PU570 Unit 7 Surveillance Data Analysis Example

Course	PU570 Chronic Disease Epidemiology
Unit	Unit 7
Paper type	Surveillance data analysis
Length	About 910 words, 6 pages double-spaced
Format	APA 7 student paper
School	Purdue University Global
Program	Master of Public Health (MPH)
Updated	October 2026

Free sample paper for PU570 Unit 7

Incidence Tripled, Mortality Did Not Move: A Unit 7 Surveillance Data Analysis Asking How Much of a Composite Registry's Thyroid Cancer Rise Was More Disease, and Finding the Answer in Tumor Size and Stage

[Student Name]

Purdue University Global

PU570: Chronic Disease Epidemiology

Unit 7 Surveillance Data Analysis

[Instructor Name]

[Date]

The registry series and its figures are composite but follow published national patterns; the diagnostic history and the research cited are real.

What this part is doingThe analysis treats the flat mortality curve as the central evidence rather than as a reassuring footnote, which is what distinguishes it from a report of rising incidence.

The Two Curves

1. Age-adjusted incidence: 4.9 per 100,000 in 1980, 14.6 per 100,000 in 2009, a near tripling.

2. Age-adjusted mortality: 0.50 per 100,000 in 1980, 0.51 in 2009, with no trend in between.

A genuine increase in a disease that kills people should move the second curve, with a lag. Thirty years is longer than any plausible lag for a cancer with this survival profile.

Reading One: A Real Increase in Disease

What it would predict. Rising incidence across all tumor sizes, including large tumors. Rising incidence of advanced stage at diagnosis. Rising mortality after a lag. Possibly a named exposure with a matching time trend, since radiation exposure is the established cause. What the data show. