| Course | IHP 525 Biostatistics |
|---|---|
| Module | Module 5 |
| Paper type | graduate milestone statistical analysis plan |
| Length | About 1,020 words, 6 pages |
| Format | APA 7 student paper |
| School | Southern New Hampshire University |
| Program | MPH |
| Updated | September 2026 |
Free sample paper for IHP 525 Module 5
Milestone Two: Statistical Analysis Plan for the Blood Pressure Program Evaluation
[Student Name]
Southern New Hampshire University
IHP 525: Biostatistics
Module Five Milestone Two
[Instructor Name]
[Date]
Milestone Two: Statistical Analysis Plan for the Blood Pressure Program Evaluation
A statistical analysis plan, written before the main analysis, commits the analyst to specific methods and protects against choosing analyses after seeing results. This milestone sets out the plan for evaluating Bayview County's community blood pressure program. It covers sample size and power, primary and secondary analyses, handling of multiple comparisons and missing data, sensitivity analyses and reporting.
Why Power Matters
Button et al. (2013) examined the consequences of low statistical power, focusing on neuroscience but with lessons for any field. Small studies have a low chance of detecting true effects, and when they do produce statistically significant results, those results are more likely to be false positives and to exaggerate the size of real effects. They argued that underpowered research wastes resources and misleads, and that sample size should be planned to give a high probability of detecting effects of meaningful size.
The Pilot Was Underpowered
Last year's pilot enrolled 30 participants and 30 comparison patients. With a standard deviation of change of about 13 mm Hg, the standard error of the difference between groups was about 13 times the square root of 2 divided by 30, or 3.36 mm Hg. To detect a true 5 mm Hg difference, the expected z statistic would be 5 divided by 3.36, or about 1.49. The power, the probability of crossing the 1.96 threshold, is the chance that a standard normal value exceeds 1.96 minus 1.49, or 0.47, which is about 32%. The pilot's p-value of 0.07 for a 5 mm Hg difference is exactly what an underpowered study of a real effect would be expected to produce, which is why it should not have been read as evidence that the program failed.
Sample Size for the Current Evaluation
To detect a 5 mm Hg difference in mean change with 80% power and a two-sided significance level of 0.05, assuming a standard deviation of 13 mm Hg, the required number per group is 2 times the square of 1.96 plus 0.84, times 13 squared, divided by 5 squared: 2 times 7.84 times 169 divided by 25, or about 106 per group. Allowing for 15% missing follow-up raises this to about 125 per group. The evaluation's 240 participants and 260 comparison patients exceed this, giving power above 95% for a 5 mm Hg difference and about 80% for a difference as small as 3.5 mm Hg.
Table 1. Power for Detecting Differences in Mean Change
| Study | Per group | Difference | Approximate power |
|---|---|---|---|
| Pilot | 30 | 5 mm Hg | 32% |
| Required minimum | 106 | 5 mm Hg | 80% |
| Current evaluation | About 220 with follow-up | 5 mm Hg | Above 95% |
| Current evaluation | About 220 with follow-up | 3.5 mm Hg | About 80% |
Note. Calculations assume a standard deviation of 13 mm Hg and two-sided alpha of 0.05.
Primary Analysis
The primary analysis compares mean six-month change in systolic pressure between groups using linear regression with group as the main predictor, adjusted for age, sex, baseline systolic pressure and number of medications. Adjusting for baseline pressure addresses imbalance and regression to the mean. The result will be reported as the adjusted difference with a 95% confidence interval and p-value. An unadjusted t-test will also be reported for transparency.
Checking Model Assumptions
Linear regression assumes that residuals are roughly normal with constant spread and that the relationship between baseline pressure and change is close to linear. Residual plots will be inspected after fitting, and a quadratic baseline term will be tested. Because each group holds more than 200 people, modest departures from normality will have little effect on the confidence interval for the group difference. If residual spread differs sharply between groups, robust standard errors will be used and reported alongside the standard results. Outlying readings above 220 or below 80 mm Hg will be checked against clinic records rather than deleted, since removing inconvenient values after seeing them is one of the choices a written plan exists to prevent.
