| Course | PHLT 8520 Advanced Epidemiological Methods |
|---|---|
| Week | Week 4 |
| Paper type | Design proposal paper |
| Length | About 1,090 words, 6 pages |
| Format | APA 7 student paper |
| School | Walden University |
| Program | PhD in Public Health |
| Updated | October 2026 |
Free sample paper for PHLT 8520 Week 4
Day Workers in the Same Jobs as the Comparison: A Design Proposal for a Record-Linked Cohort of Employee Pregnancies, Night Rosters and Preterm Birth
Student Name
Walden University
PHLT 8520: Advanced Epidemiological Methods
Instructor Name
Month Day, Year
Day Workers in the Same Jobs as the Comparison: A Design Proposal for a Record-Linked Cohort of Employee Pregnancies, Night Rosters and Preterm Birth
Aim
The study will estimate whether working one or more night shifts during gestational weeks 4 to 20 changes the risk of preterm birth among employees of a composite nine-hospital health system. The Week 2 comparison chose a record-linked retrospective cohort. This proposal specifies its population, exposure, comparison group, outcome, covariates and analysis, and it explains why the comparison group was chosen over two alternatives.
Population and Records
Eligible pregnancies are singleton pregnancies confirmed in the system's own obstetric clinics between January 2016 and December 2023 among women employed in rostered jobs throughout gestational weeks 4 to 20. Rostered jobs are those whose schedules are produced by the system's timekeeping software and can include nights: staff nurses, nursing assistants, respiratory therapists, laboratory technologists, pharmacy technicians, environmental services and security staff. Three sources are linked by employee number. Timekeeping supplies every clocked shift with its start and end times. Human resources files supply job title, grade, hire date and salary band. The obstetric record supplies the confirmation date, the dating ultrasound, prior pregnancies and the delivery.
In the composite data, 6,795 pregnancies meet these criteria. Of these, 3,105 include at least one night shift in the window and 3,690 include only day or evening shifts.
Exposure
A shift counts as a night shift when at least three of its hours fall in the late-night block that Danish payroll research used for the same question (Specht et al., 2019), from eleven at night to six in the morning. Exposure is defined as a strategy over the whole window: any night shift in weeks 4 to 20 versus none. Two secondary measures follow the dose of exposure, the count of night shifts in the window and the longest run of consecutive nights. Gestational weeks come from the dating ultrasound, so the window is anchored to the pregnancy rather than to the calendar.
The Comparison Group
Chosen: Day-Only Workers in the Same Job Titles
The comparison group is women in the same rostered job titles who worked no night shift during the window. A day-only staff nurse and a night-rotating staff nurse share training, physical demands, workplace exposures and the same employer's benefits and prenatal care. What separates them, ideally, is the timing of their hours. Restricting the comparison to the same titles removes the largest source of difference between night workers and everyone else: the kind of work they do.
Rejected: The General Population
Comparing employee night workers with all women delivering in the region would mix two contrasts, working versus not working and nights versus days. Employed women are, on average, healthier than the population that includes women unable to work, which is the healthy worker effect described by Li and Sung (1999). That difference would make night workers look protected even if nights caused harm.
Rejected: Evening Workers
Evening shifts, ending before midnight, were considered as a comparison because evening and night workers often hold the same roster patterns. They were rejected because evening work also shifts sleep and light exposure, so evening workers carry part of the exposure under study. Comparing nights with evenings would estimate only the added effect of the latest hours. Evening-only workers are kept in the day-only group for the main analysis and examined separately as a sensitivity check.
Outcome
Preterm birth is a live birth between 23 and 36 completed weeks, with gestational age from the first-trimester ultrasound where available and from the earliest ultrasound otherwise. Pregnancies ending before 23 weeks, by loss or termination, are counted and described but cannot contribute to the main outcome, a limitation addressed in next week's selection analysis. Spontaneous and medically indicated preterm births will be reported separately, since Goldenberg et al. (2008) describe their causes as partly distinct, and night work might plausibly act on one more than the other.
