| Course | PA 502 Evidence Based Medicine |
|---|---|
| Week | 3 |
| Paper type | NNT and NNH calculation |
| Length | About 740 words, 5 pages |
| Format | APA 7 student paper |
| School | Chamberlain University |
| Program | Master of Physician Assistant Studies (MPAS) |
| Updated | October 2026 |
Free sample paper for PA 502 Week 3
Number Needed to Treat and to Harm: Intensive Versus Standard Systolic Targets in SPRINT, Calculated From the Intention-to-Treat Counts
Student Name
Chamberlain University
PA 502: Evidence Based Medicine
Instructor Name
Month Day, Year
Number Needed to Treat and to Harm: Intensive Versus Standard Systolic Targets in SPRINT, Calculated From the Intention-to-Treat Counts
The Question and the Source
The aim here is two figures: how many patients must be managed to the tighter systolic goal of 120, instead of the usual 140, for one major cardiovascular event to be avoided, and how many for one extra serious adverse event to occur. The source is the Systolic Blood Pressure Intervention Trial, which randomized 9,361 adults at increased cardiovascular risk without diabetes and followed them for a median of 3.26 years before stopping early for benefit (SPRINT Research Group, 2015). All counts below come from the intention-to-treat analysis, which counts every randomized participant in the arm they were allocated to, whatever they later took. That choice matters: a per-protocol figure would describe people who stayed on their assigned treatment, not people who were offered it, and the question at the bedside is about offering it.
Event Rates, Each as a Count Over Its Total
Primary outcome, a composite of myocardial infarction, other acute coronary syndromes, stroke, heart failure or death from cardiovascular causes:
Intensive target: 243 events among 4,678 participants, a rate of 5.19%.
Standard target: 319 events among 4,683 participants, a rate of 6.81%.
The Difference and the Number Needed to Treat
Absolute risk reduction: 6.81% less 5.19%, or 1.62 percentage points over a median of 3.26 years.
Number needed to treat: the reciprocal of 0.0162 is 61.8. Rounded up, as numbers needed to treat always are, the result is 62.
Interpretation: treating about 62 people resembling the trial's participants to the tighter goal for roughly three and a quarter years lets one of them avoid a major cardiovascular event. The period stays attached to the number: an NNT of 62 over three years is not the same as 62 over one year or ten.
The Harm Side, From the Same Table
The trial reported serious adverse events in the same intention-to-treat groups. Four were more common with the intensive target.
Table 1
Numbers Needed to Harm for Serious Adverse Events, Intensive Versus Standard Target
| Serious adverse event | Intensive, count out of 4,678 | Standard, count out of 4,683 | Absolute risk increase | Number needed to harm |
|---|---|---|---|---|
| Hypotension | 110 (2.35%) | 66 (1.41%) | 0.94 points | 107 |
| Syncope | 107 (2.29%) | 80 (1.71%) | 0.58 points | 173 |
| Electrolyte abnormality | 144 (3.08%) | 107 (2.28%) | 0.79 points | 127 |
| Acute kidney injury or failure | 193 (4.13%) | 117 (2.50%) | 1.63 points | 62 |
Note. Note. Counts from the SPRINT intention-to-treat analysis; numbers needed to harm rounded up; all over a median of 3.26 years.
For every 62 people treated to the lower target for about three years, one major cardiovascular event would be prevented and, on average, about one additional serious episode of acute kidney injury would occur, along with smaller increases in hypotension, syncope and electrolyte problems. Injurious falls were not more common with the intensive target in the trial.
Rounding in One Direction
Numbers needed to treat and to harm are rounded up, never down. Rounding 61.8 down to 61 would overstate the benefit slightly; rounding it up to 62 states the benefit conservatively. Applying the same rule to harms keeps the comparison fair.
Older Participants
A separate analysis of the 2,636 SPRINT participants aged 75 or older, the group closest to my patient, reported 102 primary outcome events among 1,317 participants with the intensive target and 148 among 1,319 with the standard target over a median of 3.14 years (Williamson et al., 2016). That is 7.74% against 11.22%, an absolute risk reduction of 3.48 percentage points and a number needed to treat of 29. Because older participants had a higher baseline risk, the same relative effect produced a larger absolute benefit. The overall rate of serious adverse events in this subgroup was similar between groups, about 48% in each.
What These Numbers Cannot Tell
These numbers apply to people who resemble the trial's participants, measured the way the trial measured blood pressure, over about three years. Whether they apply to my 76-year-old patient is the question for the week 5 applicability write-up. And the composite outcome counts a heart failure admission and a cardiovascular death as one event each, though a patient would not weigh them equally.
