Drawn paths between night shifts and preterm birth, not habit, produce the covariate set in this confounding brief example, written for Week 3 of PHLT 8520. Searches like "phlt 8520 week 3 assignment example", "phlt8520 week 3 sample" and "phlt 8520 week 3 example" land here.
What a finished PHLT 8520 Week 3 confounding brief looks like
Two pages and one diagram make up the brief. The diagram places night-shift work in the first twenty weeks on the left, preterm birth on the right, and four common causes above them: household income, maternal age, parity with any prior preterm birth, and job grade. Below the main arrow sit two variables marked off limits. Sleep disruption lies on the path from shifts to birth timing, and sick leave taken during pregnancy can be produced both by shifts and by early complications. Prose beside the drawing lists each back-door path in words and names the variable that blocks it. A short table then gives the resulting adjustment set, the two excluded variables and one sentence per exclusion. A last line names a common cause the available records cannot measure at all.
How a PHLT 8520 Week 3 example is structured
Drawing precedes naming throughout, because the brief's argument is that an adjustment set is a conclusion derived from structure, not a starting inventory. Greenland, Pearl and Robins are credited for bringing causal diagrams into epidemiologic reasoning, and Pearl's back-door criterion supplies the rule applied: block every path that enters the exposure through a shared cause, and open none. Each back-door path is written out node by node, letting a reader check the blocking claim without the picture. The two exclusions receive separate treatment, one as a mediator whose adjustment would remove part of the effect under study, one as a common effect whose adjustment would open a path through unmeasured complications. d-separation is named once, plainly, as the property the final set achieves. An unmeasured common cause, financial strain behind a second job, closes the brief and hands the problem forward.
Four arrows into both ends
Household income, maternal age, parity with prior preterm birth and job grade are drawn as causes of shift assignment and of birth timing.
Paths spelled out
Every back-door path is written node by node, with the variable that blocks it named beside it.
A mediator left alone
Sleep disruption lies between night work and early delivery, so conditioning on it would strip away part of the effect in question.
A common effect left alone
Sick leave in pregnancy is produced by shifts and by early complications, and conditioning on it would open a new path.
What the records never hold
Financial strain behind a second job is named as a shared cause no available file measures.
Where marks go in PHLT 8520 Week 3
Confounding briefs at this stage are marked on derivation: could a reader reproduce the adjustment set from the drawing alone? This one can, which settles the biggest block. Writing each back-door path in words adds its own share, since a diagram whose paths are never traced reads as illustration. The two exclusions are where strong briefs pull ahead. Refusing to adjust for sleep disruption proves the author has fixed which effect is being estimated, and refusing sick leave shows collider reasoning applied to a real variable instead of a textbook letter. Naming an unmeasured shared cause earns the closing share. Terminology is checked closely: d-separation appears once, correctly. Briefs lose ground by adjusting for everything recorded, by drawing arrows with no stated reason, and by treating a finished diagram as proof that confounding is gone.
Get a PHLT 8520 Week 3 example written to your instructions
Share the Week 3 prompt and rubric, with the exposure and outcome your section is working on and any variables already under consideration. The brief draws the structure, traces each shared-cause path in words, and derives an adjustment set that keeps mediators and common effects out. Your first brief is free; allow 24-48 hours.
PHLT 8520 Week 3 questions, answered
What is the back-door criterion?
A rule from Judea Pearl's work on causal diagrams for choosing an adjustment set. A set of variables satisfies it when none of them is a descendant of the exposure and together they block every path from exposure to outcome that begins with an arrow pointing into the exposure. The brief applies the rule to one composite diagram. A custom brief applies it to the structure your own question implies, variable by variable.
Why not adjust for sleep disruption if it predicts preterm birth?
Because in this diagram it sits between night work and birth timing, carrying part of whatever effect night work has. Adjusting for it would estimate only the portion of the effect running through other routes, which answers another question entirely. Mediation analysis separates those portions deliberately, but that is a later topic. For your question, a custom brief marks every mediator before building the adjustment set.
Does the brief claim night shifts cause preterm birth?
It claims nothing about the answer. The diagram expresses assumptions about what causes what, and those assumptions belong to the author, stated so they can be disputed. No data appear on the page, and the health system behind the question is invented. Pregnant workers with questions about their schedules belong with their own clinicians, which this page does not replace. A custom brief draws whatever structure your prompt requires.