MIDTERM BRUH

nursing research

The systematic, logical, and empirical inquiry that uses rigorous guidelines to produce unbiased, trustworthy, and verifiable answers to questions about nursing practice

evidence-based practice

Involves integration of a problem-solving approach within the context of caring, considering best evidence from studies, clinical experience and expertise, and understanding of pt’s preferences and values

impact of nursing research vs. the impact of EBP

Nursing research: Generates new knowledge for practice and adds to our professions’ knowledge base through the literature

EMP: translating knowledge with a goal of improving practice

Deductive reasoning

More general → More specific, Top down, sequential process moving from a general, broad interest to a very specific research question

Inductive reasoning

More specific → More general, Bottom up, where a specific observation leads to a general focus or interest (frequently the source of evidence-based practice studies)

What is a PICO question?

a tool used to frame a focused clinical or research question using the acronym P-I-C-O, which stands for Patient or Population, Intervention, Comparison, and Outcome

PICO(T)

population/patient, how would I describe the group of patients to study?

PICO(T)

Intervention/indicator, Which main intervention, management strategy, diagnostic test, prognostic factor, or exposure am I interested in?

PICO(T)

Compare/Control, Is there a control or alternative you would like to compare to the intervention

PICO(T)

Outcome, What can I hope to accomplish, measure, improve or effect?

PICO(T):

time, What time periods should be considered?

What is sampling?  Why is it important?

• The process of selecting part of a larger group with the intent of generalizing from the smaller group, called the sample, to the larger group, the population

• Important because it is often impractical or impossible to study an entire population, making sampling a more efficient and cost-effective way to collect data

• the specific, smaller group of individuals selected from the accessible population to participate in the study

Target population

the entire group of individuals a researcher wants to study

Accessible population

the subset of the target population that is practically available to the researcher

Sampling Error 

• Difference between the sample statistic and the population parameter

• The first source is from the target population to the accessible population. The accessible population may not be representative of the target population.

• The second source is from the sampling design or selection of participants. Most research designs focus on addressing this source.

• The third source involves transitioning from the selected sample to the actual sample.

What affects sampling error?

• Heterogeneity/variance- increased heterogeneity = increased sampling error

• Sample size- increased sample size – decreased sampling error

• Sampling error tells us how precise our estimates are 

  • Small SE – more precise

  • Large SE – less precise

Point Estimates

o   the use of sample data to calculate a single value (known as a point estimate since it identifies a point in some parameter space) which serves as a best guess or best estimate of an unknown population parameter

o   Provides poor approximation of the unknown true parameters; the magnitude of the difference between the statistic and the parameter is called sampling error

Interval estimate

often shown as error bars, represent a range where the true value is likely to fall

What does a small or large standard error/confidence interval mean? 

o   Small standard error or narrow confidence interval means the estimate is more precise and likely closer to the true population value

o   A large standard error or wide CI means the estimate is less precise with more uncertainty about how close it I to true value

o   Smaller values usually come from larger samples or less variability in the the data

What do overlapping or not overlapping interval estimates mean?

o   If error bars between groups overlap, there may be no significant difference

o   If the error bars do not overlap, a significant difference is more likely

Biophysiological measures

• Objective

• Variety of technical instruments for measuring physiologic functions

• Examples: temperature, BP, labs, weights

• Limitations: Cost, require specialized training to collect and/or process, instrument malfunction/inaccuracy

Records

• Existing data are a data source

• Secondary data analysis- using data that was collected as part of a research study to answer questions other than the main study question

• Electronic health records (HER)

• Limitations: dependent on the data that was previously collected, both the content and the quality

Structured Observation

• One of the most frequent methods for collecting data is through observation of individuals or groups of individuals

• Data collected by observing

• Used to document behaviors, actions, or events

• Most often unconcealed (subject knows they are being observed)

• Observe subject in nature or laboratory

• Observations observed using videotaping, one-way mirror

• Checklists, rating scales are used to assign numeric values to the observations

• Advantages: ability to describe observed behavior as a quantitative variable

• Disadvantages: time intensive, requires considerable training, training is essential for observers because inaccurate data collection threatens the validity of research findings

Survey- Questionnaires/Self-Report

• Use questionnaires(may be one scale or contain several subscales) to elicit information from subjects

• Useful when the researcher is interested in perceptions, beliefs, attitudes, or opinions

• Subjective (self—report)- variables that cannot be directly observed or measure with physiological instruments, Ex: Pain, Quality of Life, Depression, Anxiety, Worry

• Written, online/electronic, telephone

• Rating scales, checklists, Likert scale (an ordered scale from which respondents choose one option that best aligns with their view)

