| Course | DDHA 8703 Advanced Health Analytics and Data-Driven Decision Making |
|---|---|
| Week | Week 6 |
| Paper type | Doctoral discussion post with two replies |
| Length | About 640 words, 4 pages |
| Format | Discussion post, APA 7 citations |
| School | Walden University |
| Program | Doctor of Healthcare Administration (DHA) |
| Updated | October 2026 |
Free sample paper for DDHA 8703 Week 6
Week 6 Discussion: Did the Program Cause the Drop?
The Emergency Visits Fell After Enrollment, and Would Probably Have Fallen Anyway: Three Rival Explanations for a Care Management Program's Forty Percent
Several of us have described programs that "reduced" something because a number fell after the program began, and I want to test the strongest version of that claim using one example. Northgate Health, a composite system, enrolled 220 patients with five or more emergency visits in the previous six months in an intensive care management program. In the six months after enrollment, the same patients averaged 40% fewer emergency visits. The claim, put as strongly as anyone has put it, is that a drop that large, in the patients who received the program, after it started, must reflect the program's effect. Three things could have produced the same drop without the program doing anything.
The first is who was counted. Patients were enrolled because their visits had just been unusually high, and an unusually high count is usually followed by a lower one, whatever anyone does. Morton and Torgerson (2003) describe this regression to the mean as a common source of apparent improvement when patients are selected for a high measurement. Finkelstein et al. (2020) showed how large it can be: when a widely publicized program for patients with very high hospital use was tested by randomization, readmissions fell sharply in the program group and the control group alike, and the program made no measurable difference. Patients who died or left the health plan also dropped out of the after count, which lowers the average further.
The second is a seasonal pattern. Enrollment ran from November to January, when emergency visits for respiratory illness peak, so the six months before enrollment included winter and the six months after included spring and summer. Emergency visits might have fallen for everyone over that period.
The third is attention to the measure itself. Care managers knew their program would be judged on emergency visits, and some encouraged patients to call them first, which may have routed some visits to urgent care without changing patients' health. McCambridge et al. (2014), reviewing the Hawthorne effect, found that being studied or observed can change behavior, though the size of the effect varies widely.
I am not concluding that the program failed. It may help patients in ways the visit count does not capture, and it may even reduce visits. The claim under attack is about evidence. A comparison group of similar patients who were not enrolled, followed over the same months, would tell us how much of the 40% the program can claim.
Reply to Ngozi
Ngozi, you raise a fourth explanation I had not considered: during the same months, Northgate opened two new urgent care centers near the neighborhoods where many enrolled patients live. If so, emergency visits may have fallen among enrolled patients and everyone else nearby, which would look like a program effect in a before-and-after comparison. That is a better rival than my seasonal one, because it is specific to the period and the place. It also suggests a test: comparing enrolled patients with similar patients living in other parts of the region, where no urgent care opened. Thank you for the correction.
Reply to Leif
Leif, you argue that regression to the mean cannot explain a drop this large, because patients with frequent emergency visits tend to keep having them. I agree that some patients have persistently high use, and for them the drop is more striking. Finkelstein et al. (2020), however, also selected patients for unusually high hospital use and saw readmissions decline sharply in both arms of their trial. The evidence suggests that a drop of this size is quite consistent with selection alone. I would be persuaded by the program's effect if enrolled patients fell further than a matched comparison group.
References
Finkelstein, A., Zhou, A., Taubman, S., & Doyle, J. (2020). Health care hotspotting: A randomized, controlled trial. New England Journal of Medicine, 382(2), 152-162. https://doi.org/10.1056/NEJMsa1906848
McCambridge, J., Witton, J., & Elbourne, D. R. (2014). Systematic review of the Hawthorne effect: New concepts are needed to study research participation effects. Journal of Clinical Epidemiology, 67(3), 267-277. https://doi.org/10.1016/j.jclinepi.2013.08.015
Morton, V., & Torgerson, D. J. (2003). Effect of regression to the mean on decision making in health care. BMJ, 326(7398), 1083-1084. https://doi.org/10.1136/bmj.326.7398.1083
The Walden DDHA 8703 Week 6 instructions
This thread typically asks you to examine a claim that an intervention caused an improvement. Pick a claim classmates or colleagues have made and restate it in its strongest form, so your challenge is credible. Then name specific alternative explanations that could produce the same result without the program, such as how participants were selected, seasonal or secular trends and changes in behavior caused by attention to the measure. Give each alternative a sentence of mechanism and a source. State clearly that you are evaluating the evidence rather than the program's worth. End by naming the comparison or design that would settle the question. In replies, welcome a better rival explanation if a classmate offers one, and answer objections with evidence.
