exam prep week 1 - 7 Flashcards
Define evidence-based nursing practice
When nurses make clinical decisions using the best available scientific and evidence-based research, including patient values and preferences, clinical expertise and experience
Why is EBP important to health professionals?
Ensures: accountability professional standards clinical competence and safety improved practice and service applying research to the real world of work
What are to two categories of evidence?
- non-research based
2. research based
What is the definition of research?
Is a formal, rigorous and precise process of inquiry that is planned and systematic
Why is research important for nurses?
Enables better understanding through discovery of knew knowledge
Improves practice based on evidence
Where is evidence located?
CINAHL
Joanna Briggs Institute
MEDLINE
Cochrane library
what is a systematic review?
Findings from a number of studies (usually quantitative) answering the same clinical question
Is stronger than one individual study
What is a meta-analysis?
Takes all the statistics from the systematic review to draw a conclusion
Gives a useful summary that weights study findings based upon the strength of the methodology
What is a meta-synthesis?
Involves analysis of several qualitative studies and summarises the findings
What are the 7 components of research reports?
- Title
- Abstract
- Introduction/aim
- Methods used
- Results/findings
- Discussion
- References
What does the introduction compromise of?
Defines the problem
Formulates the aim or question
What does the methodology compromise of?
Is the development and implementation of a plan for the research
What is in the results/findings?
Documentation of the findings of the data analysis
What is included in the discussion/conclusion section?
The interpretation of the findings
Draws conclusions and makes recommendations
An abstract summarizes a research report… what does it usually include?
Background Research question or aims Findings Conclusion (use it to see if it may answer your question)
Methodology includes?
The research design (qualitative, quantitative)
Sample (participant) selection from population group
Procedures and data collection methods
Data analysis
Quantitative design includes?
Hypothesis Control group Survey Random/convenience sample Variables measured by instruments or tools Reliability/validity of instruments statistical analysis Objectivity
Quantitative research includes?
A partial picture of a phenomenon is of interest
numerical information
The researcher is at ‘arms length’ from the data collection process
Validity has a specific meaning in quantitative reseach
Qualitative design includes?
No numbers
Focus is on subjective information
No control of phenomenon
Focus is on understanding complexity of people within the context of their lives
Includes naturally occurring conditions
Ensures answers will be as meaningful and accurate as possible
Ensures rigor/validity
Qualitative research includes?
A complete picture of a phenomenon is of interest
Non-numerical information
The research is NOT at ‘arms length’ from the data collection process
Rigor has a specific meaning in qualitative research
Discussion should include?
A summary of the key findings
Comparison of findings with previous findings
Speculation about the reason for the results of the study
Implications for practice
Discussion summary should …
Address the results that directly relate to the research question
Can include unexpected results
Includes results that the researcher believes are important
Discussion:
Comparison and Speculation
Compares the results with those from previous studies
Considers the findings from different theoretical perspectives
Offers explanation for the results including rationale to support this explanation
Discussion:
Implications for Practice
Interprets what the study and its results mean for practice
Makes suggestions for practice consistent with this interpretation and the supporting evidence (EBN)
Limitations…
Qualitative research – rigor
trustworthiness, confirmability, transferability, credibility
Quantitative research – reliability and validity
Measurement - reliability, validity
Design – internal and external validity
Conclusions…
Summarise:
New knowledge derived from the study
Consistency or otherwise of findings with previous knowledge
Limitations (design, sample, measures, procedures)
Recommendations (practice, education, research)
The research design identifies:
How subjects /participants recruited
What will happen during the study
How data will be analysed
When the study will end
Non-experimental Quantitative research design includes?
Correlation
Descriptive
Experimental Quantitative research design includes?
Quasi-experimental
Experimental
Qualitative research design includes?
Historical Ethnography Phenomenology Grounded theory Descriptive exploratory
Purpose of Qualitative Research
To understand personal experiences, interpretations from participant’s view
Help understand naturally occurring social phenomena
Explore attitudes, beliefs, values and experiences of participants
Searches for individuals’ perspective of reality
Interested in people’s experience of the world
Acknowledges that the researcher impacts on & is part of the research process
The Researcher in Qualitative Designs…
Visible in research process
Spends time in field
Personal contact and insight
What are the common types of qualitative research
Descriptive exploratory
Phenomenology
Grounded theory
Ethnography
Historical
What is Descriptive Exploratory?
