Biostats Flashcards
Why is statistic important in medicine
Think scientifically about health problems
help in decision making from evidence
assess possible risk associated with medical decisions
identify decision that lack scientific and logical basis
What are the two forms of statistics
Quantitative data collection
statistical methods
What are the different methods of data collection
Routine data forms
surveys
experiments
Methods of summarizing data
Tables
graphs
numerical statistic
Methods for making inferences and estimations from data
Test of significant difference
test of association
estimation
What are the three main uses of statistical methods
Collection of data
description of characteristics of a group or situation
analysis of data and drawing of conclusion
What are the two types of statistics
Descriptive statistic
inferential statistic
What is descriptive statistic
Description of characteristics of a product or process from information collected which summarize a group of scores to a single points
What is inferential statistic
The drawing of conclusion on unknown process parameters based on information in a sample which uses probability
What do you need to know in data collection
The type of research
the questionnaire design or interview guide design
sampling techniques
interviewing situations
What are the different type of numeric format for summarizing data
Measures of central tendency like mean median mode geometric mean
measures of spread like standard deviation variance range percentile
What are the different type of tabular format for summarizing data
Frequency distribution table cross tabulation
What are the different type of graphical transformers for summarizing data
Histogram bar diagram pie diagram scatter diagram frequency polygon
What are the different type of test of significance
The null hypothesis and P value t-test and t test table Z test and standard normal table the chi-squared statistics degrees of freedom confidence interval probability distribution
The different type of estimation
Regression
correlation
What is research
Systematic way of putting together information based on prior to main purpose to reach a conclusion which can be very favorable by your peers
Different type of research
Primary data collection secondary data analysis formative exploratory research pilot studies of observational studies clinical trials programmatic research translational research
What are the two types of primary research
Qualitative research
quantitative research
What is qualitative research
Based on personal stories of people through questioning in specific areas of interest
What’s techniques of data collection are involved in qualitative research
Interview guides direct observation checklist videography Notes Recording
Qualitative research that I collective is usually nonnumeric or numeric
Nonnumeric
What are the different types of qualitative research
Focused group discussion
individual direct interview
Key informant interview
special groups firsthand observation
In qualitative research what type of support is used for analysis and reporting
Analysis of themes
computer software
What is a focus group discussion
A group interview with a small number of demographically similar people who have common traits of experiences beliefs to share ideas
What are the steps in running a focus group discussion
Choose a topic develop questions from topic develop interview guide choose notetaker And recorder of session recruit participants five or 10 people obtain consents do self introduction introduce discussion topic get participants chance to speak with equal representation End the meeting in time Analyze data to provide feedback
What is quantitative research
Structured acquisition of data with participants that provide responses in pre-defined categories
No storytellin
Data collection tools in quantitative research
Forms
tally sheets
questionnaires
How are questionnaires administered in quantitative research
Through phone
self-administered
interviewer administered in face to face
mail
What are the mode of analysis and reporting of quantitative research
Numeric ,tabular ,graphical summaries
inferential statistics
computer software
does the interviewer matter in quantitative research
Yes can corrupt results
Most frequently used method of that’s a connection in quantitative analysis
person to person interview
When is a mail questionnaire advised
For sensitive to private issues with clear simple questions in the good postal system
Which one is stronger hypothesis driven research or exploratory research
Hypothesis driven research
What should be the characteristics of an hypothesis
Directional and related to underlying biological model or theoretical frameworks, justifiable
What is the 2 type of research question
Standard approach
foreground question / formative research
How do you evaluate your research question
Use the picot analysis
What does the picot analysis mean
Patients population problem intervention comparison outcome time elements /type of study
What is an aim in research
A statement derived from research question , Broads
What are objectives in research
