 guy yo need to treat at least 100 guys.*That means you have to spend 100 x500=50000*And that over 3 years.*50000x3=150000.Pretty important card believe me.
|
|
*Study is conducted to assess the activity of a new drug in patients with disease X.All patients enrolled in the study are informed that they would be treated with the drug X, they are assigned to successive dose cohorts of 400 to 1000 mg/day of the drug.*Researches monitor the results of the groups(Divided by doses they get).This study is best described as?
|
*Open-labeled Clinical trial-this is the trial when BOTH researchers but also participants know who receives what treatment.*So here both researchers and the participants know that they get drug X.
|
|
*Particular disease is more common in dizygotic twins than in monozygotic twins.what does that tell you about environmental vs genetic influence on that disease?
|
*That Environmental influence plays greater role, remember ONLY MONOzygotic twins are IDENTICAL genetically.<This is Twin concordance study.Don't confuse with adoption study-where Siblings are raised in different families(Can be biological vs Adopted family), but that study too can give you idea about genetic/environmental influences on trait/disease.
|
|
*Study that determines PREVALENCE of Heart disease in 2017 is what type of study?
|
|
|
*Study divides patients in 2 groups those on doxorubicin and those on placebo and compares the risk of developing dilated cardiomyopaty over next few months...
|
*Cohort PROSPECTIVE study, you want to check RR(How much more Likely were the patients on doxorubicin to develop Dilated cardiomyopathy over those who didn't receive that drug)
|
|
*Phase 0 clinical trial?
|
*Micro-dosing studies-when around 10-15 people receive Lower than normal doses of drugs, just receive information about PharmacoKinetics(What your body does to drug) and Pharmacodynemics(what drug does to you,mechanism of action),subtherapeutic doses(used in phase0) won't let you know how effective/toxic is this drug.
|
