Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
y
2-Information: is what happened when data are aggregated together, analyzed
bh
and interpreted into a form, useful for decision making.
So
&allow the user to understand the severity of issue &what action to be taken.
a
process and patient outcomes.
m
es
IM Process:
N
y
bh
1- Data should be defined, collected, analysed before being used in quality
management.
So
2- Not all data are useful in QM (DRIP) data rich but information poor.
3- Focus on information related to patterns of care not individual care.
a
4- Process should be monitored as well as outcomes.
m
5- Cost & quality are inseparable issues.
es
Information Resources :
N
Internal external
r/
y
4- Cannot be used when a desired
bh
change is needed .
So
a
m
&TYPES OF INFORMATION AVAILABLE
es
N
r/
D
y
bh
Language
So
Collection Principles Concepts:
a
Organizational leader identify (interdisciplinary quality team) & (cross
functional team)
m
es
1-to maximize
Use of collected data
N
2-to minimize
D
Effort duplicated
y
bh
3- Using excel to validate data collected
So
Index vs. Register
a
m
Index Registers
es
required by state laws and logic to locate cases for record listings for maintaining certain
maintenance, statistics and research فيها اهن الوعلىهات statistics, patients are possibly
r/
separated to
Examples: inpatients, outpatients, and
D
y
bh
Protected information: information that can’t be obtained by others or used in a
court of law which is protected &cannot be disclosed without the consent of the
client (privileged communicated)
So
HIPAA: Health Insurance Probability And Accountability Act
The HIPAA "minimum necessary" rule means that, access to "protected health
a
information" (PHI) is to be limited to those persons or classes of persons who
m
have a need to know in order to carry out their roles ,the patient is considered
es
the "owner" of the information in the U.S and can access and copy that
information by signing in a release form.
N
y
1- Continuity of care (by another practitioners)
bh
2- Communication among practitioners.
3- Legal protection
So
4- Data, information source
5- Patient identity
6- Diagnosis support &justification
7-
a
Documents of the course and results
m
8- Determine the reimbursement rate and justification of claims.
es
It is a screening process :
r/
y
What is the quality objective from it :
bh
1-To help decision-making process.
So
2-communicate among all functions.
3-Improve data accuracy
4-privacy , accuracy and security of data.
a
m
An integrated electronic information system that links all activities such as
quality, utilization and risk management with patient clinical and financial
es
database (this is the quality goal)
What key element must be in MIS :
N
quality strategy.
D
2- The ability of program to integrate all organization function and data base
3- flexibility of using program
4- level of computer knowledge training of staff
5- cost/effectiveness analysis
6- timely upgrade
y
Causality: the interrelationship of cause and effect, it is means that nothing can
bh
exist or happen without cause.
When there is a strong relation between risk factors and the outcome, so there is
So
causal relationship.
To say there is a causal relationship in epidemiology , it requires:
a
1- Biological plausibility must exist ( method establish causal relationship
m
between a biological factor and a particular disease )
es
2- Risk factor "exposure" must precede the outcome
• Types of causal association include:
N
Frequency Comparisons
y
health system.
bh
Incidence: Rate of new cases during a specified time period. e.g: new cases
during 1 year period ( no of new cases that develop in a given period of time )
So
New cases of transfusion reaction in year 2018 / total no of patients receiving
blood in year 2018
a
Prevalence: no of all cases of a disease that are present in a particular
m
population at a single point of time ( point prevalence ) or over a period of time (
period prevalence )in a given population )
es
Point prevalence = (Old cases + new cases) at one point in time / number of
persons in the population at the same point in time
N
Period prevalence = ( old cases + new cases ) over a defined period of time /
r/
Mortality Rate:
The proportion of deaths in a population within a time range.
= no of deaths / total population
Sensitivity: The ability of a measure, test, or tool (study design, screening tool,
or lab test) to identify and select all positive cases or specified variations or
deviations.
Sensitivity=TP/TP+FN = true positive / ( true positive + false negative )
Specificity: The ability of a measure, test, or tool to exclude
all negative cases.
Specificity=TP/TP+FP = true negative / ( true negative + false positive )
Stratification: The classification of data into homogeneous groups or
subsets.
e.g., the study of UTI by organism; by catheterization, by surgical procedure, by
nursing unit.
