39 th National Conference on Pediatric Health Care March 19-22, 2018 CHICAGO Disclosures Designs for Research Studies and Evidence Based Projects: How Do I Get Started? The presenters report no financial interests or potential conflicts of interest. Regena Spratling, PhD, RN, APRN, CPNP PC Assistant Professor, Director of PhD in Nursing Program Georgia State University Learning Objectives Discuss strategies for developing, implementing, and analyzing a research study or evidence based project. Interpret study and project results for statistical significance and clinical significance. Study & Project Design? Great question! Analyze approaches to study and project design in PhD and DNP education, and the contribution to pediatric research, evidencebased practice, and quality improvement. Study & Project Goal Study & Project Feasibility Population & Sample Timeline Resources 1
Sample Timeline & Resources Clinic patients total Clinic patients school age Eligible participants Study participants total Pilot Study Sample & Participants Timeline It always takes longer than you expect Hofstadter's Law Resources Funding Time for study/project Research Assistants Administrative Support Statistician 0 50 100 150 200 250 Spratling, R., Minick, P., & Carmon, M. (2012). The experiences of school age children with a tracheostomy. Journal of Pediatric Health Care, 26(2), 118 125. Study & Project Ethical Approvals Review process Institutional Review Board (IRB) Committee or board review Documentation of review Conduct study or project Funding Presentations Publications Resource: Foote, J. M., Conley, V., Williams, J. K., McCarthy, A. M., & Countryman, M. (2015). Academic and Institutional Review Board Collaboration to Ensure Ethical Conduct of Doctor of Nursing Practice Projects. Journal of Nursing Education, 54(7), 372 377. doi:10.3928/01484834 20150617 03 Other Considerations for Study & Project No double dipping (e.g. work & school) Your colleagues, students, and/or patients Clear delineation between roles Work (Nurse/NP) Student PI/Project Director Faculty & student Credentialing Student Research/project nurse Other considerations Instruments & surveys Reliability & validity Permission & costs Participant burden Cost effectiveness Study or project site participation Research Study Generate new knowledge Validate existing knowledge based on a theory Study or Project Evidence Based Practice (EBP) Project Translate the evidence Applying evidence to clinical decision making 2
EBP Types of EBP Projects Evidence Based Practice (EBP) External evidence (e.g. research, theories, experts) Clinical expertise (e.g. internal evidence generated through outcomes management, quality improvement [QI], & EBP projects) Patient preferences & values Evidence Based Practice (EBP) Project The DNP project prepares expert clinicians to generate internal evidence Translation into clinical practice or policy The goal is to positively influence health care, & patient & policy outcomes Quality improvement Improve future services Healthcare delivery innovation Design & evaluate new care/system methods Implement & evaluate EBP guidelines Healthcare policy analysis Describe or prescribe policy implementation Program development and evaluation Identify gap, develop program, & evaluate outcome Melnyk, B. M. & Fineout Overholt, E. (2015). Evidence Based Practice in Nursing & Healthcare (3 rd ed.). Wolters Kluwer: Philadelphia, PA. Melnyk, B. M. (2013). Distinguishing the preparation and roles of Doctor of Philosophy and Doctor of Nursing Practice graduates: National implications for academic curricula and health care systems. Journal of Nursing Education, 52(8), 442e448. Moran, K., Burson, R., & Conrad, D. (2017). The Doctor of Nursing Practice Scholarly Project: A Framework for Success (2 nd ed.). Jones & Bartlett Learning: Burlington: MA. Quantitative Designs Quantitative Qualitative Mixed Methods Research Descriptive Describe and summarize data Correlational Relationship among variables Surveys Experimental Randomized control trials Intervention Random sample, post test Quasiexperimental Convenience sample, pre test & post test Quantitative Qualitative Designs Prospective correlational Qualitative component Longitudinal component Moore, L. C., Clark, P. C., Shih Yu, L., Eriksen, M., Evans, K., & Smith, C. H. (2013). Smoking Cessation in Women at the Time of an Invasive Cardiovascular Procedure and 3 Months Later. Journal of Cardiovascular Nursing, 28(6), 524 533. Moore, L. C. (2011). Factors that influence smoking cessation in women who experience an invasive cardiovascular procedure. Available from Proquest Dissertations & Theses A&I. Ethnography Description and interpretation of cultural behavior Long periods of immersion and intimacy with that culture Phenomenology Understanding day to day lived experiences Description and/or interpretation Grounded theory Understanding actions by focusing on the main concerns Concepts emerge and theory is developed Constant comparison Case studies Narrative analysis Descriptive qualitative Munhall, P.L. (2010). Nursing research: A Qualitative Perspective (5 th ed.). Jones and Bartlett Learning: Burlington, MA. 3