Secondary Analyses and Multiple Comparisons
Six secondary analyses are planned: control rate, diastolic pressure, medication changes and effects in three subgroups defined by sex, age above or below 55 and recruitment setting. Each additional test increases the chance of a false positive; with six independent tests at 0.05, the probability of at least one false positive is about 26%. Bland and Altman (1995) describe the Bonferroni method, which splits the 0.05 threshold evenly across the planned comparisons, here 0.05 divided by 6, or about 0.008, to keep the overall false positive rate near 0.05. They note that the method is conservative, especially when tests are correlated, and can miss real effects. The plan will report unadjusted confidence intervals for all secondary results, mark which pass the Bonferroni threshold and describe subgroup findings as exploratory.
Missing Data
About 9% of program participants lack six-month readings, and those missing were younger with higher baseline pressure. Sterne et al. (2009) explained that complete-case analysis can be biased when missingness is related to participant characteristics and described multiple imputation as a way to use all available information, creating several complete data sets based on observed variables, analyzing each and combining the results. They cautioned that the imputation model must include the outcome and variables related to missingness and that results should be compared with complete-case analysis. The primary analysis will use twenty imputed data sets including all analysis variables plus recruitment site and number of program contacts.
Sensitivity Analyses
Three sensitivity analyses will test robustness: a complete-case analysis, an analysis assuming participants with missing data had no improvement, a deliberately pessimistic scenario, and an analysis excluding the two recruitment sites where measurement protocols differed. If conclusions hold across these, confidence in the main result increases.
Reporting Rules
All results will be reported with effect estimates and 95% confidence intervals. P-values will be reported as exact values rather than as above or below 0.05. The plan will be dated and shared with program leaders before analysis begins, and any deviations will be explained in the final report.
Conclusion
The analysis plan ensures adequate power, specifies the primary analysis, controls false positives across secondary tests, addresses missing data with multiple imputation and tests robustness through sensitivity analyses. It also explains why the pilot's ambiguous result was predictable, a lesson that should shape future program evaluations.
References
Bland, J. M., & Altman, D. G. (1995). Multiple significance tests: The Bonferroni method. BMJ, 310(6973), 170. https://doi.org/10.1136/bmj.310.6973.170
Button, K. S., Ioannidis, J. P. A., Mokrysz, C., Nosek, B. A., Flint, J., Robinson, E. S. J., & Munafò, M. R. (2013). Power failure: Why small sample size undermines the reliability of neuroscience. Nature Reviews Neuroscience, 14(5), 365-376. https://doi.org/10.1038/nrn3475
Sterne, J. A. C., White, I. R., Carlin, J. B., Spratt, M., Royston, P., Kenward, M. G., Wood, A. M., & Carpenter, J. R. (2009). Multiple imputation for missing data in epidemiological and clinical research: Potential and pitfalls. BMJ, 338, Article b2393. https://doi.org/10.1136/bmj.b2393
What the IHP 525 Module 5 instructions ask for
Milestone Two in IHP 525 usually asks for a statistical analysis plan: sample size and power, primary and secondary analyses, handling of multiple comparisons and missing data, sensitivity analyses and reporting. Graduate milestones commonly run five to seven APA 7 pages. Show your power calculation step by step with its assumptions, specify models precisely, explain how you will control false positives and describe your approach to missing data with its assumptions. Include sensitivity analyses and commit to reporting estimates with confidence intervals. Instructors value plans written as if they will be followed exactly, with no room for choosing methods after seeing results. IHP 525 graders notice clean headings in IHP 525 papers. IHP 525 names and dates need checking before IHP 525 submission.
How this IHP 525 Module 5 milestone two example is built
This milestone plans the statistical evaluation of a composite county's blood pressure program. A worked calculation shows the 60-person pilot had about 32% power to detect a 5 mm Hg difference, explaining its p of 0.07, as Button and colleagues' account of underpowered studies predicts. The current study needs about 106 per group and has power above 95%, shown in a table. The primary analysis is adjusted linear regression, six secondary tests use a Bonferroni threshold of about 0.008 following Bland and Altman, missing data are handled with twenty imputations following Sterne and colleagues and three sensitivity analyses test robustness. IHP 525 students can reuse this structure for IHP 525 work. IHP 525 claims here trace to cited IHP 525 sources.