• Can also conduct interviews where subjects are verbally asked open or closed ended questions  useful when more personal information is needed or only way to collect data(ex: pt is blind or cannot read)

• PROMIS- Patient Reported Outcomes Measurement Information System

• Advantages: relatively inexpensive and easy to do

• Disadvantages: Social desirability bias, Recall bias, Response bias

• Considerations: for the data collected to be valid, the measures must be both valid and reliable

• Compared to interviews, questionnaires are less expensive, more confidential, and have less interviewer bias, but provide less rich data and can have a lower response rate and more missing data

Reliability

o   Extent to which the outcomes are consistent when measure more than once

o   Consistency, Precision, Reproducibility

o   Without reliability, we have no confidence in the data we collect

o   The reliability coefficient is numerical index to describe reliability

o   A reliability coefficient of 0 indicates consistency is totally absent

o   A reliability coefficient of +1 indicates consistency is totally present

Test-retest reliability

to what degree does a person’s measure performance remain consistent across repeated testing?

Alternate-Forms Reliability

Consistency between 2 forms of the same instrument with a brief period of time in between

Internal Consistency Reliability

To what extent do the individual items that go together to make up a test or an inventory consistently measure the same underlying characteristic? Are all items in an instrument measuring the same concept or characteristic? Is the scale unidimensional (the quality of measuring a single concept)?

Interrater Reliability

How much consistency is there among the ratings provided by a group of raters

Validity

o   The extent to which the instruments used measure exactly the concept that you want them to measure?

o   Does the instrument measure what it is supposed to measure?

o   Can we make proper inferences from a given measurement?

o   Accuracy and trueness

o   Measurement validity is specifically concerned with whether operationalization and the scoring of cases adequately reflect the concept the researcher seeks to measure

o   Valid measurement is achieved when scores meaningfully capture the ideas contained in the corresponding concept

Content related validity

involves the degree to which the content of the test matches a content domain associated with the construct; typically evaluated using expert review

• Face validity: Does the instrument appear to measure what it aims to measure

• Content Validity: Does the test relate to underlying theoretical concepts?

Criterion-related Validity

correlation between the test and a criterion variable(s) taken as representative of the construct; evaluated empirically/statistically

• Concurrent validity: amount of agreement between 2 assessments

• Predictive validity: Ability to predict something it should theoretically be able to predict

• Postdictive validity: Links current measurements with previously obtained criterion scores

Construct-related validity

is the measure associated with things it should be associated with and not associated with things it should not be associated with; evaluated empirically/statistically

• Convergent validity: Degree to which the operationalization is similar to other operationalizations that it theoretically should be similar to

• Discriminant or divergent validity: Degree to which the operationalization is not similar to other operationalizations that it theoretically should not be similar to

Social desirability bias

tendency to act or report an answer in a way that the subject deems to be more socially acceptable than would be their true behavior or answer

Observer bias

tendency of an observer to “see what is expected or wanted”; knowledge of study goals or hypotheses influence researcher observations

Recall bias

inaccurate or incompleteness of recollection about past events or experiences

Response bias

individuals who complete the survey may be different from those who do not

Acquiescence bias

tendency to agree

Extreme Response bias

Naysayer- always select the negative; Pollyannas- everything is always excellent

Quantitative

• Purpose: To test relationships, assess cause and effect, and quantify data.

• Approach: Objective, numerical, measurable.

• Data: Numbers, statistics, measurable variables.

• Reasoning: Usually deductive (general theory → specific hypothesis).

• Research questions: Often begin with “What,” “How many,” “To what extent.”

• Common designs: Descriptive (e.g., surveys), Correlational, Quasi-experimental, Experimental (Randomized Controlled Trials = “gold standard”)

Qualitative

• Purpose: to explore meanings, experiences, and human perspectives

• Approach: subjective, interpretive, context-based

• Data: words, themes, observations, interviews

• Reasoning: usually inductive (specific observations  broader theories)

• Research questions: Often begin with How, Why, What is it like

• Common designs: Phenomenology (lived experiences), Grounded theory (develop theory from data), Ethnography (culture/group study), Case study (in-depth analysis of a sing case or small group)

Hierarchy of evidence

Three required properties of true experimental design

• Randomization: each subject in the study has an equal chance of being assigned to the control group or the experimental group; assumes that the potentially confounding variables will be equally distributed between the groups, minimizing bias

• Manipulation: “Doing something” to at least some of the subjects; the independent variable (predictor, experimental treatment) is manipulated

