14 ⼀BIOSTATISTICS/BIOETHICS Flashcards

Question

How should physicians approach [Complementary and Alternative medicine] with patients? (2)

Answer 1

1. have open, explorative and collaborative discussion 2. adequately research *risk, benefit,* and *[**proven** treatment alternatives] for each CAM*

Answer 2

- Excludes non-compliant participants = **analysis** of only data from participants who've followed rules **Per Protocol** - estimates true effect of an intervention but overestimate effect of intervention in a *real world* setting vs - opposite of [Intention-To-Treat analysis (which keeps participants in their allocated groups regardless of dropout or non-compliance = better estimate of real world)]

Answer 3

▶after subjects are initially randomized to tx vs control, those subjects are maintained to that intervention group regardless of what they do during study period [switch intervention/drops out]→ crossover/attrition …actually become apart of the subjects outcome data and are attributed to the intervention ▶ will provide *conservative* *(but more valid)* estimate of the intervention effect ▶preserves benefits of randomization in superiority trials

Answer 4

SPeci**F**icity= [true negative] = [TN / (TN + FP)] = *used for test con**F**irmation* = *"a test's probability (in the absence of disease) a patient test negative"*

Answer 5

evaluates association between [**1** quantitative dependent variable] and [≥2 independent variables of interest] while controlling for adjustment variables

Answer 6

*statistical significance [α = 0.05] means…* ## Footnote Researchers are willing to accept that there's a LOE5% chance that the observed differences they see will be considered significant when no difference actually exist

Answer 7

MD should maintain confidentiality of sexually active teens and ensure teen access to healthcare and counseling ***MD MUST BREAK CONFIDENTIALITY IF THERE'S RISK OF HARM TO SELF/OTHERS OR CHILD ABUSE***

Answer 8

❗️In pts who **LACK DECISION-MAKING CAPACITY** ❗️but who require 9-1-1 Tx ❗️can receive 9-1-1 tx per *[Doctrine of Implied (_NO_ _explicit_ necessary) consent]* ^{(and must be tx reasonable people in similar situation would expect)}

Answer 9

-it doesn't just account for # of events for both the [control group] and [treatment group] ... -but also the **TIMING OF THOSE EVENTS** throughout the follow-up period for these 2 groups. -This produces [TIME-TO-EVENT/DEATH] for the [control group] and [treatment group], which may reveal overall superiority between the 2

Answer 10

overestimation of **survival** only solely because patients were diagnosed earlier \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ ## Footnote *when a screening test appears to prolong survival when actually it's just informing the patient of their diagnosis earlier on* ...(and as a result...) time from diagnosis until death is longer ⼀even though there isn't actual increase in survival (just earlier detection)

Answer 11

Number of people Needed to be exposed to a tx before a harmful event occurs \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ NNH =[1 / ARI] = **{1 / [(AER^TX) - (AER^PLACEBO)]}** *AER: Adverse Event Rate | ARI: Absolute Risk Increase*

Answer 12

seeks to determine whether intervention works in real life conditions \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ retrospective observational study in which subsets of controls are matched to cases and analyzed for variables of interest

Answer 13

BEFORE MAKING ANY CHANGES, Specialist should call/discuss plan with PCP since **PCP has primary responsibility for patient's care**

Answer 14

LRs does *not* change as disease prevalence changes and LR can be used to grade clinical significance of various results when \>2 different test results are possible \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *LR = probability of a given test result occurring in a patient with a disorder compared to the probability of the same result occurring in a patient without the disorder*

Answer 15

> \>65 > > * * * > > [age\>80], female, physical impairment, psych impairment^{(dementia/depression)} > > * * * > > 1. spiral fx > 2. abrasions in unusual locations > 3. malnutrition > 4. pressure ulcers **interview patient alone and alert [Adult Protective Services] if elder abuse suspicion supported**

Answer 16

*test with:* ⇪ se**N**sitivity = ⬇︎false **N**egatives \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ ⇪ s**P**ecificity = ⬇︎ false **P**ositives

Answer 17

1. HDUS 2. PO INtolerance 3. PO meds failed 4. \< 2 yo

Answer 18

1. ✔︎ [Ok to treat temporary or correctable conditions (i.e. sepsis)] ## Footnote ----but NO **VAP** ------ 2. NO **V**entilator dependency 3. NO **A**CLS 4. NO **P**rolongation of terminal illness

Answer 19

[(mean) +/- (SD)] = [(external SD)% chance \> (mean + SD)] and [(external SD)% chance \< (mean - SD)]

Answer 20

TRUE \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *especially as to clarify medical decisions*

Answer 21

**▨ -pooling data** from several single studies in order **to INC** **power** of 1 LARGER COLLECTIVE study *(as to INC that LARGER COLLECTIVE study's ability to detect (if one exist) a difference in outcome between groups)* ▨ If an outcome is rare (or difference between groups is super small), this will be difficult for single small study to detect ➜ unlikely to reach statistical significance = use meta analysis \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ ◮ _pooling from several studies will also pool their biases and limitations all into 1 analysis_