Where the IHP 525 Module 5 rubric puts the points
Analysis plan milestones in IHP 525 are generally judged on correct power and sample size calculations, precise specification of analyses, appropriate handling of multiple comparisons and missing data, sensible sensitivity analyses, reporting commitments, scholarly support and APA 7. Strong plans state assumptions for power calculations, adjust for baseline values, explain the trade-offs of Bonferroni correction and follow best practice for imputation. Plans lose points when power calculations lack steps, many tests are run without acknowledging false positives or missing data are ignored. Linking a past ambiguous result to low power is often credited. IHP 525 marks favor careful formatting across IHP 525 sections. IHP 525 citations keep every IHP 525 argument credible.
IHP 525 Module 5 help: the mistakes that cost points
In IHP 525, analysis plans often lose points for power calculations without assumptions, for vague model descriptions, for ignoring multiple testing and for complete-case analysis without justification. A further weak spot is presenting subgroup findings as confirmatory. Calculate power with steps, specify models, control false positives, plan imputation and sensitivity analyses and commit to transparent reporting. If your study has a different outcome or design, add its details to your IHP 525 notes so the plan and calculations match. IHP 525 drafts start well from a IHP 525 outline. IHP 525 feedback already received guides IHP 525 revisions.
Get IHP 525 Module 5 written to your instructions
Send the IHP 525 Milestone Two prompt and your study details. The plan will calculate power step by step, specify primary and secondary analyses, handle multiple comparisons and missing data, add sensitivity analyses and set reporting rules, within 24 to 48 hours, free the first time. The paper above is an original model document written by our desk, not a submitted student paper and not an official Southern New Hampshire University document.
More IHP 525 papers and related MPH samples
- IHP 525 Module 1 Discussion: What a P-Value Can and Cannot Tell a Program Manager
- IHP 525 Module 2 Descriptive Statistics Paper: Describing Blood Pressure Data Without Throwing Information Away
- IHP 525 Module 3 Milestone One: Research Question, Variables and Data Plan
- IHP 525 Module 4 Hypothesis Testing Paper: A Two-Sample T-Test and a Chi-Square Test Worked Step by Step
- IHP 525 Module 6 Diagnostic Tests Paper: Sensitivity, Specificity and Predictive Values for a Home Blood Pressure Device
- IHP 525 Module 7 Milestone Three: Regression Results Adjusted for Age, Sex and Baseline Pressure
- IHP 525 Module 8 Communication Paper: Explaining Statistical Uncertainty to Decision Makers
- IHP 525 Module 9 Final Project: The Statistical Report on the Community Blood Pressure Program
- IHP 525 Module 10 Journal: Statistics as Judgment
- IHP 515 Module 3 Milestone One: Describing the Overdose Crisis by Person, Place and Time
- IHP 670 Module 7 Evaluation Design Paper: Choosing and Applying an Evaluation Framework
- IHP 640 Module 10 Journal: Measurement as a Management Discipline
- IHP 610 Module 6 Milestone Two: Comparing Policy Options Against Explicit Criteria
IHP 525 Module 5 questions, answered
Where can I find a free IHP 525 Module 5 Milestone Two sample?
IHP 525 Module 5 appears here in full: a statistical analysis plan with power calculations, Bonferroni adjustment, multiple imputation and sensitivity analyses.
How do you calculate sample size for comparing two means?
Multiply 2 by the square of the sum of the z values for alpha and power, times the variance, and divide by the square of the difference to detect.
Why are small studies unreliable?
They have low power, so true effects are often missed, and significant results are more likely to be false or exaggerated.
What is the Bonferroni correction?
Splitting the alpha level evenly across all planned comparisons to limit the overall chance of a false positive, at the cost of being conservative.
What is multiple imputation?
A method that fills in missing values several times using observed information, analyzes each data set and combines the results.