exam 2 Flashcards
What does the Pearson Product Moment Correlation Coefficient describe?
The direction and degree of relationship between 2 variables.
Reported as a single number that represents the direction and degree of relationship between 2 variables
Values are between –1 to 0 to +1 (.80 or higher is strong relationship)
Which of the following statements is true regarding unexpected statistical findings in a research study?
Unexpected results may indicate a flaw in the researcher’s logic.
A Type I error indicates
indicates that there is a significant difference when there is not.
Which situation will involve the use of inferential statistics?
An examination of the differences between control and experimental group scores
it is a conclusion or judgment based on evidence, judgments are made based on statistical result
examples:
-Pearson product-moment correlation coefficient
-regression
-chi-square
-t-test
-ANOVA
-ANCOVA
Which statistical test is used to predict the value of a variable when the value of one or more other variables is known?
Regression analysis
Which statistical test is used to examine the mean differences between two groups?
t-test
Identify a common challenge to successful adoption of mHealth.
Digital and health literacy.
Major concerns with mHealth technologies include the protection of human subjects, confidentiality, privacy, and data security.
True
Which statistical test is used to examine differences among 3 or more groups?
Analysis of variance (ANOVA)
Which best describes digital health?
The broad use of information and communication technologies (ICT) to support health and health-related fields.
prospective
means looking foward and is usually more accurate
major literature review
is conducted at the BEGINNING of the research process
limited review
is conducted AFTER the study is completed to identify studies published since the original literature review
cross-sectional design
collects data at one data point
longitudinal design
collects data at multiple points, “repeated measures design”
causality
cause and effect relationship between the variables
-descriptive and correlational do not examine
-use quasi-experimental and experimental to examine causality or a new intervention
multicausality
presence of multiple causes or “interrelating variables” for an effect
retrospective
means looking backward
-have no control over the accuracy of data
-studies that use chart reviews and are non-interventional
bias
a slant or deviation from the truth
-can distort findings within several elements of a research study
control
having power to direct or manipulate factors to achieve a desired outcome
-increasing control is a method to decrease bias
-improved accuracy of findings
increased control in quasi-experimental
greatest control in experimental
probability
the likelihood of accurately predicting an event
-the cause will probably result in an effect
manipulation
form of control used in quasi-experimental and experimental studies during the implementation of the intervention
-must manipulate the intervention (independent variable)
study validity
measure of the accuracy of findings
design validity
strengths and threats to the quality of a study design
Types:
-construct -measures what is supposed to ?
-internal - warrant casual conclusion?
-external - related to real world?
-statistical conclusion - relatioship between cause and effect?
construct validity
begins with the fit between the conceptual definition and operational definitions of variables and how is it measured
-conceptual definitions provide the basis for operational definitions
threats:
-inadequate definitions
-experimenter expectancies - rosenthal effect
internal validity
the study results are a true reflection of reality
threats:
-participant selections
-history
-maturation
external validity
the study findings can be generalized beyond the study
threats:
-interaction of setting + intervention
-interaction of selection + intervention
-interaction of history + intervention
statistical conclusion validity
conclusions about relationships between variables accurate reflection of the real world
threats:
-low statistical power type 2 error
-unreliable measurement methods
-intervention fidelity concerns
-extraneous variables
descriptive designs
-simple descriptive: used to examine variables in a single sample
-comparative descriptive: used to describe variables and examine differences in variables in two or more groups
correlational designs
-descriptive correlational: describes variables and examines relationships among them
-predictive correlational: to predict the value of one variable based on the values obtained for another variable
-model testing
what is an ePatient?
-uses technology to actively participate in his or her healthcare
-manages responsibility for his/her own health and wellness
what is eHealth?
emerging field in the medical informatics, public health and business, referring to health services for information - also a state of mind, a way of thinking, an attitude, to improve health care locally, regionally, and worldwide by using information and communication technology
driving forces for eHealth?
-personal computers and web
-access to health and wellness tracking tools
-evolution of the internet
-policy and legislative influences
Evolution of the Internet
web 1.0: read only
web 2.0: social web with a community feel
web 3.0: web with browsers and devices that behave more like personal assistants and search capabilities that harness user experience behavior to display content of interest including personally tailored advertising
web 4.0: mobile web
web 5.0: (emerging) intelligent systems that have the capacity to measure individual’s emotions, wants and desired based on detection and predictive algorithms
what are the quantified selfers?
