NURS 500 Week 5 Research Proposal Part 3, Research Methodology Example

Caregiver Teach-Back Before Discharge and Tube-Related Emergency Visits After a New Gastrostomy, Part 3: Setting, Design, Sampling, Ethics, and Data Collection

This part describes how the study proposed in Parts 1 and 2 would be carried out. It addresses the setting, the design and its threats to validity, the population and sample, the protection of human participants and the procedures for collecting data.

Setting

The study will take place at a 220-bed regional referral hospital in Southwest Virginia that serves several rural counties. Adults with a new gastrostomy are discharged from three inpatient units: a neuroscience unit that admits patients with stroke, an oncology unit that admits patients with head and neck cancer, and a general medical unit. Gastrostomy tubes are placed endoscopically by the gastroenterology service or by interventional radiology. Discharge education is currently given by the patient's assigned nurse using a demonstration and a printed booklet, and the hospital has no dedicated enteral nutrition team. The nearest emergency departments for many patients are the hospital's own and two critical access hospitals, which matters for how outcomes are found.

Design

The study will use a prospective two-group quasi-experimental design with sequential cohorts. Family caregivers enrolled in the first 12 months will receive usual discharge education and form the comparison group. During the following month, nurses on the three units will be trained in the structured teach-back protocol, and caregivers enrolled in the next 12 months will form the intervention group. Random assignment of individual caregivers was considered and rejected. Once nurses on a unit have learned teach-back, they cannot be expected to teach a randomly chosen half of caregivers the old way, so contamination would erase any difference between groups. Randomizing units was also impractical, because the three units serve different diagnoses and would not form comparable clusters.

The sequential design creates known threats to internal validity, and each is addressed. History is the main threat: a change during the second year, such as a new home health contract, could lower emergency visits for reasons unrelated to teaching. The research team will log every system change affecting discharge or home care during both years and report them. Selection is the second threat, because the two cohorts are not randomly equivalent; eligibility criteria and enrollment procedures will be identical, and patient age, diagnosis, caregiver age, caregiver relationship and distance from home to the hospital will be recorded so that Part 4's analysis can adjust for them. Instrumentation is controlled by measuring outcomes in exactly the same way for both cohorts. Shadish et al. (2002) note that a nonequivalent comparison design is strengthened when the comparison group is drawn from the same population and measured with the same procedures, and both conditions are met here.

Population and Sample

The target population is the unpaid family members and friends who take over tube care when an adult leaves the hospital with a newly placed gastrostomy. Patient inclusion criteria are age 18 or older, a first gastrostomy placed during the admission, discharge to a private home, and a plan for an unpaid family member or friend to manage tube care. Caregiver inclusion criteria are age 18 or older, identification by the patient or the patient's representative as the primary person responsible for tube care, and the ability to communicate in English or Spanish, with a professional interpreter for Spanish. Patients discharged to a skilled nursing facility or hospice, patients who will manage the tube entirely by themselves and caregivers who are paid for their care will be excluded.

The study will use consecutive enrollment, a form of nonprobability sampling in which every eligible dyad during each period is invited. This approach reduces the selection bias that occurs when nurses choose whom to approach, though it does not provide the representativeness of random sampling (Gray & Grove, 2021). Sample size was estimated for the primary question, a comparison of two independent proportions. Using the baseline rate of 24% from the hospital's 2025 records, a target rate of 10% judged clinically meaningful by the unit nurse managers, a two-tailed alpha of .05 and power of .80, a power analysis performed with G*Power 3.1 (Faul et al., 2007) indicates about 112 dyads per group. Allowing 10% for dyads lost before the day-30 follow-up, the study will enroll 125 dyads per group, or 250 in total. The hospital discharges about 150 eligible adults each year, so a 12-month enrollment period per cohort is feasible with a consent rate near 85%.

Ethics and Institutional Review Board Protection

No dyad will be approached until two review boards, the hospital's own and Liberty University's, have approved the protocol. It poses minimal risk: both groups receive education that meets or exceeds the hospital standard, and no care is withheld. Because the patient's health record will be reviewed for emergency visits, both the patient and the caregiver are participants. Caregivers will give written informed consent. Patients with decision-making capacity will consent for record review; for patients whose capacity is impaired by stroke or neurologic disease, consent will be sought from a legally authorized representative, and the patient's assent will be sought when possible. Capacity will be judged by the attending clinician, not the research team.

The study follows the three principles of the Belmont Report (National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, 1979): respect for persons through voluntary, informed consent and protection of those with diminished autonomy; beneficence through minimal risk and a plan to share the teach-back protocol with the comparison cohort's units once the study ends; and justice through enrolling every eligible dyad rather than those easiest to reach. Data will be stored under study numbers in a password-protected, encrypted database; the key linking numbers to names will be kept separately and destroyed after the analysis. As a Christian researcher, the student investigator also treats these rules as a minimum, not a ceiling. Treating each participant as one would want to be treated (English Standard Version Bible, 2001/2016, Matthew 7:12) means answering caregivers' questions honestly, never pressing a tired family to enroll and reporting results that disappoint as faithfully as those that confirm the hypothesis.

Data Collection

Data collection will follow the same sequence for both cohorts. A research assistant who does not provide discharge teaching will screen admissions daily, approach eligible dyads after tube placement and obtain consent. Demographic and clinical data will be recorded at enrollment. On the day of discharge, before the caregiver leaves the unit, the research assistant will administer the Preparedness for Caregiving Scale (Archbold et al., 1990). In the intervention cohort, the discharging nurse will also complete the teach-back fidelity checklist, which records each of the six skills and whether an accurate return demonstration was achieved.

At 30 days after discharge, the research assistant will call the caregiver using a scripted interview to ask about any emergency visits and their reasons. Emergency visits will also be identified from the hospital's record and the regional health information exchange, which includes the two critical access hospitals. Two nurse reviewers, given visit records with the cohort dates removed, will classify each visit as tube-related or not using the operational definition from Part 1, and disagreements will be resolved by a third reviewer. Part 4 describes the instruments, the measurement properties and the statistical analysis.

References

Archbold, P. G., Stewart, B. J., Greenlick, M. R., & Harvath, T. (1990). Mutuality and preparedness as predictors of caregiver role strain. Research in Nursing & Health, 13(6), 375-384. https://doi.org/10.1002/nur.4770130605

English Standard Version Bible. (2016). Crossway. (Original work published 2001)

Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39(2), 175-191. https://doi.org/10.3758/BF03193146

Gray, J. R., & Grove, S. K. (2021). Burns and Grove's the practice of nursing research: Appraisal, synthesis, and generation of evidence (9th ed.). Elsevier.

National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. (1979). The Belmont report: Ethical principles and guidelines for the protection of human subjects of research. U.S. Department of Health, Education, and Welfare. https://www.hhs.gov/ohrp/regulations-and-policy/belmont-report/read-the-belmont-report/index.html

Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-experimental designs for generalized causal inference. Houghton Mifflin.