Answer 22

**PSTQ = P**revent/**S**creen/**T**reat/**Q**uery ## Footnote [**P**rimary = **P**revent future disease (* _P_revent future MI by reducing HLD*)] [**S**econdary = **S**creen possible disease (*_S_creen possible cervical CA with pap smear*)] [**T**ertiary = **T**reat current disease PATIENT CURRENTLY HAS to ⬇︎ disability from that current disease (*_T_reat current severe CAD with CABG revascularization*)] [**Q**uartenary = **Q**uery statistics of disease: to identify pts at risk for unnecessary treatment and prevent redundant or unnecessary therapies (*shared EMR limits unnecessary or repeat procedures*)

Answer 23

seNsitivity = [true positive] = [TP / (TP + FN)] = *"a test's probability (in the presence of disease) a patient test positive"*

Answer 24

sPecificity = [true negative] = *Negative In Health* ## Footnote *= "a test's probability (in the ABSENCE of disease) a patient test negative"*

Answer 25

⭐Range of values in which a [specified probability (usually 95%)] of the [means of repeated samples] are expected to fall within ⭐CI = mean +/- Z^{(Standard Error of Mean)} *(ex: [95% CI, Z=1.96] and [99% CI, Z=2.58]* ⭐Statistical difference: a.[95% CI for (mean difference between 2 variables)] that includes 0 **=** [NSD = H0 not rejected] b. [95% CI for (odds|relative risk ratio)] that includes 1 **=** [NSD = H0 not rejected] c. CI of an [Odds Ratio] indicates (with a certain confidence level) whether a given OR is statistically different from [null OR = 1]. a CI that excludes [null OR = 1] **is** statistically significant ## Footnote H₀ = null hypothesis = [NO statistically significant difference exist]

Answer 26

| *if relationship is seen in populations outside study also??*

Answer 27

FALSE * * * - Pediatricians dating their patients' family members is ethically PROBLEMATIC. - Physicians are ethically obligated to act in best interest of patient .. so intimate involvement may blur judgment = - **these pediatricians need to transfer patient care**

Answer 28

p ≥ 0.05 means the study groups are **NOT** statistically different

Answer 29

PAE = preventable event 2/2 faulty human **P**ractice **N**AE = preventable event 2/2 **N**ot adhering to Standard of Care

Answer 30

*present* ; ([Group with Risk Factor] v [Group wihout Risk Factor]) ; **PREVALENCE of a disease** \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *How prevalent is this [disease and its risk factors] right now?**?*** \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ Collects data from a group of people to determine [frequency of disease and its risk factors] at a specific point in time *\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_*

Answer 31

when pts modify their behavior just because they know they're being studied

Answer 32

Since these influence professional judgment, it is Unethical to accept patient gifts that are \_\_\_X\_\_\_ 1. given with expectation of preferential tx (*vs genuine tokens of appreciation)* 2. LARGE 3. EXPENSIVE 4. intimate

Answer 33

If given written authorization, HIPPA allows MDs to give the **minimum necessary information** to satisfy the employer's request

Answer 34

HR = the likelihood an event will occur in a treatment group (relative to the control) [not likely/protective] \< HR 1.0 \< [LIKELY/DETRIMENTAL] \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ [EVENT less LIKELY to occur in the treatment group than the control group] \< HR 1.0 HR 1.0 \< [EVENT MORE LIKELY to occur in the Treatment Group than the control group]

Answer 35

selection bias that → unrepresentative sample

Answer 36

how generalizable the results of a study are to other populations ## Footnote *(i.e. a study for middle aged women has low external validity for elderly men)*

Answer 37

<0.0**5** \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ 9**5**% | <0.0**5**

Answer 38

▶compares the mean between ≥3 groups \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ ▶*requires [quantitative dependent variable (outcome)] and [quaLitative independent variable (exposure/risk factor)]*

Answer 39

▶adjusted measure of overall mortality *-------*▶▶*by calculating --------* [observed # of deaths in population of interest] ➗ [expected # of deaths in population of interest] \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ SMR of 1.75 = *"observed # of deaths (in this population of interest*) *is 75% higher than expected"*

Answer 40

of patients needed to treated in order to [prevent 1 disease] or [cure 1 condition] = measures efficacy of a therapy \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ NNT = [1 / ARR] = **{1 / [(failure rate_PLACEBO) - (failure rate_TX)]}** *ARR: Absolute Risk Reduction*

Answer 41

AD (living will, DNR/DNI) are invoked **ONLY WHEN PATIENT LACKS DECISION-MAKING CAPACITY** If patient still retains Decision-Making Capacity, M.D. should still ask patient if they want to refuse intervention (like Intubation) in the treatment of temporary / reversible illness (like PNA).

