| Course | HCD 501 Population Health Data Management and Analysis |
|---|---|
| Module | Module 1 |
| Paper type | Comparative research methods paper |
| Length | About 851 words, 6 pages |
| Format | APA 7 student paper |
| School | Arizona State University |
| Program | MS in the Science of Health Care Delivery |
| Updated | October 2026 |
Free sample paper for HCD 501 Module 1
Same Hormones, Opposite Answers: Data and Design in the Nurses' Health Study and the Women's Health Initiative
Student Name
MS in the Science of Health Care Delivery, Arizona State University
HCD 501: Population Health Data Management and Analysis
Instructor Name
Month Day, Year
Same Hormones, Opposite Answers: Data and Design in the Nurses' Health Study and the Women's Health Initiative
Introduction
Through the 1980s and 1990s, cohort after cohort reported that women taking hormones after menopause had fewer heart attacks. In 2002, a large randomized trial reported the opposite. This project compares a major analysis from the Nurses' Health Study (Grodstein et al., 2000) with the principal results of the estrogen plus progestin trial of the Women's Health Initiative (Writing Group for the Women's Health Initiative Investigators, 2002). It examines how each study collected and managed its data, which statistical methods it used and how its research approach shaped its conclusion, and it closes with what the disagreement teaches about using data for population health decisions.
Study 1: The Nurses' Health Study Analysis
Research approach. The Nurses' Health Study is a prospective cohort of female registered nurses that began in 1976. The analysis followed 70,533 postmenopausal women through 1996 and compared the cardiovascular outcomes of women who chose to use hormone therapy with those who did not (Grodstein et al., 2000).
Data management. Exposure data came from mailed questionnaires every two years, which recorded current hormone use, type, dose and duration and updated other risk factors such as smoking, blood pressure and body weight. Reported coronary events and strokes were confirmed through medical record review, which improves outcome accuracy. Exposure was self-reported and could change between questionnaires.
Biostatistics. Regression models estimated relative risks adjusted for cardiovascular risk factors. Current users had a relative risk of major coronary events of 0.61 compared with never-users, a 39% lower risk, while stroke risk was raised among women taking higher estrogen doses or estrogen plus progestin.
Study 2: The Women's Health Initiative Trial
Research approach. The estrogen plus progestin trial randomly assigned 16,608 postmenopausal women aged 50 to 79 who had a uterus to daily conjugated equine estrogens plus medroxyprogesterone acetate or to placebo at 40 U.S. clinical centers. The trial was stopped early, after a mean of 5.2 years, when breast cancer risk crossed a prespecified boundary and the overall balance of risks exceeded benefits (Writing Group for the Women's Health Initiative Investigators, 2002).
Data management. Outcomes were collected at regular contacts and adjudicated centrally by physicians who did not know the treatment assignment, and an independent data and safety monitoring board reviewed accumulating data against stopping rules defined in advance.
Biostatistics. Proportional hazards models estimated hazard ratios by intention to treat. Women assigned to hormones had 29% more coronary events than those on placebo (hazard ratio 1.29), with increases in breast cancer, stroke and pulmonary embolism and reductions in colorectal cancer and hip fracture.
Side-by-Side Comparison
| Feature | Nurses' Health Study analysis | Women's Health Initiative trial |
|---|---|---|
| Design | Prospective cohort | Randomized, placebo-controlled trial |
| Who chose treatment | Women and their physicians | Random assignment |
| Exposure data | Self-reported every two years | Assigned pill, adherence tracked |
| Outcome data | Self-report confirmed by records | Central blinded adjudication |
| Measure of effect | Relative risk, adjusted | Hazard ratio, intention to treat |
| Coronary result | 0.61 for current users | 1.29 for assigned therapy |
Why the Studies Disagreed
Confounding. Women who chose hormone therapy were not like other women in several respects that independently lower heart risk, such as income, education, health care use and healthier habits. Adjustment can only remove confounding from variables that were measured well. Lawlor et al. (2004) argued that confounding by socioeconomic position and related factors probably explains much of the apparent benefit in observational studies.
Timing and follow-up. The cohort mostly captured women who began therapy near menopause and had used it for years, while the trial enrolled older women, many years past menopause, and measured risk from the first dose. Early harm can be missed when a cohort counts mainly long-term current users.
Different questions. The cohort compared current users with never-users; the trial compared starting therapy with not starting it, regardless of later adherence.
Reconciling the Results
Hernán et al. (2008) reanalyzed Nurses' Health Study data to emulate the trial: they defined eligible women at repeated time points, compared women who started estrogen plus progestin with those who did not and analyzed by intention to treat. Coronary risk among starters came out at 1.42 times that of non-starters during the first two years and 0.96 across all follow-up, close to the trial's findings. They concluded that differences in time since menopause and length of follow-up could largely explain the original discrepancy. The data were not the problem; the analytic approach was.
Lessons for Population Health Data Analysis
First, design determines which questions data can answer. Second, careful data management, such as validated outcomes, does not protect against confounding in exposure. Third, defining the comparison explicitly, starting therapy versus not, helps observational analyses approach trial results. Fourth, health systems that use their own data to evaluate programs face the same risks and should design analyses as if they were trials whenever possible.
