| Course | DPH 890 Doctor of Public Health Capstone |
|---|---|
| Module | Module 4 |
| Paper type | Limitations section |
| Length | About 1,149 words, 7 pages |
| Format | APA 7 student paper |
| School | Aspen University |
| Program | Doctor of Public Health |
| Updated | September 2026 |
Free sample paper for DPH 890 Module 4
Honest Boundaries: Limitations and Future Work in a Community Health Worker Program Evaluation
Student Name
Doctor of Public Health Program, Aspen University
DPH 890: Doctor of Public Health Capstone
Instructor Name
Month Day, Year
Honest Boundaries: Limitations and Future Work in a Community Health Worker Program Evaluation
Every study has limitations, and readers need to know them to judge how far to trust the findings. Yet limitations are often mentioned briefly, vaguely or not at all. This paper writes the limitations and future work section for a capstone that found participation in a county community health worker program was associated with better blood pressure control, and it proposes studies to address what remains uncertain.
Why Limitations Are Underreported
Ioannidis (2007) examined published research and found that limitations were often not acknowledged, particularly in abstracts, and that authors tended to understate them. Underreporting leads readers to overestimate the certainty of findings and makes it harder to plan better studies.
A Structured Approach
Puhan et al. (2012) suggest describing each limitation in four parts: what the limitation is, how it may have affected results, what was done to address it and what alternative approaches could reduce it in future. This structure turns a list of weaknesses into useful information for readers and future researchers.
Summary of Limitations
Each main limitation appears in the table below.
| Limitation | Likely effect | Steps taken | Remaining uncertainty |
|---|---|---|---|
| Unmeasured confounding by motivation or support | Could inflate the estimated benefit | Propensity matching; E-value of 1.74 | A moderately strong confounder could explain part of the effect |
| Routine blood pressure measurement | Random error likely biases toward no difference | Standard protocol; sensitivity analysis by clinic | Some misclassification near 140/90 |
| Missing 12-month readings | Bias if missingness related to outcome | Multiple imputation; complete-case check | Small if data missing at random |
| Single county | Limits generalizability | Detailed description of setting | Transfer to other settings uncertain |
| Small qualitative sample | May miss some experiences | Interviewed participants who disengaged | Views of nonparticipants absent |
Unmeasured Confounding
Participants may have been more motivated or better supported than comparison adults in ways not captured in records, which could make the program appear more effective than it was. The E-value of 1.74 means an unmeasured confounder would need to increase both the likelihood of joining and of achieving control by about 74% each to explain away the effect entirely (VanderWeele & Ding, 2017). Such a confounder is possible but would have to be fairly strong and independent of the matched factors.
Measurement
Blood pressure came from routine clinic visits, where technique varies. Random measurement error would tend to make groups look more similar, biasing results toward no difference. Readings near the 140/90 threshold may have been misclassified. The continuous systolic outcome, less sensitive to threshold effects, supports the main finding.
Missing Data
About 10% of 12-month readings were missing. Multiple imputation assumed that missingness depended on observed characteristics. If residents with poor control were less likely to return, results could be biased; the similar complete-case estimate provides some reassurance.
Generalizability
The study took place in one county's east-side neighborhoods with a particular population, health system and program design. Results may differ in rural areas, among other populations or with different staffing. Detailed description of the setting helps readers judge whether findings apply to their context.
The Qualitative Sample
Fifteen interviews offered insight into mechanisms but did not include adults who declined to join. Their reasons might reveal barriers the program has not addressed. Future interviews should include them.
Future Work: A Stepped Wedge Evaluation
As the health department expands the program to new neighborhoods, it could introduce it in a randomized order over time, a stepped wedge design. Every neighborhood would eventually receive the program, but the staggered start would allow a stronger test of its effect than the capstone's matched comparison.
Future Work: Costs and Long-Term Outcomes
A cost analysis using program budgets and health care use would estimate cost per additional person reaching control and potential savings from fewer hospitalizations. Longer follow-up linking records to strokes, heart attacks and deaths would show whether blood pressure gains translate into fewer events.
Future Work: Components
Analyses of visit-level data and a factorial pilot testing, for example, home monitors with and without pharmacy support, could show which components matter most. This would help the county refine the program and help other agencies adopt its most effective parts.
Presenting Limitations Without Undermining the Study
Honest limitations strengthen credibility. The section states each limitation clearly, explains its likely direction and size and notes what was done, without apology or exaggeration. Readers are more likely to trust findings when authors show they have considered what could be wrong.
Selection Into the Program
Residents chose to enroll, and those who enrolled may differ from those who did not in ways that also affect blood pressure, such as readiness to change or family support. Matching addressed measured differences only. Interviews with participants suggested that many joined because a clinician referred them, which may reduce, though not eliminate, self-selection.
Contamination
Some comparison adults may have received similar support from clinic staff or other programs, which would narrow the difference between groups and make the program look less effective. Records did not capture all such support, so the estimate may understate the program's effect.
Limits of the E-Value
The E-value summarizes how strong confounding would need to be but does not say whether such a confounder exists. It should be interpreted alongside knowledge of the setting. Here, known strong predictors of control, such as baseline pressure and insurance, were matched, which makes an unmeasured confounder of that strength less likely.
Ethical Considerations for Future Studies
A stepped wedge design would delay the program in some neighborhoods, but only until their scheduled start, which would occur anyway given limited staff. Ethics review would consider whether the order could be based on need rather than chance, balancing fairness and rigor.
Limitations of the Literature Review
The literature review was a systematized review by a single student with partial second screening, limited to English. Relevant studies, particularly from Latin America, may have been missed. This affects how completely the capstone's findings can be compared with prior evidence.