Answer 42

MD altogether should **avoid soliciting financial support from patients** (espeicially for personal endeavor) *talking about personal experiences is fine, as long as its relevant to patient*

Answer 43

They use (1-specificity) so [⇪ (1-specificity) by 12%] actually = [**⬇︎ specificity** by 12%] \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ moving up-and-right on ROC = [⇪ N but ⬇︎P]

Answer 44

*PCPAF* ## Footnote **P**atient **C**linical_Status **P**lan of Care/Summary with anticipated d/c **A**nticipated problems with recommended mitigation **F**ollow ups/Pending Actions/TO DOs

Answer 45

1. IRB requires **ADDITIONAL OVERSIGHT** than gen pop (to ensure fairness to this vulnerable population) 2. otherwise, same rights to refuse or consent to participate

Answer 46

1. **ERRORS DISCUSSED WITH PATIENT AND FAMILY IN TIMELY MANNER = OLBIGATION!** 2. Empathetic apology 3. Explanation 4. Event prevention discussed

Answer 47

**REPORT SUSPECTED (*****SEXUAL*****) MISCONDUCT** TO STATE MEDICAL BOARD FOR INVESTIGATION = ETHICAL OBLIGATION \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *advising your colleague to terminate their physician-patient relationship is NOT ENOUGH*