Conclusion
The two studies reached opposite conclusions about hormone therapy and coronary disease not because one dataset was wrong but because their research approaches asked different questions and handled confounding differently. Analyzed like a trial, the cohort agreed with the trial.
References
Grodstein, F., Manson, J. E., Colditz, G. A., Willett, W. C., Speizer, F. E., & Stampfer, M. J. (2000). A prospective, observational study of postmenopausal hormone therapy and primary prevention of cardiovascular disease. Annals of Internal Medicine, 133(12), 933-941. https://doi.org/10.7326/0003-4819-133-12-200012190-00008
Hernán, M. A., Alonso, A., Logan, R., Grodstein, F., Michels, K. B., Willett, W. C., Manson, J. E., & Robins, J. M. (2008). Observational studies analyzed like randomized experiments: An application to postmenopausal hormone therapy and coronary heart disease. Epidemiology, 19(6), 766-779. https://doi.org/10.1097/EDE.0b013e3181875e61
Lawlor, D. A., Davey Smith, G., & Ebrahim, S. (2004). Commentary: The hormone replacement-coronary heart disease conundrum: Is this the death of observational epidemiology? International Journal of Epidemiology, 33(3), 464-467. https://doi.org/10.1093/ije/dyh124
Writing Group for the Women's Health Initiative Investigators. (2002). Risks and benefits of estrogen plus progestin in healthy postmenopausal women: Principal results from the Women's Health Initiative randomized controlled trial. JAMA, 288(3), 321-333. https://doi.org/10.1001/jama.288.3.321
Reading the HCD 501 Module 1 assignment instructions
HCD 501's Collective Research Approach Project is due in Module 6 and carries 80 points, 16% of the final grade, matching the midterm. The syllabus states its purpose plainly: to synthesize the biostatistics, data management and research approaches used in two studies on very similar topics that came to very different conclusions. Synthesis means more than describing each study; the project should explain why the conclusions differ. By Module 6 the course has covered descriptive statistics, Excel analysis, t-tests, ANOVA and chi-square and the strengths and limits of randomized trials, so draw on those modules. Canvas holds the full instructions and grading criteria. Read the original articles themselves; secondhand summaries leave out the design and analysis details the project turns on.
How the HCD 501 Module 1 example is put together
The sample introduces the two studies and the puzzle they pose, then describes each under the same three headings named in the syllabus: research approach, data management and biostatistics. A comparison table sets their features side by side. The analysis explains the disagreement through confounding, timing and the different questions the studies asked, citing a commentary on confounding. A reconciliation section reports how cohort data, reanalyzed to mirror the trial's design, came close to the trial's answer, and the paper closes with four lessons for population health analysis. Four sources, including both original studies, support the project. The paper also explains effect measures in plain terms, for example that a relative risk of 0.61 means a 39% lower risk, which shows the reader understands the numbers.
Reading the HCD 501 Module 1 grading rubric
The project earns up to 80 points under its Canvas rubric. Projects of this kind are typically credited for choosing two studies that genuinely conflict, accurate description of each study's design, data collection and statistical methods, a structured comparison, a convincing explanation of why the results differ, correct interpretation of effect measures such as relative risks and hazard ratios and lessons that connect to population health practice. Projects lose credit when they summarize the studies separately without synthesis, when they declare one study simply wrong and when they misread effect sizes. Clear tables and consistent headings make the comparison easy to grade, and accurate citation of the original studies, not only secondary summaries, shows careful work.
HCD 501 Module 1 help with common mistakes
Pick studies whose disagreement is documented, so you can find commentary that explains it. Read the methods sections closely for how exposure and outcomes were measured, since data management differences often matter. Report effect sizes with their meaning, not just their numbers. Use the same headings for both studies so your comparison is clean. If your two studies are in a different field, the desk can help you find reanalyses or commentaries that explain the disagreement. Before writing, build a two-column table of each study's design, data sources, measures, statistical methods and results; most of the synthesis will come from the rows where the columns differ. Then ask which of those differences could change the answer.
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 Arizona State 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 HCD 501 and MS in the Science of Health Care Delivery sample papers
- HCD 575 Module 2: Final Project Paper: A Transformation and Leadership Challenge
- HCD 511 Module 1: Final Summative Case Study: Applying Health Economics to a Health Care Issue
- HCD 602 Module 1: Article Analysis
HCD 501 Module 1 questions, answered
Where can I find a free HCD 501 Module 1 sample paper?
This page has a full HCD 501 Module 1 sample: a project comparing the Nurses' Health Study and the Women's Health Initiative on hormone therapy and heart disease.
What is the HCD 501 Collective Research Approach Project?
An 80-point project synthesizing the biostatistics, data management and research approaches of two studies on similar topics with very different conclusions.
Why did observational studies and the WHI disagree on hormone therapy?
Mainly confounding, differences in timing since menopause and follow-up, and different comparisons, current users versus never-users or starting versus not starting.
What is target trial emulation?
Analyzing observational data as if it came from a randomized trial, by defining eligibility, treatment start and intention-to-treat comparisons explicitly.
When is the HCD 501 project due?
In Module 6 of the seven-module course, after the modules on statistical tests and randomized trials.