Balancing Limitations and Contributions
The section closes by restating what the study contributes despite its limits: evidence from a health department-run program, using routine data, with results consistent across checks. Readers should leave with an accurate sense of both what the study shows and where caution is needed.
Limitations the Committee May Raise
Committee members may also ask about the choice of 140/90 rather than 130/80, the 12-month time frame and whether the program changed during the study. Each is addressed briefly: the threshold matched clinic practice, 12 months balanced timeliness and effect and program changes were minor and documented.
Conclusion
The capstone's main limitations are possible unmeasured confounding, routine measurement, missing data, a single setting and a small qualitative sample. Describing each with its likely effect, the steps taken and remaining uncertainty, and proposing a stepped wedge evaluation, cost and long-term studies and component testing, gives readers an honest boundary around the findings and a path for future work.
References
Ioannidis, J. P. A. (2007). Limitations are not properly acknowledged in the scientific literature. Journal of Clinical Epidemiology, 60(4), 324-329. https://doi.org/10.1016/j.jclinepi.2006.09.011
Puhan, M. A., Akl, E. A., Bryant, D., Xie, F., Apolone, G., & ter Riet, G. (2012). Discussing study limitations in reports of biomedical studies: The need for more transparency. Health and Quality of Life Outcomes, 10, Article 23. https://doi.org/10.1186/1477-7525-10-23
VanderWeele, T. J., & Ding, P. (2017). Sensitivity analysis in observational research: Introducing the E-value. Annals of Internal Medicine, 167(4), 268-274. https://doi.org/10.7326/M16-2607
What the DPH 890 Module 4 instructions ask for
Aspen's DPH 890 description includes summarizing and defending the project, and with the fourth module's wording restricted to people in the course, this example writes an honest limitations section. Limitations assignments usually ask you to identify weaknesses, explain their likely effects and propose future research. Describe each limitation in four parts: what it is, its likely effect, what you did and what could reduce it. Use a table. Quantify where possible, for example with an E-value. Consider generalizability. Propose specific future studies. Balance limitations with the study's contribution so readers leave with an accurate picture. Explain what the E-value can and cannot tell readers. Keep the tone factual, not apologetic.
How this DPH 890 Module 4 example is built
The section runs about a thousand words across twenty headings, from why limitations are underreported and a structured approach to a four-column table. Unmeasured confounding, measurement, missing data, generalizability and the qualitative sample are discussed in turn, followed by three kinds of future work and how to present limitations honestly. Selection into the program, contamination, limits of the E-value, ethics for future studies, the literature review's limits, balancing limitations with contributions and questions the committee may raise are added. A margin comment explains why evidence of underreporting opens the section. Each row of the limitations table is expanded in its own section with the likely direction and size of bias. Future work is specific and feasible. Ethics of the proposed design are considered.
Where the marks sit in the DPH 890 Module 4 rubric
Limitations sections are marked on honesty, specificity, reasoning about direction and size of bias and useful proposals for future work. This paper cites research showing limitations are poorly acknowledged, a proposal for structured reporting of limitations and the E-value method, in APA style. The table gives each limitation's effect and mitigation. Contamination is recognized as a bias toward no difference. A stepped wedge proposal turns program expansion into a stronger future test, which instructors value. The section also admits the literature review's own limits, a level of candor committees respect. Contamination and selection are both addressed. The tone is candid but not self-defeating.
DPH 890 Module 4 help: mistakes that cost marks
Students often list generic limitations such as small sample size without explaining their effect, or apologize so heavily that readers doubt the study. Name specific limitations. State the likely direction of bias. Describe what you did about each. Propose concrete future studies. Balance weaknesses with contributions. If you struggle to judge the direction of a bias, a tutor can help you reason it through. Close with what the study still shows despite its limits, and check that your abstract mentions the main limitation too. For each limitation, write one sentence on what a reader should do with that information. Avoid generic limitations. State the likely direction of each bias.
Write yours, or have the desk draft it
This paper is an original model document written by our desk, not a submitted student paper and not an official Aspen University document. Read it for the moves, then write your own to the instructions in your classroom. If you want one built to your exact prompt and rubric, the first custom sample is free and arrives in 24 to 48 hours.
More DPH 890 and Doctor of Public Health sample papers
- DPH 890 Module 1: Assembling the Complete Manuscript
- DPH 890 Module 2: Results Chapter With Tables and Figures
- DPH 890 Module 3: Discussion and Implications for Practice
- DPH 890 Module 5: An Executive Summary for Decision Makers
- DPH 890 Module 6: The Defense Presentation
- DPH 890 Module 7: A Dissemination Plan, Manuscript and Brief
- DPH 890 Module 8: Comprehensive Exam Preparation
- DPH 801 Module 4: Community Models and Diffusion
- DPH 840 Module 3: A Strategic Plan With Goals and Measures
- DPH 830 Module 8: Human Rights and Ethics in Global Health
- DPH 820 Module 2: The Right to Health as a Framework
DPH 890 Module 4 questions, answered
What does DPH 890 Module 4 usually ask for?
Aspen's DPH 890 covers summarizing the capstone, so a limitations and future work section is a typical assignment. Confirm with your classroom prompt.
How should each limitation be described?
State the limitation, its likely effect on results, what was done to address it and what could reduce it in future studies.
What is a stepped wedge design?
A design in which an intervention is introduced to groups in a randomized order over time until all receive it.
Where can I find a free DPH 890 Module 4 sample paper?
This page presents the limitations and future work section with a table of limitations, effects, steps taken and remaining uncertainty.
How do you write limitations in DPH 890 Module 4?
Describe each limitation, its likely effect on results, what was done to address it and what could reduce it in future studies.