Answer 48

```ratio between (probability of event^ET : probability of event^cp)] ``` RR = [I^ET/ I^cp] *RR = [Incidence of disease in (Exposed/Treated)* *group**] / [Incidence of disease in (control/placebo)* *group**]*

Answer 49

TRUE * * * ⼀(True ***if there is standardized proof of need -** but note: waiving copays for _ALL_ patients = insurance fraud due to total charge discrepancy and potential inducement of unnecessary medical services)*

Answer 50

⭐detecting a difference between groups when a difference does NOT in fact exist lol or (AKA) ⭐: Mistakenly rejecting [a null hypothesis] … that is actually true and should NOT be rejected = false positive(*since rejecting = positive*) ⭐{probability of [Type 1 error] occurring = α } and this reflects significance of a test

Answer 51

**cOIN** **c**entral tendency graph = 1: m**O**de 2: med**I**an 3. mea**N**

Answer 52

Odds Ratio [(a**D**) / (bc)] *[(**D**ISEASE^OOE)/ (control^OOE)]* \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *• OOE = Odds of Exposure* *• first draw contingency table | • [(OR \>1) indicates there IS an association between a **D**ISEASE and the [exposure or risk factor] in question*

Answer 53

For (Event J)… *Case fatality rate* = [proportion of **people with (Event J)** who die from (Event J)] \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *Mortality rate* = [proportion of general population who die from (Event J)]

Answer 54

[solid employees \> half time ≥1y] of [Large companies \> 50 employees] have legal federally protected [≤12 weeks *unpaid* leave] if : - postpartum - serious illnes - caring for **immediate family** with serious illness

Answer 55

*describes* [**Direction** {neg inverse} vs {positive direct}] *and* [**STRENGTH** (-1 0 1) - furthest from 0 = stronger] * of two quantitative variables to each other* * \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_* * [{r\<0} = {neg inverse} = {A⇪ ⼀B⬇︎}]* 🆚 *[{r\>0} = {pos direct} = {A⇪ ⼀B⇪ }]*

Answer 56

**prevalence** of disease; obtain snapshots = examines associations between risk factors and outcomes in a well-defined population at one **particular time** =takes snapshot of a population and measures freq of risk factors and outcomes simultaneously *although this leaves out temporal relationship between the two*

Answer 57

1. Uninsured pts who don't qualify for [Medicaid federal health insurance] **but** also can't afford [private health insurance] – can purchase insurance through the [*ACA* Health Insurance Marketplace Exchange] * * * 2. *ACA* asserts all insurance plans adhere to 3 guidelines: a. can not deny coverage/INC premium due to preexisting health condition b. must cover*^{full or partial}* the [10 **essential health benefits** (i.e. Rx, mental health)] ⼀pt may incur copay on certain benefits c. must cover*^FULL*100% all [**Preventative Health Care**] services = $0 copay * * * 3. *ACA* applies to both HMO and PPO \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *HMO: Health Maintenance Organizations / PPO: Preferred Provider Organizations*

Answer 58

Hospital ETHICS Committee

Answer 59

✏️IRB = *Institutional Review Board* ✏️primary authority on human research ✏️that approves research protocol initiation and/or modifications

Answer 60

⇪ Prevalence = ⇪ PPV and ⇪ Prevalence = ⬇︎ NPV

Answer 61

**P**CD = [**P**rooF|ConFirmation|Diagnostic] \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ s**P**eci_F_icity= [true negative] = [TN / (TN + FP)] = used for **P**CD test = *"a test's probability (in the absence of disease) a patient test negative"* ## Footnote **P**CD = [**P**rooF|ConFirmation|Diagnostic]

Answer 62

TRUE * * * ***For [Preventative Health Care (i.e. screening mammogram)],** Federal law _mandates_ ALL insurance plans **cover 100% the cost** so that patients incur $0 out-of-pocket co-pay!*

Answer 63

1. Patients must be notified **IN WRITING** about: 2. breach mechanism 3. information disclosed 4. actions taken to [fix current breach] 5. actions taken to [prevent future breach] \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *breach = disclosure of patient info to unauthorized person or for improper purpose*

Answer 64

EMTALA = [Emergency Medical Treatment Active labor Act] enacted by Congress 1986, imposes [3 (*SSS*) primary requirements of Hospitals to EVERY patient before discharge (regardless of payment)]. Hospitals MUST : 1. *SCREEN* patient- medical exam 2. *STABILIZE* patient prn 3. *SEND* patient/transfer ***only after patient stabilized***

Answer 65

measures **inter-rater RELIABILITY (*or Concordance*)**. represents the extent to which inter-rater agreement is an improvement on chance agreement alone.

Answer 66

[compares *MOQV^{Group 1}* to *MOQV^{Group 2}*] (*I.e. comparing mean of a construct score between 2 intervention groups* \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *MOQV: [mean (of the quaNtitative variable)]*

Answer 67

\<0.010 \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ 99% | \<0.010

Answer 68

**survival analysis** that [real-time analyzes⼀throughout the study] period from [entry into study] to the [event of interest (*death*)] and compares the rate at which [events of interest] occur for different participants. - HR - CI that EXcludes [null = HR = 1] is statistically significant

Answer 69

a sample is unrepresentative of the target population (this leads to incorrect measures of association) ; Surveys / Polls \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *In Survey/Polls, nonresponses may cause selection bias, as responders may have responded 2/2 a particular bias = sample group is unrepresentative of target population*

Answer 70

acting in pts best interest; do no harm \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ * ONLY IF no other physician is available.. you can give * **ACUTE** care ONLY * **LIMITED** care ONLY ## Footnote \*\*_writing a prescription automatically establishes a medicolegal physician-patient relationship; making MD liable for any/all outcomes"_ *"I would like to help you, but I am uncomfortable prescribing for someone I am not treating."*

Answer 71

diagnostic accuracy; Greater AUC = Greater Accuracy and discrimination (outperforms)

Answer 72

- occurs in the absence of blinding, - observers misclassify data due to preconceived expectations about the tx - usually occurs when studied outcome is QUALitative