Recordability: ability to identify, capture & measure needed information
y
bh
Validity: الصالحيهThe capability of the indicator or collection tool to measure
what it is supposed to measure
So
Reliability: الذقه في االداءThe ability of the indicator or collection tool to
measure in a reproducible way what it is supposed to measure (" interrater
reliability").
a
m
Usability: سهىله االستخذامThe relative ease with which the indicator can be
understood or the tool can be used.
es
Other definitions
N
Goal: a numerical value that defines the significance level of the data that is
desired for decision making
r/
y
Focused: data collection occurs when only certain topics are the focus of data
bh
collection.
Intensive data collection , so the Highest prices &need resources
So
Population &sampling
Population:100% of the possible group to be studied (individuals, objects,
Events) 2 types(static or dynamic).
a
m
Sample :a subset of a population or a group drawn from a larger population.
es
The purpose of sampling:
N
Factors to consider:
D
Sampling techniques
Probability sampling:
High probability that findings can be generalized to the whole population.
Non probability sampling:
Low probability that findings can be generalized to the whole population.
TYPES OF PROPABILITY SAMPLING TECHNIQUES
(RANDOM SAMPLING)
y
1) Simple Random
bh
Uses a table of random digits (available in all statistical software) to
So
select the case from a list every case in 1 population
( Each case has an equal chance of being selected)
a
2) Stratified simple Random m
Sampling utilizes 2 or more homogenous categories ( strata ) of a
es
population, then choose one of each category randomly. بنقسم األشخاص
لمجموعتين أو أكثر ثم نختار عشوائي من كل مجموعة
N
3) Systematic Random
r/
(NON-RANDOMIZED SAMPLING)
1) Convenience
Sampling utilizes data that is most readily available الداتاالمتاحةوالسهلة
Selection bias , so may not reflect whole population
2) Quota
Sampling utilizes portions or percentages of persons in a stratified
population , the researcher identifies the strata according to his
y
certain criteria, and he determines the proportion needed from
bh
each strata.
3) Purposive( judgment )
So
Sampling selects persons because they demonstrate a desired
characteristics.
a
m
4) Expert Sampling :
es
N
MEASUREMENTS
r/
D
Data basics
y
zero is absolute value
bh
zero is benchmark
So
Examples Race, satisfaction Weight, height, time, length
degree(v.satisfied- of stay ,Ratio ,Age , Charges
satisfied -neutral-un
a
m
satisfied -v. un satisfied)
es
Statstical Test % in each category Mean, Mode, Median, Max
CHI SQUARE test percentile
N
T.Test
r/
y
A process is in good statistical control When:
bh
1) Stable over time (demonstrate through measurement data).
So
2) Operated in a stable, consistent matter with no arbitrary changes in
process steps or conditions.
a
3) The process aim is set and maintained at the proper level based on
m
quality specification and target values.
es
4) The average or normal process variations (control limit) falls within the
specification limits.
N
r/
Basics Statistics:
D
Descriptive Statistics:
Data type Distribution Central Tendency Variability
Nominal µ, percentage Mode
Ordinal µ, percentage Median, Mode Range,min/ max
Interval\Ratio Mode,Median,Mean Range, min/max,
SD, variance
Central Tendency:
Describes a set of measures that indicates what is the middle value.
1-MEAN= Average=
For Continuous data
Used with: interval and ratio types of data
When Astronomical value or outliner data (the value is very different from
remaining values ex. 1, 3, 5, 60) So, the mean is pulled toward astronomical
y
value. So, if data skewed, we use Median instead of Mean.
bh
Disadvantages: doesn’t really indicate the middle of data
So
2-MEDIAN
a
The middle value in the list of numbers ( represents a location)
m
Used with: Ordinal, Interval / ratio data types.
es
3-MODE
Is the most Frequent appearing number
Used with: Nominal, Ordinal, Interval, Ratio data
Like: Yes or No questions, Percentage, Proportion
Example: 29, 56, 109, 110, 360 > No mode
NOTES:
1) If normal distribution curve: Mode=Median=Mean.
2) If asymmetrical (skewed distribution): Mode in highest point.
and Mean toward the tail and Median between Mode and Median
So, it is better to utilize Median.