Covariates
Covariates are the four read from the Week 3 diagram, covering job, pay, age and illness before conception, while parity and prior preterm birth wait for the Week 8 decision. All are measured before week 4. Variables that occur after exposure begins, such as early pregnancy complications, are deliberately excluded from adjustment.
Analysis Plan
Robust standard errors will account for women contributing more than one pregnancy during the study period. Because the outcome affects about one pregnancy in ten, risk ratios rather than odds ratios will be reported, keeping the estimate on the scale readers can interpret.
Table 1
Planned Analyses
| Analysis | Contrast | Method | |
|---|---|---|
| Primary | Any night shift in weeks 4-20 vs none | Log-binomial regression for the risk ratio, adjusted for the Week 3 set | |
| Dose | Count of night shifts, in four groups | Same model with indicator terms | |
| Pattern | Longest run of consecutive nights | Same model | |
| Subtype | Spontaneous and indicated preterm birth | Separate models | |
| Sensitivity | Evening-only workers moved to a separate group | Same model | |
Note. Analyses planned for the composite cohort; numbers are not yet estimated.
Power, Briefly
With about 3,100 exposed and 3,700 unexposed pregnancies and a baseline preterm risk near 9%, the study can detect a risk ratio of 1.2 with about 80% power, using a conventional two-sided 5% significance level. Palmer et al. (2013) found that larger, higher-quality studies point to modest or null effects of shift work on preterm delivery, so the study is sized to detect a modest effect, not a large one.
Data Governance
Linking timekeeping to obstetric records joins two kinds of information employees never expected to see side by side, so the linkage will run inside the health system's research data warehouse, and the analytic file will carry study numbers rather than employee numbers. Managers will have no access to the linked file, since a schedule matched to a pregnancy outcome could be misused in staffing decisions. The protocol will seek review board approval, including a waiver of consent justified by the retrospective design and the impracticality of contacting former employees, and with a plan to report results only in groups of at least eleven pregnancies. Those safeguards also shape what the study can say: small job titles may need to be combined before results are published.
Limitations Carried Forward
The proposal leaves three questions for later weeks. Women who were never confirmed in the system's clinics cannot enter, and those who leave night rosters early in pregnancy may do so for reasons tied to risk. Timekeeping may miss shift swaps. Income outside the employee's salary is not recorded. Each will be examined in turn.
References
Goldenberg, R. L., Culhane, J. F., Iams, J. D., & Romero, R. (2008). Epidemiology and causes of preterm birth. The Lancet, 371(9606), 75-84. https://doi.org/10.1016/S0140-6736(08)60074-4
Li, C.-Y., & Sung, F.-C. (1999). A review of the healthy worker effect in occupational epidemiology. Occupational Medicine, 49(4), 225-229. https://doi.org/10.1093/occmed/49.4.225
Palmer, K. T., Bonzini, M., Harris, E. C., Linaker, C., & Bonde, J. P. (2013). Work activities and risk of prematurity, low birth weight and pre-eclampsia: An updated review with meta-analysis. Occupational and Environmental Medicine, 70(4), 213-222. https://doi.org/10.1136/oemed-2012-101032
Specht, I. O., Hammer, P. E. C., Flachs, E. M., Begtrup, L. M., Larsen, A. D., Hougaard, K. S., Hansen, J., Hansen, Å. M., Kolstad, H. A., Rugulies, R., Garde, A. H., & Bonde, J. P. (2019). Night work during pregnancy and preterm birth: A large register-based cohort study. PLOS ONE, 14(4), Article e0215748. https://doi.org/10.1371/journal.pone.0215748
What the PHLT 8520 Week 4 instructions ask for
Design proposals in doctoral epidemiology courses ask students to spell out their study so fully that a stranger could run it from the page. Expect required sections on population, data sources, exposure definition, comparison group, outcome definition, covariates and analysis, often with a power or sample size estimate. Many prompts single out the comparison group for justification because it shapes every estimate. Instructors look for operational definitions with numbers attached, such as hours and dates, rather than general descriptions. Lengths vary by section, and a table of planned analyses is often requested. Ethical and data governance issues may be required, especially when records are linked. Students are often asked to list limitations that later assignments will examine, so the proposal reads as the start of a sequence.