Answer 73

NLR⊝ = [(1-N) / P] \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ INDEPENDENT ## Footnote "**N**umber **1** **N**igga, **P**ositivity".... **P**ositive LR = se**N**sitivity / (**1** - s**P**ecificity) **N**egative LR = (1 - se**N**sitivity) / **P**

Answer 74

[ecologic correlational study] ; population-level information

Answer 75

- patients with untreatable disease and poor quality of life, who've already tried different therapies, may sometimes refuse to accept *any* types of treatment (including general). Under principle of autonomy, this is ok. * -Unless court rules pt is incompetent*, **principle of autonomy** includes a patient's _right to refuse treatment_, which must be respected

Answer 76

⬇︎Warfarin by 25% (since amiodarone inhibits CYP2C9)

Answer 77

modifying sensitivity of a study's criteria .. and then repeating the initial analysis ⼀in order to determine if those modifications played a significant role initially

Answer 78

[progressive benign] disease cases have LONGER lifetime duration -\> they are more likely to be detected **incidentally** by a screening xm -\> artificially inflates the "detection success" of that screening xm

Answer 79

kickback; **UN**acceptable \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *kickback = financial incentive paid to referring physician for each referral sent*

Answer 80

▶occurs when magnitude or direction of the [independent variable's effect] on the [dependent variable outcome] varies based on level of a third variable = *EFFECT MODIFIER* ▶ Separate Stratified Analysis should be conducted for each level of the *EFFECT MODIFIER* * * * Example: “Study showing Vitamin D's outcomes on Cancer Mortality in patient groups separated by BMI” - independent variable = Vitamin D - dependent variable = Cancer Mortality - *EFFECT MODIFIER* = *BMI^{[<25 / 25-29 / ≥30]}* * ⼀each level (i.e. BMI\<25, BMI 25-29, BMI ≥30) will have separate stratified analysis for [independent Vitamin D's effect] on [dependent outcome Cancer Mortality]*

Answer 81

Doctors have ethical duties to prioritize patient global well being (beneficence) while also [**AVOIDING** (either directly or indirectly by inaction) causing unwarranted **patient harm**] (nonmaleficence) *(ie so...DOCTORS must EXPLICITY and clearly state recommendations against or in support) Patient autonomy does NOT extend to the recommendations by the Doctor*

Answer 82

**life threatening emergencies** ▶i\ [inadequate time]^{*to obtain consent*} _or ▶i\ [inadequate capacity ^{of patient's mentation}]^{*to obtain consent*} | *🔎i\ = in the setting of*

Answer 83

*[past (retrospective)] and [future (prospective]* ; ([group WITH Risk Factor] v [group WITHOUT Risk Factor]) ; [**INCIDENCE (development)]** of a disease \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ {Analyzes records between *[GWF] and [GWOF]* to determine... 🔲 ⛞*retrospective = who **developed** the disease?* OR ⛞*Prospective = Who **WILL develop** disease?* ## Footnote _{*"Smokers (exposed/risk factor group) had higher RISK of developing COPD than Nonsmokers (NOT exposed/NO risk factor group) had" -- in the Prospective cohort study*}

Answer 84

HIGH SPECIFICITY High Negative Predictive Value

Answer 85

A: [c**A**se report] , [c**A**se series], [c**R**oss-sectional] ## Footnote \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ B: [_{Observational}*descriptive*] studies describe disease patterns by time, place and person and **GENERATE HYPOTHESIS** about the association between risk factors and disease * (i.e. study showing high proportion of people with lung CA also smoked cigarettes)*

Answer 86

For minors, in **NON**-emergent situations, **informed consent should be obtained from parent or legal guardian** (written \> verbal) before intervention \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *if 911 situation: consent does NOT need to be obtained before intervention (as long as reasonable attempts to contact parent/legal guardian are made)*

Answer 87

Case-Control Study \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *allows quick localization of outbreak source*

Answer 88

have family establish [primary POINT OF CONTACT who will communicate concise questions/concerns from family] to careteam

Answer 89

* STATEWIDE interoperable *(clinician, pharmacy, inpatient, outpatient)* database with patient's **complete** controlled substance hx * accessed by clinicians to avoid duplicate Rx and to identify high-risk drug use * clinicians accessing PDMP database (*for patient care)* = exempt from HIPPA = **does NOT require patient consent**

Answer 90

Blinding ## Footnote (Observer bias = Observer [researcher] alters elements of the study **(like over reporting a dz)** either consciously or subconsciously)

Answer 91

1. Surgical (operation/procedural) 2. [**MEDICATION** = MOST FREQUENT in non-surgical population (1/5 of all Adverse Hospital Events)] 3. Gen Care 4. Hospital-acquired infection

Answer 92

If sharing a patient's PHI with family can/will resolve an _imminently_ modifiable risk to those family members = **potential duty to relatives**

Answer 93

* IV = c**ausality** (did your [independent variable-EFFECT] *Cause* the ∆ in the [dependent variable-OUTCOME] you observed? * If Yes = [⇪ Internal Validity] 👍 * If No ➜ [solution = double blind]

Answer 94

* EV = **GENERALIZABILITY** ( is the observed relationship seen in the study GENERALIZEABLE (meaning can we see the SAME observed relationship *OUTSIDE THE STUDY-IN THE REAL WORLD*? * * If Yes ➜ ⇪ External Validity)

Answer 95

[_*(some)* c**R**_Poss-sectional]^{*uses Risk Factors to determine Disease Prevalence [who HAS disease right now]*} \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ [c**A**_Rse control]^{*Determines Risk Factors for a Disease*} \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ [C**O**_DHORT R/P] ^{*Uses Risk Factors to determine who DID and who WILL develop Disease*} ## Footnote \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ [*Analytical* observational studies] **TEST HYPOTHESIS** about association between risk factors and disease *(i.e. study determining that smoking cigarettes is a statistically significant risk factor for Lung CA)*

Answer 96