Statements derived from the end which are more specific contains action verbs and are listed as primary or secondary
What are the characteristics of objectives in research
They have to be SMART Specific measurable attainable realistic time bound
What are the different type of data
Continuous Data
categorical data
binary data
What is continuous data
 uninterrupted range of values so data can only be collected in integers
Like bp, weight, age
What are examples categorical data
Race
ethnicity
country of birth
degree of agreements
What are example of the binary data
Yes or no
positive or negative
dead or alive
What are nominal data
Data can be labeled
What is ordinal data
Data that is put an order or on a rating scale
What is discrete data
Values that are distinct or separated and can only assume certain values with no intermediate values
It’s continues that are measurable or countable
Measurable
Is discrete data countable or measurable
Countable
What is a variable
A name, a symbol, a letter used to represent an item , category, in a data set
What are the different types of variables
Quantitative
Qualitative
What is quantitative variable
Measured numerically
Either discrete or continuous
What is qualitative variable
Description of category using characteristics of interest
No numerical value
4 scales of measurement
Nominal
Ordinal
Interval
Ratio
Nominal scale of measurement
Categorical Dichotomous Qualitative Data divided into different categories, no order
Ordered scale of measurement
Inhérent or prédétermined order in categories but nominal
Divided into categories that can be ranked
What is interval scale of measurement
Like ordinal scale but distances between two points are of known sizes
Arbitrary Zero point
Ratio scale of measurement
Like interval scale but with Zero point, meaningful & definite
Why are health indicators difficult to evaluate
Mostly not quantifiable
More than one factor involved in results
Sources of dlatafor development of health indicators
Vital events register Population & censures Routine heath service & records Sample surveys Disease registers
. Characteristic of a good health indicator
Relevant, valid, specific, reliable l sensitive
What is data
Collection of information
Two main sources of data to health personnel
Routine data coming from health personnel ( routine hospital forms , registers )
Periodic ( research data like surveys and experiments ,clinical trials data )
3 methods of data collection
Routine forms
Sample surveys
Experiments
Examples of data sources
Vital events register Population censuses housing censuses Routine health service records Epidemiological surveillance data Sample survey data Disease registers Experiment
Types of routine data collection forms
Laboratory form OPD forms Death register Cancer register Admission form Diagnosis index card
examples of primary source of data
Registers Telephone records Interview guides Clinical trials Questionnaire
Secondary source of data example
Annual reports
Publications
What are important factors to consider when collecting data
Good scientific procedure to determine sample
Sample size
Representativeness of sample \
Good instrument to serve purpose of data collection
What is the research protocol ( research steps when writing )
Title Abstract Introduction (background, problem rationale ) Literature review Aims and objectives Methodology Plan for data analysis Budget References
2 types of epidemiological studies
Observational
Experimental
What is observational study
Non experimental procedure in which nature determines exposure
2 types of observational studies
Descriptive
Analytical
Example of descriptive study
Case report
Case series
Case studies
Cross sectional studies in general
Types of analytical study
Case control
Cohort
Retrospective cohort
Prospective cohort
Example of experimental studies
Clinical trials
What is close ended questionnaire
Possible answers are provided for choosing
What is Open ended questionnaire
Respondent provides their own answer
What are the difficulties of open ended questionnaire
Difficult to code and make comparisons
What is reliability
Extent to which measures are repeatable
Consistency of measuring instrument
What is validity
Extent to which instrument does what it is intended to do
What is internal validity
How instrument measures content under investigation
What is external validity
Generalization to total population
What are the important points of the numberg code
Voluntary consent essential Yield fruitful results No unnecessary suffering No predictable death / disability Risk not greater than problem Participants can stop anytime Investigator stopped if harmful
Which declaration says “ interest of subject must prevail over interests of science and society “ and enforces privacy and informed consent
The declaration of Helsinki at world health assembly in 1964
3 basic principles of clinical research
Respect for persons with informed consent and privacy
Bénéficience out-weights maleficence
Justice
Code of conduct used by Europeans and Japanese in experiment
ICH guideline
Code of conduct used by USA in research
Code of federal regulations (CFR)
What is informed consent
Process by which fully informed participants in bio medical research make choices about participation in study