Has overtaken other methodologies
Not a specific approach
A generalised approach: tends to include common aspects of others
Collects “rich narrative data from small samples”
Describes situations
Describe Phenomenology
Studies phenomena in their natural setting
Aims to describe experience
Discovers the ‘lived’ experience eg. living with cancer
Involves in-depth interviews (2-3 hours talking to someone)
Research questions: Phenomenology
Provides us with in depth data about specific lived experiences
eg. Describe the experience of the before and after of losing a partner
What is the lived experience of … ?
What is the meaning of …. ?
What is the quality of…..?
Describe grounded theory
Examines processes
Examines inter-relationships among concepts to develop theory
Interviews, examination of documents & observation
Focuses on social processes, meaning and developing theory
Recognizes links - builds theory from that
Grounded theory is used when?
When little study done in the area
To find new understandings or perspectives on unknown or unclear phenomena
To look at and understand social process that come from our behavior
eg. Behaviors of family members in ICU
Research questions of grounded theory would include?
What does … ? (recovery mean to ICU patients)
How do …. ?
What is the process of ….? (social processes, interventions and developing theories from that)
What is Ethnography?
Doesn’t have to be race or religion, could be people who use illicit drugs
Understand the culture of a group of people
Insider (emic) as well as outsider (etic) approach
Researcher often participant observer-embedding within culture
Observations often supported with interviews
Time taken to establish trust
Research question: Ethnography
Usually about patterns of behavior within the social context of a culture/subculture
Nursing: questions that concern how cultural knowledge, norms, values influence the health experience
eg. Culture of different wards; ICU, theater, students / a group that has something in common
Historical Research
Looking what has happened in the past and relating it to the future
Examines past as perceived at the time
Historical documents (eg diaries) and recordings often examined
Description of the past based on documented information
Research question: Historical
Any questions relating to exploring the facts, events, ideas and people’s lives in history
What was it like to be a person in ….?
eg. the use of side rails historically and in the future
Rigor in Qualitative Research
Trustworthiness/rigor interchangeable
Confirmability - audit trail
Transferability - external check
Credibility - data collection
(researcher need to be trusted by the participants)
Research Question example
What allows some people to adapt or cope with illness when others don’t?
Research question could be:
“Does previous experience with serious illness improve a person’s ability to cope with subsequent serious illness
Hypothesis
A tentative prediction about relationship between two or more variables in study
Must be testable - eg. is there a relationship between smoking and acncer
Can be more than one hypothesis in a study-each tested separately
Testable research hypotheses state the expected relationship between the independent variable (cause) and the dependent variable (affect) within a population
Hypothesis Example
(Does the independent variable cause a change in the dependent variable?)
Question: Does the ‘temperature of normal saline’ (independent variable) used in endotracheal (ETT) suctioning affect ‘heart rate alterations?’ (dependent variable)
Hypothesis: Room temperature normal saline used in ETT suctioning results in greater decline in heart rate alterations than body temperature normal saline
Hypothesis testing:
Allows generalisability of findings
Answers questions eg:
- What is the relationship between the variables? - What are the changes over time?
The statement of the hypothesis should
include:
Variables of the hypothesis
Population being studied
Predicted outcome of hypothesis
Sources of Research Problems
Clinical experience
Nursing literature
Social issues
Theories
Critiquing Research Problems, Questions & Hypotheses
Does the report answer: “why ask the question” & “what do we know”?
Does report include clearly identified background/literature review?
Does background discuss clinical problem?
Does literature review explain importance of question?
Is literature review important & current?
If theory presented, is it relevant?
Is problem/question/hypothesis connected to literature/theory?
Is question relevant to clinical problem?
Hierarchies of Evidence are for?
Traditionally based on quantitative methods
Minimise bias and remove effect of confounding variables
Systematic reviews of Randomised Controlled Trials (RCTs) at top
Opinion at the bottom
List the hierarchies of evidence…
Systematic reviews - meta analysis (Quant)
RCTs
Cohort studies
Case - control
Cross - sectional studies
Case series, case reports
Ideas, opinions, editorials, anecdotal (Qual)
How Do We Evaluate Qualitative Methodology?
Qualitative research: ? Lacks rigor/legitimacy/generalisability
Generally ranked low on evidence pyramid
A different way of evaluation needed?
