Reading a Non-Inferiority Results Table: Descriptive Statistics, Confidence Intervals, and What "Not Significant" Does Not Mean for Saline Port Locks
[Student Name]
University of Phoenix
NSG/455: Evidence-Based Practice and Statistics
Week 3 Assignment
[Instructor Name]
[Date]
The results table interpreted here is a composite teaching table modeled on published non-inferiority trials; it does not report any real study's data.
The PICOT question developed in Week 1 asks whether locking implanted ports with saline results in no higher rate of occlusion or infection than locking with heparin. The search in Week 2 found randomized non-inferiority trials addressing that question, including one in adults with cancer that used this design (Goossens et al., 2013). This paper interprets a composite results table modeled on such trials, to build the skills needed to read the real studies in Week 4. Reading a results table is not a matter of finding the p value; it is a matter of knowing which number answers the question the study asked.
The Table
The composite trial randomized 400 adults with implanted ports in an outpatient oncology setting to saline (n = 200) or heparin 100 units/mL (n = 200) locks and followed them for a median of 312 days. The table reported the following.
Baseline: mean age 61.2 years (SD 11.4) in the saline group and 60.7 years (SD 11.9) in the heparin group; 58% and 56% women; 71% and 73% receiving active chemotherapy.
Follow-up: median 312 days (interquartile range 180 to 420) in each group, about 62,400 catheter-days per group.
Primary outcome, occlusion requiring intervention: 8 of 200 (4.0%) with saline and 7 of 200 (3.5%) with heparin; risk difference 0.5 percentage points, 95% confidence interval -3.2 to 4.2; non-inferiority margin 5 percentage points; p for difference = .79.
Secondary outcome, catheter-related bloodstream infection: 2 of 200 (1.0%) with saline and 3 of 200 (1.5%) with heparin.
Baseline Descriptive Statistics
The baseline rows describe the groups before treatment. Means with standard deviations summarize continuous variables such as age: the average age was about 61 in both groups, and the standard deviations of about 11 to 12 years show that most participants were within roughly 23 years of the mean on either side. Percentages summarize categorical variables such as sex and chemotherapy status. The purpose of the baseline rows is to check whether randomization produced comparable groups; here the groups look similar, which supports attributing differences in outcomes to the lock solution rather than to differences in the patients (Polit & Beck, 2021).
Follow-up is reported as a median with an interquartile range rather than a mean, because follow-up times are often skewed: some ports are removed early, and some last years. The median tells the typical follow-up, and the interquartile range tells the spread of the middle half.
Rates per Catheter-Day
Counts and percentages depend on how long patients were followed. Rates per 1,000 catheter-days adjust for that. With 8 occlusions in about 62,400 catheter-days, the saline rate is 8 divided by 62,400, multiplied by 1,000, or about 0.13 per 1,000 catheter-days; the heparin rate is 7 divided by 62,400 times 1,000, or about 0.11. Occlusion was uncommon in both groups, which matters when judging the practical importance of any difference.
The Primary Outcome and the Confidence Interval
The risk difference of 0.5 percentage points is the saline percentage minus the heparin percentage, 4.0 minus 3.5. It is the study's best single estimate of how much more often occlusion occurs with saline. The 95% confidence interval, -3.2 to 4.2 percentage points, is the range of true differences compatible with the data. It includes zero, meaning the data do not rule out no difference, and it includes values from saline being 3.2 points better to 4.2 points worse (Grove & Cipher, 2020).
Non-Inferiority, Not Superiority
The p value of .79 comes from a test of whether the two groups differ. It is not significant, but that does not show that saline and heparin are equivalent; absence of evidence of a difference is not evidence of no difference, especially in a small trial with few events. This is the most common misreading of results like these.
The study was designed to answer a different question: is saline not worse than heparin by more than the prespecified margin of 5 percentage points? The answer comes from the confidence interval, not the p value. Because the upper limit of the confidence interval, 4.2 points, is below the margin of 5 points, the data exclude a difference as large as the margin, and saline is judged non-inferior. Reporting guidance for non-inferiority trials emphasizes exactly this: the margin must be specified in advance and justified, and the conclusion depends on where the confidence interval lies relative to it (Piaggio et al., 2012).
