44 Statistics and Patient Safety

Question 1

type 1 error

Answer 1

rejects the null hypothesis incorrently, falsely assumed there was a difference when no difference actually exists

Question 2

type 2 error

Answer 2

accets null hypothesis incorrectly, 2/2 small sample size, tx are interpreted as equal when there is actually a difference

Question 3

null hypothesis

Answer 3

hypothesis that no difference exists between groups

Question 4

p value

Answer 4

convention is <0.05 rejects the null hypothesis … mean 95% likelihood that the difference between the populations is true … <5% likelihood that the difference is not true and occurred by chance alone

Question 5

variance

Answer 5

spread of data around a mean

Question 6

parameter

Answer 6

population

Question 7

mode

Answer 7

most common value

Question 8

mean

Answer 8

average

Question 9

meadian

Answer 9

middle value, 50th percentile

Question 10

trials and studies: list types

Answer 10

RCT, double-blind controlled trial, cohort study, case-control study (retro), meta-analysis

Question 11

randomized controlled trial

Answer 11

prospective study with random assignment to treatment and non-tx groups, avoid treatment biases

Question 12

double-blind controlled trial

Answer 12

prospective study in which patient and doctor are blind to treatment, avoids observational biases

Question 13

cohort study

Answer 13

prospective study - compares disease rate between exposued and unexposed groups (random assignment)

Question 14

case-control study

Answer 14

retrospective study in which those who have the disease are compared with a similar population who do not have the disease, the frequency of the suspected risk factor is then compared between the 2 groups

Question 15

meta-analysis

Answer 15

combines data from different studies

Question 16

list quantitative vs qualitative variables

Answer 16

quant = student’s t test, paired t test, ANOVA … qual = nonparametric statistics, chi-squared, kaplan-meyer

Question 17

student’s t test

Answer 17

2 independent groups, variable is quantitative - compares means

Question 18

paired t test

Answer 18

variable is quantitative, beore and after studies (i.e. weight before and after drug vs placebo)

Question 19

ANOVA

Answer 19

compares quant variables (means) for more than 2 groups

Question 20

non-parametric stats

Answer 20

compare categorical (qualitative) variable (i.e. race, sex, medical problems and diseases, meds)

Question 21

chi-squared test

Answer 21

compares 2 groups with categorical (qualitative) variables - i.e. number of obese patients with and without DM versus nonobese pts with and without DM

Question 22

Kaplan-Meyer test

Answer 22

small groups, estimates survival

Question 23

relative risk

Answer 23

incidence in exposed / incidence in unexposed

Question 24

power of test

Answer 24

probability of making the correct conclusion = 1 - prob of type 2 error (accepts null incorrectly) … likelihood that the conclusion of the test is true … larger sample size increases power of test

Question 25

prevalence

Answer 25

number of people with disease in a population (i.e. number of pts in US with colon CA), long-standing disease increases prevalence

Question 26

incidence

Answer 26

number of new cases diagnosed over a certain time frame in a population (i.e. number of pts in US newly diagnosed with colon CA in 2003)

Question 27

TP, TN, FP, FN 2x2 table

Answer 27

L: disease present and absent …. top: test positive and negative

Question 28

sensitivity

Answer 28

ability to detect disease = TP / (TP + FN) … note denominator is the actual number of pts in the population with the disease … indicates the number of people who have the disease who test positive …. with high sens, a negative test results mean pt is very unlikely to have the disease

Question 29

specificity

Answer 29

ability to state no disease is present = TN / (TN + FP) … note denominator is true number of pts in population without disease … indicates the number of ppl who do not have the disease who test negative … which high spec, a +test result means pt is very likely to have the disease

Question 30

PPV

Answer 30

TP / (TP + FP) … denom is all + tests … likelihood that with a positive test, the patient actually has the disease

Question 31

NPV

Answer 31

TN / (TN + FN) … denom is all neg tests … likelihood that with a negative test result, the pt does not have the disease

Question 32

accuracy

Answer 32

(TP + TN) / (TP + TN + FP + FN) … denom is whole population … note denominator

Question 33

predictive value

Answer 33

depends on disease prevalnce

Question 34

sensitivity and specificity - relation to prevalnce

Answer 34

independent

Question 35

goal of NSQIP

Answer 35

NSQIP = national surgical quality improvement program … seeks to collect outcome data to measure and improve surgical quality in the US … outcomes are reported as observed vs expected ratios

Question 36

JCAHO prevention - use, protocol

Answer 36

aim to prevent wrong site, procedure, patient … protocol = preop verification of patient and procedure, operative site and side (marking if L or R or multiple levels, must be visible after the pt is prepped) … time out before incision is made (verify pt, procedure, position site and side, availability of implants or special requirements)

Question 37

what promotes a culture of safey

Answer 37

confidential system of reporting errors, emphasis on learning over accountability, flexibility in adapting to new situations or problems

Question 38

risk factors for retained object after surgery, MC object

Answer 38

MC retained sponge … emergency surgery, unplanned change in procedure, obesity, towel used for closure

Question 39

define sentinel event

Answer 39

defined by JCAHO - unexpected occurrence involving death or serious injury or the risk thereof –> hospital undergoes RCA to prevent and minimize future occurences (i.e. wrong site surgery)

Question 40

GAP protection technique

Answer 40

gaps in care (i.e. change in caregiver, divisions of labor, shift changes, transfers) can lead to loss of information and error … prevention = structured handoffs and checklists (face to face if possible), standardizing orders, reading back orders if verbal