• Control: Subjects in the control group receive the usual treatment or a placebo(substance or treatment that seems to be “real” but is designed to have no therapeutic value

blinding

Both groups will be treated identically in all respects except for the intervention being tested and to this end pts and investigators will ideally be blinded to which group an individual is assigned

allocation concealment

Hides the sorting of trial participants into groups so that this knowledge cannot be exploited

intervention fidelity

Participants receive the intervention or instructions exactly as described in the study protocol

intention to treat analysis

Pts are analyzed within the group to which they were allocated, irrespective of whether they experienced the intended intervention → allows the investigator to draw accurate conclusions regarding the effectiveness of an intervention at the level of adherence in the study   

Independent variable

the variable that is manipulated, controlled, or categorized by the researcher to see how it affects another variable

Dependent variable

the outcome or effect that is measured in response to the independent variable

What makes a quasi-experimental design different than an experimental design? 

An empirical intervention (active manipulation) study used to estimate the causal impact of an intervention on target population without random assignment; there may be a control group but not necessary

confounding

Variable that influences both the dependent variable and independent variable, causing a spurious association

bias

• Any trend or deviation from the truth in data collection, data analysis, interpretation, and publication which can cause false conclusions

• Lack of randomization can affect the validity of findings

Strengths and weakness of quasi-experimental designs

• Strengths

  • Practical, less expensive, generalizable, and sometimes the only feasible alternative because these designs are more adaptable to real-world settings

  • Replication of a study by different investigators at different locations can strengthen evidence

• Weaknesses

  • Unable to demonstrate clear cause-and-effect relationships

  • Change we see in groups could be attributed to another factor other than the intervention

Difference between observational and experimental/quasi-experimental designs

• The investigator simply observes

• No interventions are carried out by the investigator; only direct observation of individuals in their natural setting

• The most important difference is the lack of manipulation by the investigator in observational designs

Cohort

• Can not show causes and effects

• Individuals with exposure of interest and individuals without exposure of interest are identified

• Groups are studied to find out which groups are more likely to develop the outcome

• Typically prospective- individuals are followed over time and data about them is collected as their characteristics of circumstances change

• Objective

  • Incidence- new cases of outcome during a specified time period

  • Relative risk- compares the rate of an outcome in 2 groups

  • Prognosis/natural history- course of a disease

• Advantages:

  • Used for research on risk factors- often mandatory when RCTs are unethical

  • Establish a sequence of events as cohort studies measure potential causes before the outcome has occurred the study can demonstrate that these causes preceded the effect

  • Single study can examine various outcome variables

• Disadvantages

  • Inefficient for rare outcomes

  • Loss of subjects to follow-up can significantly affect the outcome

  • Inability to control for potentially confounding variables

  • Sampling bias

  • Expensive and often take a long time for sufficient outcome events to occur to produce meaningful results

Case-control study

• Individuals with outcome of interest (cases) and individuals without outcome of interest (controls) are identified

• Groups are studied retrospectively to compare the frequency of the exposure to a risk factor in order to estimate the relationship between the risk factor and the subsequent outcome

• Objective:

  • Odds ratio: likelihood of developing the outcome for a person who is exposed to the factor as compared to that who is not exposed

• Advantages

  • Simple to organize

  • Good to studying infrequent events- where the outcome is rare or has a long latent period between exposures and disease, may be the only feasible approach

  • Cheap and efficient- relatively few subjects are required compared to other designs

  • Useful for generating hypotheses that can be tested in another study

• Disadvantages          

  • Limited to 1 outcome

  • Confounding variables

  • Sampling bias

  • Recall bias

  • Observation bias

Cross-sectional study

• Collect data simultaneously on both exposure and outcome at one given point in time

• Only one group is used with no reference to exposure or outcome

• Subjects are neither deliberately exposed, treated, or not treated and seldom ethical difficulties

• Data is collected once at one point in time

• Often use questionnaires or surveys

• Objective:

  • Determines if exposure is related to the outcome by comparing the prevalence of the outcome in exposed and unexposed individuals

  • Prevalence is vitally important to the clinician because it influences considerably the likelihood of any particular diagnosis and the predictive value of any investigation

• Advantages

  • Quick and cheap, no follow-up so less resources are required

  • Best way to determine prevalence and are useful at identifying associations that can then be more rigorously studied using a cohort study or RCT

• Disadvantages

  • Difficult to differentiate cause and effect from simple association

  • Often there are a number of plausible explanations for results

  • Rare conditions cannot be efficiently studied because even in large samples there may be no one with the disease

  • Sampling bias

internal validity

• Degree to which change in the dependent variable can be definitely attributed only to the independent variable, and not to extraneous variables

• Did the “exposure” cause a difference in the outcome (high internal validity) or was a difference in the outcome caused by systematic error in the study (low internal validity)

• Can be compromised by not having a control group or by having a control group that is not comparable to the exposed group in measurable or unmeasurable ways

• The 2 common sources of threat are bias and confounding variables

external validity

• Generalizability of findings of an experimental study to other people and settings

• It is the degree to which the conclusions in a study would hold for other persons in other places and at other times

• One indication that a study lacks external validity is if the sample us not representative of the target population

What is a systematic review? 