How the PHLT 8520 Week 4 example is put together
An aim paragraph frames the proposal as the next step after last week's design choice. A population section names the eligible job titles, the study years and the three linked record systems, then gives the composite counts of exposed and unexposed pregnancies. The exposure section borrows a payroll definition of a night shift and anchors the window to ultrasound dating. Three subsections defend the comparison group and reject two alternatives. The outcome section separates spontaneous from indicated preterm birth. Covariates follow the Week 3 diagram. Table 1 lists five planned analyses with their methods. Power, data governance and limitations carried forward make up the final sections, and margin notes flag the comparison argument and the aim.
PHLT 8520 Week 4 rubric: what earns full marks
Points in a design proposal go first to operational precision, and the sample earns them with definitions that specify hours, gestational weeks and record sources. The comparison group defense carries the next block: naming what the chosen group holds constant and what each rejected group would mix in shows design reasoning rather than preference. A planned analysis table that matches the question, including risk ratios rather than odds ratios for a common outcome, earns analytic credit. A power estimate tied to realistic effect sizes from published reviews shows feasibility thinking. Data governance and a limitations list earn smaller shares. Formatting, tables and references take the rest. Proposals that leave exposure as "night shift work" without hours or dates typically lose most of the definition credit.
PHLT 8520 Week 4 help: mistakes that cost marks
Vague definitions are the biggest risk in this proposal. Write exposure with clock hours and gestational weeks, and say which record supplies each. Defend your comparison group by naming what it holds constant, then explain why each alternative was rejected. Avoid comparing workers with the general population unless you can address the healthy worker effect. Choose an effect measure suited to how common your outcome is. Size the study for a realistic effect, not a hoped-for one, and cite where your expected effect comes from. Include how records will be protected if you link sources. List what you will examine later so the reader sees the plan. Our team can draft a first proposal around your own data sources at no charge.
Get a PHLT 8520 Week 4 example written to your instructions
Attach your Week 4 prompt and rubric, plus the question under study and any setting or data source your section named. The proposal specifies population, exposure and outcome, argues the comparison group against its rivals, and lists what it still lets through. A first proposal is free of charge and arrives in 24-48 hours.
More PHLT 8520 papers and PhD in Public Health samples
- PHLT 8520 Week 1: Error Taxonomy Brief
- PHLT 8520 Week 2: Design Selection Paper
- PHLT 8520 Week 3: Confounding Brief
- PHLT 8520 Week 5: Selection Analysis
- PHLT 8520 Week 6: Peer Challenge Reply
- PHLT 8520 Week 7: Measurement Error Brief
- PHLT 8520 Week 8: Control Strategy Comparison
- PHLT 8520 Week 9: Sensitivity Argument
- PHLT 8520 Week 10: Validation Check
- PHLT 8520 Week 11: Final Methods Paper
- PHLT 8500 Week 4: Coefficient Reading
- PHLT 8450 Week 4: Participation Plan
- PHLT 8331 Week 4: Paired t-Test Write-Up
- PHLT 8270 Week 4: Surveillance System Evaluation
PHLT 8520 Week 4 questions, answered
Where can I find a free PHLT 8520 Week 4 sample paper?
The full Week 4 design proposal sits on this page, defining a record-linked cohort of employee pregnancies and defending a same-job comparison group, with tables and checkable sources.
Why compare night workers with day workers in the same jobs?
Same-job day workers share training, physical demands, workplace exposures and benefits, so the comparison isolates the timing of work hours rather than differences in the kind of work.
Why not compare employees with the general population?
Employed women are healthier on average than the whole population, which includes women unable to work. That healthy worker effect would make night workers look protected even if nights caused harm.
Should I report odds ratios or risk ratios for preterm birth?
Because preterm birth affects roughly one pregnancy in ten, odds ratios overstate risk ratios. The sample plans log-binomial models so results are reported as risk ratios.
How is a night shift defined in this proposal?
As any shift with at least three working hours between 11 p.m. and 6 a.m., the definition used in Danish payroll research on night work and pregnancy.