If asking quantitative question (treatment effectiveness): quantitative pyramid for intervention studies appropriate
If asking question about how patient’s experience a particular health issue/intervention: qualitative pyramid would help guide an evidence review
Evidence Pyramid for Qualitative Studies
Level I Generalizable studies Level II Conceptual studies Level III Descriptive studies Level IV Single case study
Single case study:
interviews with only one participant. Poor evidence for practice. May lead to more robust studies based on ideas generated
Descriptive studies:
usually not guided by theory. Usually conducted to capture personal experience of participants
Conceptual studies:
use theoretical concepts to guide sample selection- based in the literature on the clinical question
Generalisable studies:
also use theory to guide sample selection but extend beyond one specific population to capture diversity of experience
Implications for Nursing
Nurses and qualitative researchers tend to approach situations form a holistic viewpoint
Nurses and qualitative researchers appreciate that there is no single reality- reality changes over time and is unique to each person
Therefore evaluating levels of evidence of qualitative research can highlight unique contribution nursing makes to patient care (different to medicine)
Types of quantitative research studies
Descriptive
Correlational
Quasi-experimental
Experimental
Quantitative Descriptive design
Accurate portrayal of phenomenon of interest
Used to answer research questions that seek to describe - (eg. of teenage mothers, how many are using the postnatal services?) interviews, questionnaires etc. numerical… if it was qualitative it would ask their experience of this
May use observation, survey, questionnaires, interviews
Longitudinal/cross-sectional - Retrospective/prospective… both elements of time
Standing back and not getting involved personally
Correlational design
Used to answer research questions that seek to link or connect (eg. the correlation between heart bypass and stress)
Enable examination of relationships between two or more variables & comparison between groups
Used to quantify strength of relationship between variables
Quasi-experimental design
Allows generalisation (most of the elements of the experiment but with something or one thing missing)
Experimental treatment manipulated
“Natural experiment”
Some characteristic of true experiment missing:
- control
- randomisation
Experimental design
Involves observation & data collection with specific criteria and protocol
Three identifying properties: - randomisation (randomly select population) - control (eg. give drug A - to ...) - Manipulation (Give drug B - to...)
Used to test cause & effect relationships (eg. smoking causing cancer)
Double blinded randomized controlled trial = the best
Experimental designs types
Randomised Controlled Trials (RCT)
Blinded RCTs (participants are anonymous)
Double-blinded RCTs (researcher and participants are both anonymous) = no bias
Time in quantitative studies
Retrospective: data collected about past events - for as long as it takes within reason
Prospective: data collected about events as they occur
Cross sectional studies - data collected at one point in time
Longitudinal studies - data collected over a period of time (days/weeks)
Cross-sectional - Prospective = + (Today - specific infectious diseases in child care)
Cross-sectional - Retrospective = ++
Longitudinal - Prospective = ++
Longitudinal - Retrospective = + (Over period of time in the past)
Control in Quantitative Research
Researcher needs to have control over study
Procedures to ensure uniform/constant conditions (homogeneous - sample is the same)
Control over variables (independent/dependent/extraneous)
eg. A is controlling B (not something else)
A is fixing B
(need to make sure the measuring tool is accurate)
Variables
Variable: aspect of interest that differs among different people/situations
Independent variable: manipulated - give to the controlled group - cause - smoking
Dependent variable: measured - measure what happened when we gave then cigarettes - effect - cancer
Extraneous variable: needs to be controlled
A has an affect on B
Validity and reliability
Study is valid if it measures what it claims to measure
Measurement instrument accurately measures what it is supposed to measure
Two main types:
internal validity: accurately measures - cause or relationship
external validity: generalisable (can we generalize the results with the population?)
is there a relationship between group A and group B?
REMEMBER!!
The appropriate research design is the one that will best answer the research question
Define POPULATION:
Well-defined group with specified characteristics
ALL individuals researcher interested in studying
Target population: the population to study/generalise findings to
Inclusion for Sampling Criteria:
Identify how individuals selected
Stated characteristics ensuring homogeneous group of subjects
Should be common characteristics
Enable generalisability of findings
EXCLUSION CRITERIA:
Characteristics that deem a participant inappropriate for inclusion in the study
May be stated or implicit
eg those cognitively impaired or with language issues
Sampling is:
Process of selection of study subjects
Represents entire population
Purpose of Sampling:
Increase efficiency of study
Maintain representativeness
Minimise bias (quantitative)
Represent the characteristics of interest of target population
It is important to Understand Sampling Because You Want to Know:
Does the study address the population about whom we have a clinical question?
Does the sample accurately reflect /form a part of the population addressed in the study?
Does the sampling approach limit the usefulness of the findings?