3) If there are repeated samples of the same types.
y
Mean is more stable value from sample to sample.
bh
Mode is the least consistent value.
Weighted Mean = Weight*Mean
So
To calculate total performance score using weighted mean
1-RANGE:
r/
The difference between the highest and lowest numbers in a set, but it doesn’t
D
2-FREQUENCY DISTRIBUTION:
A logic and systematic arrangement (Rank Ordering) of numerical values from
the highest to lowest or from the lowest to the highest
There are 4 types of frequency distribution
Simple Frequency distribution Grouped Frequency distribution
125 5 Scores frequency
124 3 100-150 2
123 2 151-200 4
y
N=10 201-250 3
bh
بيشوف كل رقم من دول اتكرر كام مره و i=interval=50, N=2+4+3=9
بعدين يجمع التكررات و مجموعها يساوي عدد للتسهيل في حاله االرقام الكتيره لكن نفس فكره
So
العينه اللي قبلها
Cumulative Frequency distribution Relative Frequency distribution
a
Calculated like grouped frequency
m Proportion calculation of percentage
then add the value of data to all to whole relationship
es
smaller points
N
100-150 2 2
D
151-200 4 (2+4)=6
201-250 3 (3+4+2)=9
N=2+6+9=17
3- RATIO Or Proportion
A Fixed relation in number or degree between 2 similar things
Ex) calculation of the difference between 2 ratios
y
σ=SD=√
bh
Variance: is the arrhythmic mean of the squared difference between each
So
value and mean value
Normal distibution bell curve
a
m
es
N
r/
D
y
indicate real (significant) difference
bh
So
How To Choose A Hypothetical Testing ? ( in – brief )
Any numerical data ( interval / ratio ) which I can calculate a mean from T –
test or ANOVA according to number of samples
a
If you couldn’t calculate a mean ) Nominal / Ordinal ) data Chi –Square Test
m
Both Chi Square and T.Test has P-Value
es
TYPES OF ERRORS:
y
bh
difference
Ha Alternative hypothesis ( assumes there is a difference )
So
P-VALUE Assumes null is true , it is the probability of obtaining
results at least as the Null , so high P value means high probability
a
m
that the null is true , so the results fails to reject null hypothesis and
es
low P-value means acceptance of alternative hypothesis
The probability of making alpha error ( incorrectly reject null ) usually
N
5 % = ,05 ,SO
r/
Regression Analysis:
Is a statistical technique that allows one to
compare
y
m a straight line.
bh
If r=+1 Strong positive relationship>> both increase or decrease together
If r=-1 Strong negative relationship>> one increase and another decrease
So
If r=zero>> No significant relationship = No linear relationship ( may be a
circulinear relationship or any other relationship)
a
The scatter diagrams
m
es
One way to display the possible relationship between 2 variables (2 sets of
data), looking at how closely they correlate
N
Regression Equation:
D
The formula for the line that best fit dots of the scatter diagram
Can be used to predict the expected value of one variable based on values of
others
MULTIPLE REGRESSION ANALYSIS:
Similar to simple regression analysis but includes multiple independent values.
Interpercentile range: (Interquartile range)
y
bh
So
Refers to the data between the 1st quartile (25th percentile) and the 3rd quartile
(75th percentile) or, Sometimes describes the middle 50% (50th percentile) of
a
the data values
m
The most Commonly used IQR is in Baby Growth Chart
es
N
r/
D
Process Variations
y
When to say that the process is statistically controlled process?
bh
When the process is
So
1) stable over time
2) operated in a stable consistent manner (reliable)
a
3) control limits fall within the expectations
m
But there are 2 types of process variations
es
Random or Common cause Assignable or Special cause
variation variation
N
y
bh
Quality Improvement Tools
So
The appropriate display of information is a key responsibility of the quality
professional, So the decision of performance improvement will be made and it
a
will directly affect process effectiveness and outcome.
m
In every tool we should know answers of these following questions:
es
1) What is this tool? (in exam they may bring a display to recognize which
N
tool is it?)
r/
y
2) Pie chart 2) Brain storming
bh
3) Frequency plots 3) Task list
So
a) steam and leaf 4) Gantt chart
b) dot plot 5) Cause and effect diagram
c) Histogram
a 6) Affinity diagram
m
d) bar chart 7) Flow chart
es
e) pareto chart 8) Lotus diagram
N
STATISTICAL TOOLS:
1) Tables
organized and summarized data for a sample, population, a given set of
criteria, screens.