Physicians are ethically obligated to report colleagues (to State Medical Boards) who are * impaired * incompetent * unethical * or who [subject patients to potentially harmful tx]

Answer 97

▶Confounder = **extraneous variable** a/w both the [dependent variable-OUTCOME] ... and the [independent variable-EXPOSURE] ...... individually ▶If Study does not correct for the potential skewed effect these confounders may have (*Randomization helps to remove confounding variables*) = Confounding Bias

Answer 98

1. smaller sample size 2. more precise estimates of treatment effect than studies with parallel design - (since crossover = each subject gets ALL TREATMENTS = acts as their own control) (*versus* *parallel = subjects receive single treatment thoughout entire study*) 3. takes longer to complete (*since subjects have to move from 1 phase to another multiple times during the study*) 4. carryover effect (*1st treatment effect carrysover into the 2nd treatment phase - can be avoided with lengthy washout*)

Answer 99

sampling pts in a NON-random manner -\> lky to exclude certain members of the target population

Answer 100

TRUE ⼀physicians are **NOT** ALLOWED TO ACCEPT travel expense reimbursement from pharmaceutical companies \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ …*unless*….they are the [faculty physician **LECTURER**] ⼀who can accept reasonable honoraria and/or travel reimbursement

Answer 101

allows patients that are both [infirmed (weak in health)] and [*homebound\**\*] to receive**skilled** care at home \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ ## Footnote * \*\*homebound = ≥1 of* * -[uses mobility device (wheelchair/cane/walker)]* * -can only leave home with assistance* * -leaving home medically contraindicated*

Answer 102

^{***"PCPAF"***}systematic template checklist

Answer 103

- [(*r*) correlation coefficient] range from {-1 to +1} with [0 = NO CORRELATION AT ALL]. - The closer to +1 = VERY *POSITIVELY* CORRELATED - The closer to -1 = VERY *NEGATIVELY* CORRELATED

Answer 104

d. [deductible = (Out-Of-Pocket **minimum**)] = **min** patient has to pay before health plan starts to cover *any* part of health expenses e. [Out-Of-Pocket MAX] = **MAX** patient has to pay before health plan starts to cover *ALL* (100%) of health expenses

Answer 105

exposure and outcome * * * RR = observational / experimental / follow up studies / cohort OR = case-control / cross-sectional studies

Answer 106

**DP dRIVRS**

Answer 107

10-20 ; ≥45 *note: Chelation does not improve cognitive complications*

Answer 108

SIR describes if occurrence of a disease in a small population is high or low relative to an expected value derived from a larger comparison population \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ SIR = [observed cases] / [expected cases]

Answer 109

acting in pts best interest; do no harm \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *disadvantages friend/family from MD possibly \_\_\_* 1. treating conditions beyond expertise/training 2. lack of follow up and documentation 3. inadequate assessment of sensitive H/P

Answer 110

*Interview Patient ALONE*

Answer 111

*Interview Patient ALONE*

Answer 112

Compares tx groups in a randomized trial by including 1. [all subjects], 2. [their first (randomized) tx group allocation] and **3. any personal acts of attrition (dropout, loss to f/u, crossover)** to overall benefit of the drug = conservative estimates of effect if acts of attrition are significant, however, *results will better reflect expectefd effect in practical clinical (i.e. real world) setting*

Answer 113

★ [interdisciplinary ⬇︎of unnecessary medical interventions] while [**⇪ quality of life** for terminal/seriously ill patients (and their family)]. ★ Can occur concurrently with life-prolonging tx

Answer 114

**RQ1. [PATIENT (*****or Patient's Surrgate*****) FOREGOS ALL _CURATIVE_ TX]** **RQ2. prognosis LOE6mo** \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ 3. interdisciplinary palliative team 4. Patients can leave hospice for curative tx then return later if desired 5. [comorbid conditions /non-curative interventions are OK] 6. [Full informed consent from patient is NOT required (*since, if needed, Patient's Surrogate can offer substituted judgment on pt behalf*)]

Answer 115

P = N / (1-P) ## Footnote "**N**umber **1** **N**igga, **P**ositivity" ..... "**P**ositivity... **N**umber **1** **P**lan". **N**egative LR = (**1** - se**N**sitivity) / **P** **P**ositive LR = se**N**sitivity / (**1** - s**P**ecificity) INDEPENDENT

Answer 116

keeping individuals involved **unaware** of participants' treatment assignments \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ [*Double* Blinding] ⇪ internal validity

Answer 117

- failure to detect a difference between groups when a difference does in fact exist - or (AKA): failure to reject a null hypothesis … that is actually false and should be rejected = false negative (*since rejecting = positive*) * * * * * * ---{probability of [Type 2 error] occurring = β } and this is inversely related to [***power*** *(which is how large a study has to go to detect a difference when a difference does exist)**]* - smaller study = ⬇︎power = [⇪ β = INC chances of Type 2 Error occurring] * \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_* [T2E = β = (1 - *power*)]

Answer 118

occurs when, even though [pt Test A] and [pt Test B] both die 5 years after the same disease.. bc [pt Test A] test diagnosed their dz 2yrs earlier...it'll SEEM like [pt Test A] had longer survival time when actually they're both only 5 years

Answer 119

serious transgressions against physician safety and/or well-being (such as unwanted touching) \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ 1st: reinforce hospital code of conduct with patient 2nd: **REASSIGN PHYSICIAN** 3rd: document alert in patient's chart

Answer 120

**inaccurate recall of past exposure by participants** ➜ MISCLASSIFICATION OF EXPOSURE (*applies mostly to case control studies*) \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *individuals with dz are more likely to search their memories for (possibly inappropriate) association between an exposure and their dz ⼀compared to individuals w/out dz*

Answer 121

the ability to detect an effect \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *(depends on sample size = larger sample size ➜ INC power)*