Sampling definitions:
Sample: subset of overall population
Element: most basic unit
Sampling frame: list of population elements/individuals
Sampling examples:
Population: All Childcare Centres CCCs (in Australia)
Sample: 14 (CCCs) in Perth
Element: each CCC
Sampling frame: List of all registered CCCs in Perth
Probability Sampling:
Objective
Includes:
- simple random sampling - stratified random sampling - cluster sampling - systematic sampling
Distinguishing Characteristics of Probability Sampling:
Each element has possibility of being chosen to take part in study
The probability of being chosen is unknown
All members of the population are physically present or listed in sampling frame
Each element appears only once in sampling frame
Advantages of sampling
No researcher bias
Maximises representativeness
Summary of Probability Sampling:
Primary characteristic:
a random selection of elements from the target population- each element has an equal & independent chance (probability) of being selected in the sample
Simple Random Sampling:
All population elements identified
Random number generation to select sample
Disadvantages of simple random sampling:
Time consuming & inefficient
Difficult to obtain complete list of population
Stratified Random Sampling:
Aim: to increase representativeness of sample
Population divided into homogeneous subsets (strata)
Elements randomly selected from strata
Proportional stratified: sample size from each strata proportional to size of strata
Proportional Stratified Sampling
The undergraduate population of students = 5% Aboriginal
20% Asian-born
75% Caucasian
Then the proportionate sample of 100 students = 5,20,75
Often used if comparisons between strata of unequal membership wanted
Can be stratified according to any number of attributes eg gender, age, ethnicity, religion
Similar to quota sampling
Cluster Sampling:
Successive random sampling of units/clusters
Progresses from large to small
Must all meet sample eligibility criteria
Clusters may be selected by simple random or stratified random methods
Can make it easier to obtain a random sample
eg.
1st stage = states of Australia
2nd stage = cities, suburbs, street blocks
3rd stage = households
Systematic Sampling
Involves selection of members of a population for a sample at fixed intervals i.e. every “nth” case
Similar to simple random sampling- more convenient & efficient
Individuals to be chosen must NOT be listed in some order that creates bias
eg.
Population = 5000
Select every 100th case on list (sampling frame)
NB case 1= randomly selected
eg if 73 is first no. randomly selected, then those matching 173, 273, 373 etc included in sample
Sample Size: Power Analysis
Quantitative: sample size determined prior to starting study
No of factors influence sample size eg design, need for generalisability, cost
Power analysis: mathematical strategy
Allows researchers to determine sample size needed
Allows researchers to determine how large a sample size is need
To determine sample size needed to detect a real relationship or difference in the phenomenon under study if it exists
Study suitable “powered” provides confidence in interpreting findings
Sample Size: Qualitative Research:
Qualitative studies do NOT usually begin with predetermined sample size
No formal criteria for determining sample size
Therefore no rules for determining when small/large enough
Richness of data more important
Around 8-15 but variable
Data saturation: point at which data repetitive- no new information received
Non-Probability Sampling:
Subjective judgements contribute to choosing sample
Every member of population does NOT have equal chance
Less rigorous than probability sampling
Limits generalisability
Main types:
- convenience - quota - purposive - snowball - theoretical
Non-Probability Sampling: Quantitative Research
Small exploratory studies
Useful when information on total population unknown/unavailable
Less rigorous
Tends to produce less accurate & less representative samples
Limits generalisation
Convenience Sampling
Uses most readily available subjects
Common in clinical research
Advantages: easy & inexpensive
Disadvantages: increased risk of bias
Self-selection also leads to bias
Snowball Sampling
Type of convenience sampling
Use original participants contacts: social networks
Used when difficult to otherwise identify participants
Useful for accessing diversity of experience
Quota Sampling
Goal: to make sample more representative
Used to access different subgroups of as population
Characteristics important to study used to set quotas
eg.
5,000 nurses in a city:
20% diploma (100) 40% certificate (200) 40% degree (200) = total 500 =10% of 5,000 using proportional quota sampling
Purposive Sampling
Intentional selection
Cases included- handpicked
Can be used when unusual group under study
e.g. rare genetic disease
Goal: focus on particular aspect of phenomenon
Theoretical Sampling
Usually in Grounded Theory studies
Generate data for theory generation
Starts from homogenous(small) sample
Moves to heterogenous (larger) sample
Occurs sequentially alongside data analysis
Sampling Bias
Bias: unintended factor changes results- incorrect conclusions
Distorts findings
Difficult to interpret results
Sampling Goals: Qualitative
Find best sources of data relevant to study aims/objectives
Obtain insights into the phenomenon of interest
Obtain sample representative of population of interest
Obtain sample that allows effects of specified variables to be distinguished from other variables
