BUSI 418 Week 4 Discussion Board Forum 5 Example

Discussion Board Forum 5: Interpreting Quantitative Results

BUSI 418 Research and Evidence Based Practice in Healthcare, Discussion Board Forum 5

Thread by [Student Name]

The hospital, its committee and the classmates are composites; the textbook, Scripture and the peer-reviewed articles cited are real.

Our hospital's safety committee is deciding whether to fund a workplace violence prevention program modeled on a published study, and our unit manager admitted she could not tell whether the study's results were impressive. Hall and Roussel (2017) explain that reading quantitative research means understanding what the statistics say about the size and certainty of an effect, not only whether the authors called it significant. Two studies make the point well, one with a clear and moderate effect and one with a very large sample and a small effect.

Kim et al. (2026) evaluated a program on four inpatient units that screened patients for aggression and added proactive rounding by a behavioral health nurse. Violent events fell from 5.30% of patients at baseline to 3.71% during the full program, a relative risk of 0.69 with a 95% confidence interval of 0.53 to 0.90. In plain terms, violent events were about 31% less frequent, and because the whole interval falls below 1.0, the reduction is hard to attribute to luck. The interval also shows the uncertainty: the true reduction could be as small as 10% or as large as 47%. For our committee, even the smaller figure would mean several fewer assaults a year on our units, which helped them see the program as worth funding.

Statistical significance is not the same as importance. Guglielminotti et al. (2023) studied more than 21 million birth records and found that Medicaid expansion was associated with a statistically significant 1.6% relative decrease in eclampsia compared with nonexpansion states, but the authors noted that its clinical significance remained to be determined. With very large samples, even small effects become significant, so a careful reader sets the p-value aside long enough to ask how big the difference is and what it would mean for real patients and budgets.

From these examples, I suggested three questions for any result: How big is the effect in real terms? How certain is it, judging by the confidence interval? Does the design allow a causal claim? Proverbs urges, whatever you get, get insight (English Standard Version Bible, 2001/2016, Proverbs 4:7). A significant result tells you the effect is probably real; only its size tells you whether it matters.

Reply to Brianna

Brianna, you wrote that a p-value below 0.05 shows an intervention works well. It tells you that a difference this large would rarely appear by luck if the intervention did nothing, but it says nothing about how well the intervention works or whether the effect is large enough to justify its cost. Rihari-Thomas et al. (2026) found in a stepped-wedge trial that a nurse-led assessment and handover intervention significantly reduced pressure injuries by 41%, while changes in falls and emergency team calls were not significant, so the same program had a meaningful effect on one outcome and no proven effect on others. Jeon et al. (2026) reported that Medicare drug price negotiation reduced projected spending by 8.8%, a figure whose importance depends on context rather than on any p-value. Hall and Roussel (2017) encourage readers to look at effect sizes and intervals. James counsels that anyone lacking wisdom should ask God, who gives generously (English Standard Version Bible, 2001/2016, James 1:5). Which outcome in your project matters most to patients, and is the program's effect on that outcome both significant and large enough to matter?

Reply to Owen

Owen, you concluded from a staff survey that heavy workloads cause nurses to quit. The association is plausible, and it matches what many nurses say, but a cross-sectional survey cannot prove cause, because it measures workload and intent to leave at the same moment. Muir et al. (2026) linked heavier nurse workloads to burnout and intent to leave across 132 hospitals, but because the data were collected at one time, other factors, such as leadership, may explain both. Xu et al. (2025) similarly reported correlations between second victim experience and turnover intention, not causal effects. Hall and Roussel (2017) describe how design determines the claims a study can support. Proverbs contrasts the discerning heart, which goes looking for knowledge, with the fool who is content to feed on nonsense (English Standard Version Bible, 2001/2016, Proverbs 15:14). Could your hospital test a workload change on one unit and compare turnover with a similar unit over a year? That design would come closer to showing cause than any survey.

References

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

Guglielminotti, J., Daw, J. R., Friedman, A. M., Landau, R., & Li, G. (2023). Medicaid expansion and risk of eclampsia. American Journal of Obstetrics & Gynecology MFM, 5(8), 101054. https://doi.org/10.1016/j.ajogmf.2023.101054

Hall, H. R., & Roussel, L. A. (2017). Evidence-based practice: An integrative approach to research, administration, and practice (2nd ed.). Jones & Bartlett Learning.

Jeon, Y. K., Hansen, R. N., Gabriel, N., Martin, K., & Sullivan, S. D. (2026). Spending impact of the first 10 drugs selected for Medicare drug price negotiation in the United States. Value in Health, 29(7), 1228-1234. https://doi.org/10.1016/j.jval.2026.03.010

Kim, S. C., Vejnovich, C., Hall, L., & Hultman, S. (2026). Workplace violence prevention program targeting patients at risk of violence in nonpsychiatric inpatient units: A prospective interventional study. Joint Commission Journal on Quality and Patient Safety. Advance online publication. https://doi.org/10.1016/j.jcjq.2026.08.010

Muir, K. J., Aiken, L. H., Brom, H., Hovsepian, V., Lasater, K. B., & McHugh, M. D. (2026). Safe minimum nurse staffing requirements for hospitals: Evidence from Pennsylvania. Medical Care, 64(7), 447-455. https://doi.org/10.1097/MLR.0000000000002332

Rihari-Thomas, J., Whittam, S., Thornton, A., Goncharov, L., Slade, D., McMahon, J., Coventry, A., Coughlan, K., Steadward, Y., Taylor, J., Liu, M., Walsh, S., Riddell, K., Tuqiri, K., Olesen, K., Dahm, M. R., Chien, L., Kelly, P., Ball, J., . . . Middleton, S. (2026). Assessment and communication excellence for safe patient outcomes (ACCELERATE): A stepped-wedge cluster randomised trial evaluating the effectiveness of a nurse-led assessment and handover communication intervention on patient adverse events. International Journal of Nursing Studies, 183, 105652. https://doi.org/10.1016/j.ijnurstu.2026.105652

Xu, Y., Jin, Q., Zhou, Q., Zhang, R., & Ding, W. (2025). Relationship between second victim experience and turnover intention in Chinese nurses: Assessing the mediating role of posttraumatic growth using structural equation modeling. Journal of Nursing Management, 2025, 7327139. https://doi.org/10.1155/jonm/7327139