• A review that is conducted according to clearly stated, scientific research methods, and is designed to minimize biases and errors inherent to traditional, narrative reviews

• It is designed to answer a focused clinical question and employs a predetermined explicit methodology to comprehensively search for, select, appraise, and analyze studies

• The scientific rigor of this process decreases bias and is what makes a systematic review research and distinguishes it from a narrative review

What is a meta-analysis?

• A unique type of systematic review and focuses on the quantitative synthesis of research findings

• Quantitative statistical approach for systematically combining results of previous research to arrive at conclusions

• Focuses on direction and magnitude of the effects across studies

• Individual effect sizes from each study are combined and a summary or overall effect size and 95% confidence interval is reported

how to interpret a forest plot

• Vertical line = no effect (effect of control = effect of treatment)

• Box/Diamond to left of vertical line = outcome of interest more desirable in intervention than control group

• Box/Dimond to right of vertical line = outcome of interest more desirable in control than intervention group

• If 95% CI touches or crosses “no effect” line, then the difference on the outcome between the treatment and control group was not statistically significant

• Width of diamond (overall pooled effect size) is the 95% CI. If it DOES NOT touch or cross the “no effect” line, then the meta-analysis indicates a statistically significant difference on outcome between treatment and control group

Nominal

classify data into categories; gender

Ordinal

relative rankings of data; disease severity

Interval

rank on a scale with equal intervals; can be positive or negative; temperature

Ratio

ranking on scale with equal intervals and with an absolute 0; height/weight

Confidence intervals

• An interval of scores around a statistic

• Computed by adding and subtracting a specific value to a calculated statistic; adding the specific value yields an upper limit, subtracting the specific value produces a lower limit

• The confidence level is indicated by a percentage, usually 90, 95, or 99%

• If the confidence level is 95%, then 95 out of 100 confidence intervals calculated will contain or capture the population mean

Type I error

• Error of the first kind

• Alpha error

• False positive

• Reject null hypothesis that is actually true; accept premise that there is a difference when there actually is no difference

• Alpha level is usually set at 0.05 but may be set at 0.01

Type II error

• Error of the second kind

• Beta error

• False negative

• Accept null hypothesis that is actually false; accept premise that there is no difference when a difference actually exists

Statistical significance

results say nothing about clinical importance or meaningful significance of results; always determine whether statistically significant results are substantively meaningful

Clinical significance

refers to whether the results are meaningful or important in real-world practice; make a noticeable difference in pt outcomes

Emergent design

Initial plan for research cannot be tightly prescribed; researcher may need to change or shift after entering the field and beginning to collect data

Reflexivity 

Researcher is conscious of the biases, values, and experiences that they bring to the study

Purposive sampling

Researcher selects individuals and sites for the study because they can purposefully inform an understanding of the research problem and central phenomenon in the study

Data saturation

Data collection stops when no new information is being obtained and redundancy is achieved

Triangulation

Researchers make use of multiple and different sources, methods, investigators, and theories to provide corroborating evidence for validating the accuracy of their study

Memoing

a record of a researcher's conversation with the data, helping to challenge assumptions and build a deeper understanding

Bracketing

Researcher tries to set aside personal interpretation/bias

Coding

The process of aggregating the text or visual data into small categories of information, seeking evidence for the code from different databases being used in a study and then assigning a label to the code

5 common qualitative designs

o   Narrative research

o   Phenomenology

o   Ground theory

o   Ethnography

o   Case study

Narrative research

study of stories or narrative or descriptions of a series of events that accounts for human experiences

Phenomenology

describes the common meaning of experiences of a phenomenon (or topic or concept) for several individuals; the researcher reduces the experience to a central meaning or the “essence” of the experience → “lived experience”

Ground theory

researcher generates a theory that explains some action, interaction, or process

Ethnography

study of cultural or social groups based on observations and a prolonged period of time spent by the researcher in the field; the researcher listens and records the voices of informants with the intent of generating a cultural portrait

Case study

Study of a case within a real-life contemporary context or setting; follows one person