References
SPRINT Research Group. (2015). A randomized trial of intensive versus standard blood-pressure control. New England Journal of Medicine, 373(22), 2103-2116. https://doi.org/10.1056/NEJMoa1511939
Williamson, J. D., Supiano, M. A., Applegate, W. B., Berlowitz, D. R., Campbell, R. C., Chertow, G. M., Fine, L. J., Haley, W. E., Hawfield, A. T., Ix, J. H., Kitzman, D. W., Kostis, J. B., Krousel-Wood, M. A., Launer, L. J., Oparil, S., Rodriguez, C. J., Roumie, C. L., Shorr, R. I., Sink, K. M., . . . Pajewski, N. M. (2016). Intensive vs standard blood pressure control and cardiovascular disease outcomes in adults aged 75 years or older: A randomized clinical trial. JAMA, 315(24), 2673-2682. https://doi.org/10.1001/jama.2016.7050
The Chamberlain PA 502 Week 3 instructions
In the third week you calculate the number needed to treat and, where relevant, the number needed to harm from a published trial. Say which column of results you used and why, since intention-to-treat and per-protocol figures answer different questions. Write each event rate as a count over its own total so a reader can repeat the arithmetic. Show the absolute difference, the reciprocal and the rounded result. Always round up, and keep the follow-up period attached to every number. Take the harm figures from the same table and population as the benefit. Interpret the results in plain sentences, and note what the numbers cannot tell a clinician about an individual patient.
How the PA 502 Week 3 example is put together
The calculation uses SPRINT's intention-to-treat analysis: 243 of 4,678 participants with the intensive target and 319 of 4,683 with the standard target had a primary event, rates of 5.19% and 6.81%. The difference of 1.62 points gives an NNT of 61.8, rounded up to 62 over a median of 3.26 years. A table calculates harms from the same groups: numbers needed to harm of 107 for hypotension, 173 for syncope, 127 for electrolyte problems and 62 for acute kidney injury. Injurious falls were not more common. The older subgroup's 102 and 148 events give an NNT of 29. A closing section notes that composites count unequal events equally.
Where the points sit in the PA 502 Week 3 rubric
More points go on the wrong column than on any arithmetic error. A per-protocol result is easier to find and reads better, but it describes people who completed treatment rather than people who were offered it. The second heavy loss is rounding down, which overstates benefit. Numbers needed to treat with no period attached, harms taken from a different population or time frame and event rates given as percentages without counts also cost points. Strong calculations name the column and why, show counts over totals, show each step, round up, keep the period attached, take harms from the same table and interpret both sides plainly. A short note on what the numbers cannot say about one patient completes the work and is often what lifts a good answer to a strong one.
Common PA 502 Week 3 mistakes, and how to avoid them
Number needed to treat calculations are simple arithmetic, but the choices around them, which column, which period, which harms, are where marks are won or lost. If you share the trial your instructor assigned, we can help you find the intention-to-treat counts, lay out each step and calculate harms from the same table. We can also help you write interpretations a patient could follow. Your calculations remain your own. If the trial reports only hazard ratios and event rates per year, we can help you work out whether a number needed to treat can be calculated honestly from what is given, and say so plainly if it cannot.
Get a PA-502 Week 3 example written to your instructions
Send the article your section named along with the assignment instructions, since every figure on the page runs off that specific table and a calculation built on another study returns numbers you could not reproduce. A custom PA-502 Week 3 calculation is written to your rubric and returned within 24-48h, first one free.
More PA 502 papers and MPAS samples
- PA 502 Week 1: Clinical Question Post
- PA 502 Week 2: Bedside Question Brief
- PA 502 Week 4: Surrogate Outcome Critique
- PA 502 Week 5: Applicability Write-up
- PA 502 Week 6: Guideline Comparison
- PA 502 Week 7: Decision Aid Draft
- PA 502 Week 8: Journal Club Brief
- PA 501 Week 3: State Practice Act Analysis
PA 502 Week 3 questions, answered
What if the paper only reports a relative figure?
Then the event rates have to be found further down, usually in the results table rather than the abstract, and the calculation is built from those. Where a paper truly prints no counts anywhere, say so and name what is missing. Sections differ on whether an estimate read off a graph is acceptable, so check yours before taking numbers from a figure.
Does the calculation need a confidence interval around the number?
Many sections want one and it is worth including even where they do not, because an interval on the underlying difference translates into a range around the number itself. Report that range and say what its wide end would mean for somebody deciding. Where the difference crosses no effect, the upper bound stops being a finite figure, and that fact needs a sentence of its own.
Can I use invented figures if my section did not assign a paper?
Some sections supply a constructed table on purpose, and an example built on one works the same way. What has to hold is that the numbers behave like real ones: totals matching the groups described, an event count smaller than its total, and an adverse event rate that is not identical to the benefit rate. Say in a line where the figures came from.
Where can I find a free PA 502 Week 3 sample paper?
This page has a full PA 502 Week 3 calculation of numbers needed to treat and harm from SPRINT, with each step shown.
What NNT does the PA 502 Week 3 sample calculate?
62 over a median of 3.26 years for the whole trial, and 29 over about three years for participants aged 75 or older.