Mixed Methods Designs Quantitative data can support qualitative research Identifying participants or outlying cases Providing baseline information to help researchers select patients to interview Qualitative data can support quantitative research Developing conceptual model or instrument Understanding barriers & facilitators Interpreting, clarifying, describing, & validating quantitative results Evidence Based Project Quality improvement Implemented and facilitated adoption of a practice change Local problem Doorenbos, A. Z. (2014). Mixed Methods in Nursing Research : An Overview and Practical Examples. Kango Kenkyu. The Japanese Journal of Nursing Research, 47(3), 207 217. Bowen, C., Stanton, M., Manno, M. (2012). Using diffusion of innovations theory to implement the confusion assessment method for the intensive care unit. Journal of Nursing Care Quality, 27(2), 139 145. Evidence Based Project Quality improvement Implemented data registry to identify all adolescents 14 years of age and older with Type 1 diabetes Completed baseline transition assessment on all adolescents identified by the registry Final thoughts Study or project? Study & project quandaries Little, J. M., Odiaga, J. A., & Minutti, C. Z. (2017). Implementation of a Diabetes Transition of Care Program. Journal of Pediatric Healthcare, 31(2), 215 221. doi:10.1016/j.pedhc.2016.08.009 Intervention study Patient focused Health outcomes Intervention Study or project? Intervention and control group EBP project System focused practice change Quality indicators Single group DNP Project Quandaries Lack of faculty knowledge of EBP & QI Lack of consensus on the DNP project Lack of faculty resources for DNP projects Challenges with clinical sites for the DNP project Students scholarly writing skills Data sources Instruments Interviews Data sources Provider surveys Chart reviews Dols, J. D., Hernández, C., & Miles, H. (2017). The DNP project: Quandaries for nursing scholars. Nursing Outlook, 65(1), 84 93. doi:10.1016/j.outlook.2016.07.009 4
39 th National Conference on Pediatric Health Care March 19-22, 2018 CHICAGO Disclosures Analyzing, Interpreting and Reporting Research Results. The presenters report no financial interests or potential conflicts of interest. Monica Roosa Ordway, PhD, APRN, PPCNP BC Assistant Professor Yale University School of Nursing Learning Objectives Describe the purpose of analyzing data including do s and don ts Present common statistical tests Explain the interpretation of results Purpose of Analyzing Data Obtain usable and useful information Describe and summarize data Identify relationships between variables Compare variables Identify the difference between variables Predict or estimate outcomes Types of Data There are 4 types of data: nominal, ordinal, interval, and ratio Necessary to understand how to categorize different types of data to be analyzed Nominal: labels data with not meaningful numeric or quantitative relationship (e.g. gender, hair color) Ordinal: the order of the values is important but difference between them is not known (e.g. Likert scale of concept like satisfaction, happiness) Interval: numeric scale with known order and difference between values (e.g. temperature, time). However, there is no true zero Ratio: Like interval data with an absolute zero, can be meaningfully added, subtracted, multiplied, divided (e.g. weight, height) Classification of Analysis: Descriptive and Inferential Purpose of statistics: to summarize and reduce data for interpretation Within the context of the theoretical framework that guided the research question Descriptive statistics: organize and describe the sample characteristics Do not attempt to draw conclusions about the sample population Inferential statistics: aim to draw conclusions about an additional population based on your sample 1
Measures of Central Tendency Statistical analysis based on assumption of normal distribution (the bell curve) 3 main measures of central tendency: mean, median, mode Mean: sum of all values in the group divided by the number of values in that group Median: midpoint in the data set (half data fall above and half below the median) Mode: value that occurs most frequently Variance is the dispersion of the values collected around the measure of central tendency key to interpreting measures of central tendency Hypothesis Testing Empirically testable statements about a relationship between 2 or more variables Null hypothesis: there is no relationship between variables of interest Significance level is important as it affects the likelihood of rejecting the null hypothesis Set a priori Type I error: rejection of the null hypothesis when it is true; false positive Type II error: failure to reject the null hypothesis; false negative Sensitivity and Specificity Sensitivity: likelihood that an instrument, measurement, or medical test will correctly identify those with a particular attribute A true positive Specificity: probability that an instrument, measurement, or medical test will correctly identify the absence of a particular attribute A true negative Confidence Intervals A range of values associated with the probability that a variable will fall within that range The larger the CI, the less precise the measurement CI of 95% and 99% are most common in nursing and medical research Example: blood pressure This Photo by Unknown Author is licensed under CC BY NC SA Common Statistical Tests Used in Nursing Which Statistical Test to Choose? First identify the measurement scale of the variable Are the data normally distributed? Understand the difference between independent and dependent variables ANOVA Regression analysis evaluates the relationship between a dependent variable with specific independent variables https://www.google.co.uk/search?q=statistical+test+decision+chart&es_sm=122&source=lnms&tbm=isch&sa=x&ei=u9invbq5ncvealdygpgh&ved=0cacq_auoaq&biw=1366&bih=643#imgdii=uehrvbcn9kapfm: &imgrc=j5mkojpnhygg9m:&spf=1518497605695 2