It is not preferred to show it to leaders because It is hard to understand
Ex) infection control surveillance (statistics related to antibiotics)
Specific measurement data
Ex) medication use, waiting times
y
bh
Identifying a problem
So
Outcome evaluation
2) Pie chart
a
A relative frequency (%) of the proportional relationship with a data set
m
Not frequently used unless if proportion it used
es
N
r/
D
3) Frequency plot
How often a value occurs and repeated
a graph designed to display the location, spread and shape of the data
a) steam and leaf
b) dot plot
c) Histogram
d) bar chart
e) pareto chart
A)stem and leaf plot تقسماألعدادحسباآلحادوالعشرات
65,67,68,71,74,108,103,103 109,112,112,115,125,129
10 9 6 5,7,8
y
11 225 7 1,4
bh
10 8,3,3
12 59
So
B) DOT PLOT
a
A graph utilizes a dot for each unique value
m
es
N
r/
D
y
bh
A bar graph of the frequency of one continuous variable, x-axis is
So
independent variable and y-axis is dependent variable
Data analysis
a
Outcome evaluation
m
D) BAR CHART
es
N
r/
D
E) PARETO CHART
A special form of vertical bar chart with bars in rank order of occurrence
from highest to lowest.
Mainly used to Prioritization after comparison.
identify the problem
y
data analysis
bh
outcome evaluation
There are 3 types of pareto chart
So
1) simple pareto chart 2) advanced pareto chart 3) pareto drill down
a
4) Scatter diagram
m
es
Fof data analysis
outcome evaluation
N
5) Run Chart
r/
D
Types of variations:
A) Common Cause Variations
is acceptable
y
quality department should monitor only because it is stable process or
bh
controlled process .
So
B) Special Cause Variations
if we have a special cause in run chart , so Next step is to do deep
a
analysis or root cause analysis or peer reviwe to investigate about the
m
special cause
es
identify the problem
data analysis
N
outcome evaluation
r/
y
break a trend.
bh
So
a
m
es
N
y
bh
So
a
m
es
N
r/
a) Generate ideas
y
b) Understand the process
bh
c) Peritorize improvement efforts
So
1) Brain Storming
To create many ideas in a short time, there are 3 types of brain storming
a
Structured
m
unstructured rapid
es
resolution
D
3) Multi-voting
To priotorize a long list of alternatives to end up with the critical few ideas
upon with the team should focus
multi-voting is best done after Brainstorming
4) Delphi-technique
Participant should support the consensus idea even if they don’t agree.
5) Cause and effect diagram (Ishikawa) (fishbone)
y
bh
sub causes> can be identified by using 5WHYS
So
Or 5P (people, provision, polices, procedures, place)
a
m
es
N
r/
D
6) interrelationship diagram
A tool allows a team to analyse all the interrelated cause and effect
relationships and factors involved in a complex problem.
Used in the beginning of team work to order the jobs after brainstorming
y
Affinity means close relationship or connection or similar structure
bh
We should differentiate between primary issues and major related
So
subgroups in order to group the appropriate relationship and links
8) Lotus diagram
a
m
A tool to expand thinking around a single topic (major) to organize ideas
from brainstorming or to organize discussion during planning (for open
es
ended ideas )
N
r/
D
9) Flow Chart
It is a very detailed table (SIPOC) to identify value and non value steps in a
process from start to the end of process
y
bh
It is similar to process map but value stream map is for process
standardization
So
11) process map
A story boards
a
m
A problem-solving approach to structure thinking
es
Used For Change, if you want to apply change what is the driving
forces(strengths) and restraining forces(weaknesses)
13) Check list\ Task list
y
bh
Includes a list of tasks (process steps) and estimates of time of people of
resources required to complete the quality efforts
So
a
m
es
N
r/
D
bh
So
Data collection tools
a
m
- If observation, interview, focus group> qualitative
es
- The ADDLE training model (analyse, design, develop, implement,
evaluate)
N
r/
D