HS860 Unit 8 Doctoral Project Feasibility Appraisal Example

The Calendar, the Count and the Fields: A Unit 8 Feasibility Appraisal of a Lymphedema Surveillance Doctoral Project

[Student Name]

Purdue University Global

HS860: Doctoral Project I: Project Proposal

Unit 8 Feasibility Appraisal

[Instructor Name]

[Date]

The cancer center, record fields and counts are composite and shown as placeholders; no project data have been collected, and the sources cited are real.

Why Feasibility Now

The committee will approve a proposal it believes one candidate can complete. A feasibility appraisal tests the project against the constraints that most often stop doctoral work: time, numbers and data, the practical side of the feasibility criterion researchers apply to any question (Hulley et al., 2013). Each test below states its question, the evidence used to answer it and the result.

Test One: The Calendar

Question: can the project be completed within the program's remaining time?

Evidence: the program allows [14] months from proposal approval to final defense. The site access memo estimated four to six months from approval to the first extracted record, mostly waiting for the analytics office. Interviews depend on the records results, as the sequential design requires (Ivankova et al., 2006), and will take about three months including recruitment. Analysis, writing and committee review need about four months.

Result: pass, with little margin. The schedule works if data extraction takes no more than five months. The candidate will submit the extraction specification to the analytics office for a non-binding estimate as soon as the board submission is made, so that delays are known early.

Test Two: The Count

Question: are there enough eligible women to answer research question one with useful precision, and enough to recruit for question two?

Evidence: the breast program's operative log, reviewed in aggregate with permission, shows about [95] axillary lymph node dissections a year, so two years should include about [190] women. After excluding women treated at other centers for part of their care and women who died within four months of surgery, about [170] should remain.

Result: pass. With about [170] women, the proportion completing a step can be estimated within roughly seven percentage points either way. For interviews, if the largest loss occurs at the preoperative step, as local figures suggest, well over [100] women will be eligible to approach, more than enough for the recruitment floor of [10].

Test Three: The Fields

Question: does the record capture each step of the pathway in a form that can be extracted?

Evidence: the candidate reviewed the record build with an analyst, without viewing any patient data. The surveillance referral is a structured order with a date. Lymphedema clinic visits are structured scheduling records with dates and visit types. Arm measurements taken at the lymphedema clinic are recorded in a structured flowsheet. But preoperative measurements, when taken in the surgical clinic, are recorded in the surgeon's free-text note, sometimes as circumference values and sometimes only as a statement that measurements were done.

Result: partial fail. Two of three steps can be extracted automatically. The preoperative step would require reading about [170] surgical notes by hand, interpreting inconsistent wording.

A step recorded only in free text is not impossible to study, but it changes the study from counting to reading, and reading takes a different budget of time.

The Decision

Three options were considered. Reading every surgical note would add an estimated [60] hours and introduce interpretation errors, which would need a second reader. Dropping the preoperative step would remove the step local figures suggest is most often missed. The chosen option narrows rather than drops: the candidate will read the notes for a random sample of [60] women, with a second reader checking [20] of them, to estimate the proportion with a documented baseline. The other two steps will be counted for all women. Research question one is revised to say that the baseline proportion is estimated from a sample. The note reading will use a short coding sheet with three categories, measurement values recorded, measurement stated without values, and no mention, so that the two readers apply the same rules and their agreement can be reported.

What Was Cut

The original plan included a secondary aim to compare arm volume at lymphedema diagnosis between women who had and had not attended surveillance. Volume at diagnosis is recorded only for women seen in the lymphedema clinic, and the comparison would mean tracking every woman for the full two-year risk window. That aim is cut. It would also have required a larger sample than the center can supply to detect a meaningful difference.

The Remaining Eleven Months

Months one and two: board review and extraction request. Months three to six: extraction, note reading for the sample, and analysis of question one. Months six to eight: interview recruitment and data collection. Months eight to ten: analysis and integration of the two strands in a joint display (Creswell & Plano Clark, 2018). Month eleven: writing and committee review.

Limits of This Appraisal

Estimates of time are based on the analytics office's past turnaround, which varies. The count assumes the next year's surgical volume matches the last. The field review examined the record build, not real records, so some data may be missing in ways the build does not reveal.

References

Creswell, J. W., & Plano Clark, V. L. (2018). Designing and conducting mixed methods research (3rd ed.). SAGE Publications.

Hulley, S. B., Cummings, S. R., Browner, W. S., Grady, D. G., & Newman, T. B. (2013). Designing clinical research (4th ed.). Lippincott Williams & Wilkins.

Ivankova, N. V., Creswell, J. W., & Stick, S. L. (2006). Using mixed-methods sequential explanatory design: From theory to practice. Field Methods, 18(1), 3-20. https://doi.org/10.1177/1525822X05282260