What information should a person have in informed consent
Not a one time activity Research goals Research procedure s Why person targeted Benefits and risks Voluntary Confidentiality ‘ Who to contact
8 elements of informed consent
Research and its procedure , methodology Risks, discomfort Benefits Alternate treatments Confidentiality Voluntary participation Compensation in case of injury Contact in case of emergency
What is source document
First place where research data is written out
What are the essential documents
Informed consent form Cv of investigators Copies of protocol Correspondance with sponsor Local lab values
Who provide assent form in Ghana
Children between 10 to 18 years
Sponsor definition
Entity that takes responsibility to initiate , manage , finance the clinical trial but does not conduct it
What are the responsibilities of sponsors
Record keeping Recruit qualified persons to conduct trial Handling data Verification of data Data analyses Preparation of reports
What is population
Total persons or items observed in given study
What is sampling
Some members of population are selected to observe purposes of drawing conclusions
What is a parameter
Characteristic of a population
Benefits of sampling
Save manpower , Time, finances
what is the law of statistical regularity
Sample from population possesses characteristics of larger group
What is the law of large numbers
The bigger the sample size the more accurate and steady the characteristic
Qualities of a good sample
Representative Randomly selected to minimize bias Large enough to increase precision All characteristics of interest Good research design
2 types of sampling
Probability sampling
Non probability sampling
Types of probability sampling
Simple random sampling Systematic sampling Stratified sampling Cluster sampling Multistage sampling
Types of non probability sampling
Quota sampling Snowball sampling Purposive Sampling Saturation sampling Convenient sampling
Sample frame
Entire list of sampling members of population from which sample will be taken
Sampling unit
Individual members of sampling frame
What is simple random sampling
Each member of population has equal chance of being selected into sample
Is sampling frame required for simple random sampling
Yes
Three ways of obtaining simple random sampling
Balloting
Random digit table
Computer
What is systematic sampling
Sample selected at interval
Is sampling frame necessary in systematic sampling
Not necessary but can be used
Procedure of systematic sampling
Choose sample size
divide population to get k
Choose item between 1 and K at random
go down the frame and pick Kth item every time until you exhaust the frame
What is cluster sampling
Sampling unit is group instead of individual like households in villages
Is sampling frame necessary in cluster something
Yes
Procedure of cluster sampling
Villages selected at random
list of head created
sample selected from a list of head using simple random sampling
everybody related to head of household selected part of sample
What is stratified sampling
Separate heterogeneous population into homogeneous strata to allow evaluation of underrepresented groups
What is multistage sampling
multiple stage to sample
What is quota sampling
Population divided into groups which ideally represents what exist in the population
what is the limit of quota sampling
Selection bias
can’t calculate sampling error
What is a snowball technique in sampling
Previously identified group members identify other members of the population until most of the population is named
In what cases can you use snowball technique
For drug users
illegal aliens
prostitutes
community opinion leaders
What is the situation technique in sampling
Interviewer select individuals into study while partially analyzing information collected and stuff when realize that no new information is added to the study
What cases are saturation technique used
In qualitative research
What is purposive sampling
Nonrandom method of sampling in which a group of people are sitting with particular characters are selected
What is the convenience sample technique
Simple drawings from population at hands
When is convenient sample useful
For pilots testing and quantitative work
What does minimal sample size depends on
Objective of study
level of confidence to estimate population parameter
level of acceptable error
consideration with available resources
What are some assumptions made to calculate sample size
Single statistic calculated normal distribution of the statistic sampling using simple random sample only type one error have 2 tail situation with null Hypothesis and alternative hypothesis different from zero
What data doyou need before calculating sample size
Objective of study is SMART
level of confidence
level of acceptable error
available resources
What is a type one error
Null hypothesis is rejected when it’s true
What is type ii error
Null hypothes is accepted when its false and alternate hypothesis is rather true
What is the formula of sample size using Cochran formula
n = (z^2 * p(1-p) ) / d^2
What is the formula of sample size using Slovinsformula
n= N/ 1+ Ne^2