-Blood Pressure
-exercise
-sleep
-dietary intake
21st century cares act
2016 - expected to increase choice and access
-required to promote interoperability of health information and prohibit information blocking
-clinical notes are on electronic info that must not be blocked and made available for free
when was the EHR adoption and patient access through personal health record portals (PHR) promoted
2009
guided discovery
developing a SHARED hypothesis and including the ePatient in creating a plan to manage care’
-a way for ePatients to interact with healthcare
challenges to eHealth
-current models may not be properly structured to support patient-centeredness
-clinicians may find it difficult to find the time and resources to fully engage with the ePatients who come with a well-prepared agenda
-patient collected data through digital health are not yet integrated into their medical record
intervention
“independent variable”
-expected to impact the dependent variable (physiological, psychosocial, educational or any combination)
-should be consistent
experimental group
this group is given the intervention/treatment and acts on the dependent variable
control/comparison group
are not exposed to the intervention and receive standard care and acts on the dependent variable
quasi-experimental designs
used when control is not possible
types:
-pretest/posttest design with comparison group - most commonly used
-posttest only design with comparison group - threats to validity and has no comparisions
experimental design
more control than quasi-experimental through randomization
types:
-pretest/posttest designs with experimental and control group
-post-test only with control group design
Randomized Clinical Trial (RCT)
-gold standard
-may use blinding
-uses large # of subjects to test a treatment’s effect & compare results with a control group who did not receive the treatment
-RANDOMIZED subjects is essential
-usually multiple geographic locations are used
subjects come from a reference population
Essential elements of experiments
-random assignment of subjects to groups
-precisely defined intervention/independent variable
-researcher - controlled manipulation of independent variable
-control/comparison group
-clearly identified sampling critieria
-carefully measured dependent variable/outcome
2020 CARES act
coronavirus aid relief
EHR Interoperability and Standards
- vocabularies and terminologies
- data or document content
- transport of messages
- privacy and security
- unique identifiers
administrative applications
Registration systems:
demographics, insurance, procedures, allergies
Financial applications:
billing claims
Other:
staffing, scheduling, human resources
ancillary applications
labratory
radiology
pharmacy
cardiology
resp. therapy
physical therapy
Computerized Provider Order Entry (CPOE)
- Demonstrates reduction in medication errors
- allows orders to be entered, processed, tracked, updated, completed
- uses alerts to assist in decision support
- can flag abnormal results or reminders
Specialty (niche) Applications
information systems:
-maternity
-newborn
-surgical
-ER department
clincial decision support
Alerts:
-duplicate orders
-allergies
-medication dosing errors
-changes in patient condition
-abnormal labs
-diagnostic test results
-reminders
DANGER of alert fatigue
EHR benefits
-benefits all healthcare providers
-cost savings
-access to patient info
quality and safety
EHR challenges
-privacy and confidentiality
-standard language
-documentation burden
-consumer access EHR
-ownership
-patient-generated
-data integrity
P-Value
the probability that a difference is due to chance alone - not standard error
-results are statistically significant between 2 or more groups
-significant IF < .05
EXAMPLE:
*r = 0.56 (p = 0.03) - moderate relationship statistically significant
*r = –0.13 (p = 0.2) - weak relationship, not statistically significant
*r = 0.65 (p < 0.002) - moderate relationship, statistically significant
Type 1 error
Reject Ho when Ho is true
the results indicate that there is a SIGNIFICANT DIFFERENCE, when in reality there is not
Type 2 error
Fail to reject Ho when Ho is false
the results indicated that there is NO significant difference when there is a difference
-more likely to occur where there is a small sample size
Pearson Product Moment Correlation Coefficient (r)?
-measure of strength of a linear association between 2 variables
-reported as a single number that represent the direction and degree of the relationship
-1 to 0 to +1
.80 or higher is a STRONG relationship
Regression analysis
used to predict the value of one variable when the value of one or more other variables is known
multiple regression - predicting 2 values of variables??
Chi-square test of independence
determines whether 2 variables are independent of related
look at the X2 number and p-value
T-test
analyzes the difference between 2 means
There is a significant difference between the mean of 2 groups
independent t-test
paired t-test
ANOVA
analysis of variance
a statistical test for testing mean differences among 3 or more groups
ANCOVA
analysis of COvariance
-allows the researcher to examine the effect of a treatment apart from the effect of one or more cofounding variables
Significant and predicted results:
-Agree with those predicted by the researcher
-Support the purpose, questions or hypotheses, variables, framework, and measurement tools
-Must consider the possibility of alternative explanations for the positive finding
Nonsignificant results
called “negative” results and may be a true reflection of reality
-the reasoning or theory used to develop the hypothesis is in error
-may stem from a type 2 error
-the study failed to find any relationships or differences
mixed results
-most common outcomes
-one variable may uphold predicted characteristics and another one does not
connected health
mobile health AKA mHealth is a catalyst for healthcare change
Digital health
includes mHealth, telehealth, and telemedicine
-connects and empowers people and populations to manage health and wellness, augmented by accessible and supportive provider teams working within flexible, integrated, interoperable, and digitally enabled care environments that leverage digital tools, technologies and services to transform care delivery
MHealth
the WHO’s global observatory for eHealth - “medical and public health practice supported by mobile devices like phones, patient monitoring devices, personal digital assistants, and other wireless devices “
Tools: mobile phones with video capability can be used for telehealth to deliver care at a distance
Applications: mobile apps are important tools for phone and tablet devices
Sensors: provide links from mobile phones to an external device for longitudinal data collection or patient monitoring - data can be synced to mobile devices for monitoring (AliveCor - heart monitor )
Client education and behavior change
Focuses on improving knowledge, modifying attitudes, and supporting behavior change
Examples in Healthcare Settings: Smoking cessation programs, Medication adherence support, Appointment reminders
Data collection and reporting
Allows data to be directly deposited into central servers from mobile device
Examples of Functions: SMS, Voice communication, Digital forms
Examples in Healthcare Setting: Health surveys, Disease registration
Electronic health records
Examples of Function: Digital forms, Mobile web (WAP/GPRS)
Examples in Healthcare Setting: Personal and healthcare facility-based
Driving Forces in digital health
- technology (access, reduced cost, ad its increasing functionality)
- the consumer/patient engagement movement
- global health or connected health (expanding health care services)
- research, policy, and business (cost savings and earnings potential)
benefits to mHealth
potential to address and overcome disparities i health service access, shortage of healthcare providers, health inequities, and high costs for healthcare
Direct Measures (concrete)
weight
blood pressure
temperature
Indirect measures (abstract)
pain
coping
depression
Content Validity
the measurement method or scale includes all major elements or items relevant to construct being measured
Construct Validity
whether the instrument is actually measuring the construct (which examines the fit between conceptual and operational definitions)