Inside the DDHA 8703 Week 6 example
The post opens by restating the claim about 220 enrolled patients whose emergency visits fell 40 percent. The second paragraph explains selection at a high point and dropout from the count, citing Morton and Torgerson and Finkelstein and colleagues' randomized evaluation, where readmissions fell in both arms. The third notes that enrollment in winter means the comparison spans seasons. The fourth describes care managers steering patients toward phone calls and urgent care, citing McCambridge and colleagues on observation effects. The final paragraph says the program may still help and names a comparison group as the missing evidence. The reply to Ngozi accepts a new explanation, urgent care openings, and proposes a regional comparison. The reply to Leif answers a doubt about selection with the trial's result.
Where the marks sit in the DDHA 8703 Week 6 rubric
Graders look for a challenge that is fair and specific. Restating the claim in its strongest form earns credit because it shows the writer is not attacking a straw man. Each alternative explanation should have a mechanism and a source. Common rivals, such as regression to the mean, seasonality and observation effects, are expected, but applying them concretely to the case matters more than naming them. Explicitly separating a critique of evidence from a judgment about the program shows maturity. Naming the design that would settle the question is often decisive. Replies are evaluated on openness to better alternatives and on evidence-based responses to objections, and citations should follow APA style.
Common DDHA 8703 Week 6 mistakes, and how to avoid them
A common weakness is listing threats to validity from a textbook without connecting them to the specific program. Tie each alternative to a feature of the case, such as when enrollment happened or how patients were chosen. Avoid concluding that the program failed; the evidence may simply be unable to tell. Say what comparison would help. In replies, accept a stronger rival explanation gracefully and say how to test it. If you are asked to evaluate a program at your workplace and want a post modeled on its details, our team can draft one to your prompt.
Get a DDHA 8703 Week 6 example written to your instructions
This thread comes back written to your own Week 6 prompt and rubric, initial post and replies together, nothing charged the first time and a turnaround of 24-48 hours. The claim it contests will be an invented one. A finding circulating inside the place you work is yours to handle, and no version of it appears here.
More DDHA 8703 papers and DHA samples
- DDHA 8703 Week 1: Unused Report Discussion Post
- DDHA 8703 Week 2: Metric Specification
- DDHA 8703 Week 3: Data Source Review
- DDHA 8703 Week 4: Data Quality Audit
- DDHA 8703 Week 5: Descriptive Analysis
- DDHA 8703 Week 7: Comparative analysis
- DDHA 8703 Week 8: Visualization critique
- DDHA 8703 Week 9: Predictive model brief
- DDHA 8703 Week 10: Data governance memo
- DDHA 8703 Week 11: Decision brief
- DDHA 8601 Week 6: Interoperability Discussion Post
- DDHA 8503 Week 6: System Boundary Post
- DDHA 8303 Week 6: Single-Site Claims Post
- DDHA 8203 Week 6: Tradeoff Discussion Post
DDHA 8703 Week 6 questions, answered
Where can I find a free DDHA 8703 Week 6 sample paper?
The causal claim post and both replies, with references, are free to read here.
What claim does the DDHA 8703 Week 6 example challenge?
That a care management program cut emergency visits by 40 percent because visits fell after patients enrolled.
What are the three rival explanations?
Selection of patients at an unusually high point, a seasonal fall in visits and care managers' attention to the measure.
Does the post conclude the program failed?
No; it says the evidence cannot show the program's effect without a comparison group followed over the same months.
What rival explanation does a classmate add?
Two urgent care centers opened near enrolled patients during the same period, which could lower visits for everyone nearby.