Do Plan analysis a priori Focus on quality of data Interpret data Present limitations Do s and Don ts Don t Wait until data is collected to determine analysis plan Assume quantitative analysis is preferred method Present data in isolation Overstate findings Steps to Data Analysis and Interpretation 1. Data integrity, quality, and organization Organize all forms/questionnaires in one place Follow security measures detailed in IRB Check for completeness and accuracy Plan for missing data should have been determined in analysis plan Maintain a record of decisions regarding incomplete data or data that does not make sense (e.g. hair cortisol levels that are biologically implausible) Assign a unique identifier to each form/questionnaire Steps to Data Analysis and Interpretation Steps to Data Analysis and Interpretation 2. Data entry Hand entry Computer entry Excel spreadsheet Microsoft Access Statistical packages SPSS, R, SAS, RedCap https://www.predictiveanalyticstoday.com/top statistical software/ 3. Discussion Interpretation of data Definition: the process of attaching meaning to the data Numbers do not speak for themselves Fair and careful judgement (6 blind men and the elephant) Team approach to encourage thorough thought processing of data Who are the stakeholders? Engagement of the community? Lessons learned, ah ha moments, what does the study add that is new or supports previous studies Were findings expected or surprising Areas for future investigation; recommendations Steps to Data Analysis and Interpretation 4. Discussion limitations and conclusions Present a balance of strengths and weaknesses Discuss limitations Careful to avoid suggesting causation without a true experimental design Do not generalize without a random sample Qualitative Analysis Methods Direct interaction with individuals: 1:1 or group interaction Interviews; structured, semi structured Focus group discussion Observation Pros: richness of data and deeper insight into phenomena under study Cons: Time consuming This Photo by Unknown Author is licensed under CC BY NC ND 3
Handling data Audio/video recording vs. note taking Transcribing data Time consuming; ratio of time required is 5:1 Who should transcribe Blinded to treatment group?verbatum Language barriers Qualitative Data Consider the tone and inflection of interviewer Positive/negative continuum Certainty/uncertainty Constant comparative analysis Collection and analysis of data simultaneous Differences noted in earlier versus later interviews Analyzing qualitative data Goal is to summarize data and present results to effectively communicate the most important features Interested in the big picture Start with labeling or coding every item of information to identify differences and similarities Create coding sheet Analyzing qualitative data Content analysis Categorization of verbal or behavioral data Coding and classifying data at two levels Basic/manifest what was actually said Higher/interpretive what was meant by the response/ latent level of analysis Content Analysis Steps 1. Read transcript and take notes 2. Make a list of different types of information gathered from notes 3. Categorize the listed items 4. Identify how the categories are linked (major themes) 5. Compare and contrast categories 6. Repeat above for each transcript Content Analysis Next Steps Collect the extracts from interviews that you have put together in one category Review categories and move items from one category to another if required Confirm uniqueness of each category or ability to combine categories Check your initial notes and consider an possible excluded data This Photo by Unknown Author is licensed under CC BY NC ND 4
Atlas/ti Nvivo NUD*IST Data Analysis Software Presenting Results Review themes and categories and organize results accordingly Can be set up as a list or diagram in the beginning Themes are the main findings of the study Support themes with direct quotations Range of quotations similarities and differences between respondents Conclusion This Photo by Unknown Author is licensed under CC BY NC References Giuliano, K. K., & Polanowicz, M. (2008). Interpretation and use of statistics in nursing research. AACN Advanced Critical Care, 19, 211 222. doi: 10.1097/01.AACN.0000318124.33889.6e Clamp, C., Gough, S., & Land, L. (2004). Resources for Nursing Research (4th ed.). London: SAGE Publications, Ltd. Retrieved from http://methods.sagepub.com/book/resources fornursing research. doi: 10.4135/9780857024633 5