NR-612 Week 3 Documenting Implementation Fidelity: How to Write It

What NR-612 Week 3 asks for

The most useful artifact in any project audit is the delivery log, and the most common reason an audit stalls is that nobody kept one. A reviewer asks how many eligible patients received the intervention and the answer arrives as a range with an apology attached. In a population health capstone, that answer costs points in two places at once: the implementation rows, which cannot be evidenced, and the interpretation rows, which then have nothing to interpret against. The stage of the session where delivery is running is therefore also the stage where the written record has to be built, not remembered.

Fidelity has four written components, and a strong document addresses each by name. Reach is the proportion of the eligible population who received the intervention at all. Dose is how much of it they received compared with what was specified. Adherence is whether it was delivered as designed, in content and in sequence. Adaptation is what was deliberately changed during delivery, and why. Students who write only about reach produce a thin document; students who write about all four produce something a grader can score across the whole implementation band.

The second thing this stage asks for is honesty about drift. Interventions bend once real staff deliver them inside real workflows. A five-minute conversation becomes ninety seconds when the clinic runs late. A handout designed for every eligible visit gets used on the visits where a specific nurse is working. Drift is not failure and it is not something to hide; it is a finding about implementability, and population health as a discipline cares about implementability at least as much as about efficacy. What the writing has to do is name the drift, quantify it if possible, and say what it will mean for the conclusions.

The third thing is the barrier and facilitator account. This is where a capstone report earns its systems-level voice. A barrier is not that people were busy; it is that the eligibility list arrived on Fridays while the visits it applied to happened on Mondays, so the list was three days stale every week. A facilitator is not that the team was supportive; it is that a specific role agreed to run a two-minute check at the start of each huddle, which raised the proportion of eligible visits flagged before the patient arrived. Specificity is what converts this section from filler into analysis.

Where our help stops in a practicum course

The 72 practicum hours attached to this course are your own, and so is everything that documents them. Hour logs, encounter counts, site attendance records, preceptor or mentor evaluations and signatures are never drafted, reconstructed, or estimated with help, and no part of this page should be read as an offer to do so. The delivery itself is likewise yours: the huddles you attended, the staff you trained, the patients your team reached. A fidelity account is only worth writing because it describes work that genuinely happened, and a fabricated one is worse than no project at all.

Where support is legitimate is in the shape of the written record: how to structure a fidelity account, how to present reach and dose so a reader can weigh them, how to write drift as analysis rather than confession, and how to turn a set of scattered observations into a barriers and facilitators section that reads like graduate work. Every patient detail in that writing must be de-identified before it reaches the page. When you illustrate a barrier with a case, remove names, record numbers and exact dates, and check that the remaining details would not let a colleague identify the person, particularly when the eligible group is small.

The NR-612 Week 3 method, step by step

Six moves that turn a fortnight of delivery into a document a grader can audit.

1. Rebuild the delivery record from primary traces, not memory

   Sign-in counts, materials distributed, entries in the field you specified, the huddle list. Write down which trace supports each number so your report can name its own source, and note where a number is an estimate rather than a count.

2. Report reach as a proportion of the eligible denominator you defined earlier

   Not how many people you reached, but how many of the people who should have been reached. The same denominator that governs your outcome measure governs this, and using a different one makes the two sections incomparable.

3. Compare delivered dose against specified dose in the same sentence

   The specification said one structured conversation per eligible visit; the record shows conversations occurred at roughly two of every three eligible visits. Paired like that, the gap is visible and quantified rather than implied.

4. Separate deliberate adaptation from unplanned drift

   Adaptation was decided, with a reason and a date. Drift happened. Label each occurrence as one or the other, because a grader reads them differently and so should your later interpretation.

5. Write each barrier as a mechanism rather than a mood

   State what specifically prevented delivery, at which point in the workflow, affecting roughly how many cases. Then say what was tried in response and whether it worked. Barriers with mechanisms attached are analysis; barriers described as attitudes are not.

6. Close with the implication for the evaluation you have not written yet

   One short paragraph saying what this level of fidelity will and will not allow you to conclude in a few weeks. It costs eighty words now and buys a defensible interpretation section later.

A layout and word budget for a fidelity account

Our frame for an implementation fidelity write-up, sized for roughly 1,000 to 1,300 words with a delivery table alongside. It is our own outline rather than anything the university issues, and your week's rubric outranks it wherever they disagree.

Evidence craft for implementation writing

Use a named implementation framework and stay inside its vocabulary. Published frameworks for evaluating delivery give you categories a reader already recognizes, and using one consistently is worth more than assembling headings of your own. Attribute it with its year and apply the same category labels throughout the document rather than switching terms between paragraphs.

Label estimates as estimates. Where a count came from a complete record, say so; where it is your best reconstruction from partial traces, say that instead and give the range. A capstone that distinguishes counted from estimated figures reads as disciplined, and one that presents both as facts loses credibility the moment a grader asks where a number came from.

Anchor barriers in the literature where you can. Most implementation barriers in population health work are well described, and citing a study that reports the same obstacle turns your local observation into a recognizable pattern. Keep it to one or two citations placed where they do work, and name the setting the cited study ran in so the reader can judge transferability.

Every proportion carries its base. Reach of 68 of 104 eligible visits over three weeks, not roughly two thirds. In this course the base is not decoration; it is the thing that lets a reader recalculate your figure and trust the rest of the report.

Five mistakes that cost points in this week's territory

• Writing the plan again in the past tense. An account of what was supposed to happen is not evidence that it did, and graders separate the two ruthlessly.
• Reach without a denominator. Forty-two patients received the intervention says nothing until the reader knows how many were eligible.
• Treating drift as embarrassment. Unreported drift resurfaces in the results as an unexplained flat line, and by then it looks like concealment rather than context.
• Barriers written as personalities. Staff resistance is a label; a workflow step where the flag is invisible until after the patient has left is a mechanism you can act on.
• No implication paragraph. A fidelity section that stops at description forfeits the analytic credit it was in the best position to earn.

Before you submit

• Every delivery figure names the record it was drawn from
• Reach is expressed against the same eligible denominator as your outcome measure
• Delivered dose is compared with specified dose explicitly
• Deliberate adaptations are separated from unplanned drift and dated
• Each barrier names a mechanism and a point in the workflow
• Estimated figures are labelled as estimates with a range
• No case example carries identifying detail

Questions students ask about this stage

Keep going