Answer 122

A = [inpatient (includes SNF, Hospice and Home health)] B= Outpatient C = (*Medicare Advantage*) ⼀allows private health care to provide Medicare benefits D= prescription drugs

Answer 123

[chi-square] determines whether distribution of a categorical variable is different across ≥2 independent groups = **p-value** IF • [chi-square *p*] ≥ [significance level α] = **NO SIGNIFICANT DIFFERENCE** between distribution of categorical variable between groups = **NOT ENOUGH EVIDENCE TO REJECT NULL HYPOTHESIS = NOT STATISTICALLY SIGNIFICANT**

Answer 124

*PCP* can share PHI when patient **p-C-p**... ## Footnote 1. is [**p**resent (or otherwise available prior to disclosure)] 2. has **C**apacity 3. gives **p**ermission

Answer 125

study design that utilizes ≥ 2 interventions and all combinations of these interventions

Answer 126

[null = r = 0] ⼀*since this indicates no linear relationship \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_* * a CI for *r* is statistically significant if it EXcludes 0 * CI of a *r* indicates*,* _with a certain confidence level_, whether *r* is significantly different from its [null= *r* =0] * **so… if study states [linear relationship between 2 variables (** ***r*** **)]** **is significantly different from 0** **(i.e. statistically significant | p\<0.050)** **➜ further illustrated by [CI for that (** ***r )*** **must EXclude 0]**

Answer 127

[CV prevention] / anti-HTN / [PRN Insulin] \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ meds taken at end of life should be comfort meds only

Answer 128

[**N**LR-] = value of a negative *dtr* = [(1-**N**) / P] [**P**LR+] = value of a positive *dtr* = [N / (1-**P**)] \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ | *🔎dtr = diagnostic test result* ## Footnote 🧠LR is likelihood that a *dtr*^{*[diagnostic test result: + or -)*} will occur in the {[Diseased pts^{= rule IN_dtr}] or [Nondiscriminatory/both equally] or [Healthy pts^{= rule OUT_dtr}]} 🧠LR ranges from 0 to infinity. LR furthest away from 1 ^{*(i.e. <0.1 or >10)*} have strongest discriminatory^{*rule OUT|rule IN*} Capabilities 🧠Which → [⇪LR from 1^{*in either direction*} = ⇪ likelihood Test can establish or eXclude disease.] 🧠 🅸 {⭐LR > 10 = Test strongly ESTABlishes^{rules IN} ⤵️Disease} 🇳 {⭐LR = 1 = Test can NOT discriminate between DISEASE vs HEALTHY = USELESS ❌} 🄾 {⭐ LR < 0.1 = Test strongly EXcludes^{rules OUT} ⤴️Disease}

Answer 129

purely *observational* study in which small group of patients with [similar dx or similar tx] is **_described_** [at one point in time] or [over period of time/followed]. ## Footnote **There is no comparison group**

Answer 130

Confidence interval contain [***sample value*** **(∆ between 2 sample percentages)** **at the center of the interval**] = [sample value +/- margin of error] * * * 1st: determine *sample value* [(Drug X 74.0%) - (Drug Y 85.4%) = **(-*****11.4%)***] 2nd: determine if [p \< 0.001] --(if yes)-\> significantly different from 0 and excludes 0 = **CI% will also not include 0** 3rd: Find range that does not include 0 and has [***sample value -11.4%***] its the center = (-13.9% to -8.9%)

Answer 131

SPECIFIC; *task overlap between team members may ➜ duplication errors! [Ideal = clear specific roles with continuous refinement of clinical processes via quality improvement and by tracking outcomes]*

Answer 132

Case-Control Study \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *allows quick localization of outbreak source*

Answer 133

Confidence interval contain the [**sample value (∆ between 2 sample percentages) at center of the interval**] = [sample value +/- margin of error] \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ when comparing ≥2 groups (*i.e. treatment 1 vs treatment 2*), if their Confidence Intervals **DO NOT** overlap = There is a Statistically Significant Difference ✔︎ between those 2 groups \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ when comparing 1 group to placebo, if that groups' Confidence Interval **DO NOT** contain the null value (0) = Overall Statistically Significant \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ If 2 groups' CI overlap = there may or may not be a statistically significant difference between those 2 groups

Answer 134

[**ARR**= (risk_PLACEBO - risk_Treatment)]

Answer 135

**RRR**= [ARR/risk_CTRL] --or-- [(risk_CTRL - risk_Treatment)/risk_CTRL)] \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *[ARR: Absolute| RRR: Relative _ Risk Reduction]*

Answer 136

[**ARR**= (risk_CTRL - risk_Treatment)] [RRR = ARR/risk_CTRL ] * [ARR: Absolute| RRR: Relative _ Risk Reduction]* * * * NNT = [1 / ARR] = **{1 / [(failure rate_PLACEBO) - (failure rate_TX)]}**

Answer 137

*[extraneous (3rd) variable] that's associated with both the [independent exposure] and the [dependent outcome] \_\_\_\_x\_\_\_\_ the effect the [independent exposure] has on [dependent outcome]* ## Footnote C: [**CONFOUNDS**(obscures)] E: [**MODIFIES direction & strength**] * * * * Separate stratified analysis should be conducted for EACH level of an Effect modifier*

Answer 138

**PCD** ; [⇪ sP = ⇪ PPV] \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *[⇪ sP = ⇪ PPV = ⬇︎FP]* | **P**CD = [**P**roof/Confirmatory or Diagnostic]

Answer 139

In a [group with an outcome], the percentage of that group that can **attribute** that outcome to a specific exposure \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ ARP = [(Risk_EXPOSED - Risk_nonexposed) / Risk_EXPOSED] x 100 \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *Risk = Risk of getting outcome*

Answer 140

FALSE ## Footnote *referrals to specialist, even when specialist personally known to the referring physician, is OK **as long as it is medically indicated** and there's **no financial compensation** exchanged for said referral*

Answer 141

estimates proportion of the [*population* with disease] that actually acquired that disease from the exposure \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ PARP = [{(Prevalence) x (RR-1)}] / [{(Prevalence) x (RR-1)} + 1]

Answer 142

association between [2 categorical (quaLitative) variables] *(i.e. study evaluating association between sex and [Y/N presence of MI])* \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ FET is commonly used to analyze 2x2 contingency tables when the sample sizes are small