The margin itself deserves scrutiny. A margin of 5 percentage points means the investigators judged that saline being up to 5 points worse would be clinically acceptable given its advantages. A reader should ask whether that margin is reasonable; a clinician who believes that even 2 extra occlusions per 100 patients would be unacceptable would read the same data differently.
The Secondary Outcome
Catheter-related bloodstream infections were rare, 2 and 3 events. With so few events, no meaningful comparison is possible, and the table appropriately does not report a test. The result is reassuring that saline did not produce an obvious increase, but it cannot establish that infection rates are equal.
Implications for the Clinical Question
If real trials show results like this composite table, they would support saline as non-inferior to heparin for occlusion in implanted ports, with rare infections in both groups. The next step, in Week 4, is to appraise the actual trials and the systematic review to see whether their designs, margins and results support that conclusion for this center's patients.
Conclusion
Reading this results table required distinguishing descriptive statistics from inferential ones, recomputing rates to judge practical importance, interpreting a confidence interval and recognizing that a non-inferiority trial is answered by the position of that interval relative to a prespecified margin, not by a nonsignificant p value. Reading tables this way prevents two opposite errors: dismissing a useful finding because a p value is not significant, and treating a nonsignificant result as proof that two treatments are the same.
References
Goossens, G. A., Jerome, M., Janssens, C., Peetermans, W. E., Fieuws, S., Moons, P., Verschakelen, J., Peerlinck, K., Jacquemin, M., & Stas, M. (2013). Comparing normal saline versus diluted heparin to lock non-valved totally implantable venous access devices in cancer patients: A randomised, non-inferiority, open trial. Annals of Oncology, 24(7), 1892-1899. https://doi.org/10.1093/annonc/mdt114
Grove, S. K., & Cipher, D. J. (2020). Statistics for nursing research: A workbook for evidence-based practice (3rd ed.). Elsevier.
Piaggio, G., Elbourne, D. R., Pocock, S. J., Evans, S. J. W., & Altman, D. G. (2012). Reporting of noninferiority and equivalence randomized trials: Extension of the CONSORT 2010 statement. JAMA, 308(24), 2594-2604. https://doi.org/10.1001/jama.2012.87802
Polit, D. F., & Beck, C. T. (2021). Nursing research: Generating and assessing evidence for nursing practice (11th ed.). Wolters Kluwer.
How this NSG 455 Week 3 example is structured
The NSG/455 shelf page describes Week 3 as the week statistics reading lands: descriptives, significance and interpreting a results table. The paper reads a table the way a clinician should, from the baseline characteristics to the primary outcome to the secondary outcomes, and it recomputes key numbers so the reader can see where they come from. The section on non-inferiority is the core, because it corrects the most common misreading of results like these. Students search this week as NSG 455 Week 3, NSG455 Wk 3 or NSG/455 Wk 3; all three are the same assignment.
NSG/455 Week 3 questions, answered
What does NSG/455 Week 3 usually ask for?
The course shelf describes Week 3 as statistics reading: descriptive statistics, significance and interpreting a results table. Many sections ask you to interpret the statistics in a research article or table related to your clinical question and explain what they mean for practice.
What is a non-inferiority trial?
A trial designed to show that a new intervention is not worse than the standard by more than a prespecified margin. It is used when the new intervention has other advantages, such as lower cost or fewer side effects, so it does not need to be better, only not unacceptably worse.
Does a p value above .05 mean two treatments are the same?
No. It means the study did not find convincing evidence of a difference. The confidence interval shows the range of differences compatible with the data; if that interval is wide, a meaningful difference may still exist.
Write yours, or have the desk draft it
This paper is an original model document written by our desk, not a submitted student paper and not an official University of Phoenix document. Read it for the moves, then write your own to the instructions in your classroom. If you want one built to your exact prompt and rubric, the first custom sample is free and arrives in 24 to 48 hours.