Answer 143

**P-C-P-A-P-AAU** \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ 1. **P**t designated 2. **C**ourt appointed^{*(only occurs if multiple surrogates disagree, none is designated or surrogate acting in self interest)*} 3. **P**artner/Spouse 4. **A**dult children 5. **P**arents 6. **A**dult siblings 7. **A**dult relatives*other* 8. **U**nrelated friends ## Footnote _{*✏️Court-appointed guardianship only occurs if multiple surrogates cannot agree, no healthcare surrogate is identified, or surrogate is clearly acting in his/her own self-interest*}

Answer 144

▶**THE PATIENT THEMSELF** has to freely consent to [reproductive sterilization] on their own ▶ legal guardians/appointed health care decision-makers CAN NOT consent on patient's behalf for [reproductive sterilization] ## Footnote *Involuntary sterilization (especially in retarded patients) is unethical*

Answer 145

scree**N**ing **\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_** seNsitivity= [true Positive] = [TP / (TP + FN)] = *used for test scree**N**ing* = *"a test's probability (in the presence of disease) a patient test positive"*

Answer 146

*Since [slow progressive benign cases] outlive [rapid progressive malignant cases], **Length**-time bias occurs when a test only ends up detecting those [slow progressive benign cases (which naturally have better prognosis)] ➜ false appearance that the test actually INC survival benefits* \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ ## Footnote survival benefits of a screening test are overstated due to detection of a disproportionately higher # of [slowly progressive benign cases]

Answer 147

**P**CD ; [⇪ sP = ⬇︎False Positives] ⼀*which is why good Confirmatory test have HIGH sPecificity* \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *[⇪ sP = ⇪ PPV = ⬇︎FP]* | **P**CD = [**P**roof/Confirmatory or Diagnostic]

Answer 148

[p-value] = Chance that study results happened randomly [(At CI 95%) [p-value] < 0.05] means you can Reject Null hypothesis since it means there's lil chance the results happened randomly

Answer 149

[exposure has **lower** ODDS of causing Outcome] ⬅︎ [**OR 1.0**] ➜ [exposure has **HIGHER** ODDS of causing Outcome] ## Footnote *if (OR = 1) = exposure NOT ASSOCIATED with Outcome* \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ [1.0 - (0.351)] = (0.649) = 64.9% = [[exposure (HPylori+)] has 64.9% **lower** ODDS of causing [Outcome (stomach bleeding)] ...*as compared to [non-exposure(HP-)]*

Answer 150

**DCRSA** ## Footnote 1. Data Collection 2. [Causal Factor flow chart] (main steps leading up to event that, if eliminated, could have prevented or reduce adverse event) 3. Root Cause (reason behind each causal factor) 4. Solutions recommended (for each root cause) 5. Assessment (assess success of Recommendations)

Answer 151

Non-Statistically significant difference between the two data sets being compared

Answer 152

***S****n**N****O**ut* **S**ensitivity HIGH + **N**egative test result = rules *Out* a disease

Answer 153

TRUE ## Footnote * * * * Physician Referrals resulting in financial gain for that Physician are generally illegal and/or unethical* . * Routine referrals should only be made to facilities with evidence-based & medically-indicated care*

Answer 154

NPV= [TN / (TN + FN)] = *probability (in the presence of a Negative test result) a patient **DOES NOT actually** have the disease* = ability of test to accurately predict Negative test result_ How much can we trust this test?

Answer 155

**CAN** \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ | *"pt must have [⊕time & ⊝911 needs] before they can be terminated* ## Footnote Abandoning Patient if: 1. Patient terminated whilst in **immediate** medical need 2. Patient NOT given reasonable time to find alternate provider

Answer 156

FALLLSEE!!!! ## Footnote Active (i.e. plans to hang themself) suicidal or homocideal ideation warrants breaking confidentiality and informing parents or whomever

Answer 157

using gold standard testing to verify/confirm results of a preliminary test (which ➜ over/underestimation of seNsitivity or sPecificity)

Answer 158

publication

Answer 159

SMR of 1.75 = *"observed # of deaths (in this population of interest*) *is 75% higher than expected"*

Answer 160

If the [disease strength] increases as [risk factor exposure level] increases = *dose response relationship* = likely causative {although this is **not** necessary to infer causation}

Answer 161

[acknowledge their resistance] + [open ended questioning/motivational interviewing] \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *"You have reasons for not wanting to contact your brother; what are some possible benefits of sharing the results?"*

Answer 162

investigators must have analyzed [**TIME-TO-EVENT/DEATH**] data instead and saw that, although both groups had 80% die ultimately, the treatment group must have had longer [TIME-TO-DEATH] = Treatment group lived longer = Treatment overall more effective ## Footnote *[TIME-TO-EVENT/DEATH] often used in survival analysis*

Answer 163

**UNETHICAL PROFESSIONAL MISCONDUCT** \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *SEX/ROMANANCE between physician and [patient (current OR former✳)] is UNETHICAL PROFESSIONAL MISCONDUCT* | ✳ = if MD exploits pre-existing knowledge

Answer 164

*current patient =* **UNETHICAL 100%** \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *former patient* = Unethical if MD exploits knowledge or influence derived from previous professional relationship

Answer 165

PPV = [TP/ (TP+FP)] = *probability (in the presence of a Positive test result) the patient **actually** has the disease* = ability of test to accurately predict Positive test results_ \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ _PPV = ability of test to accurately predict Positive test results_ How much can we trust this test?

Answer 166

[(a**D**) / (bc)] *[(**D**ISEASE^OOE)/ (control^OOE)]* \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ *• OOE = Odds of Exposure* *• first draw contingency table | • [(OR \>1) indicates there IS an association between a **D**ISEASE and the [exposure or risk factor] in question*

Answer 167

POSITIVE skew | ( *"it's skewing away from the POSITIVE direction* = POSITIVE SKEW) ## Footnote mea(N) > med(i)an > m(O)de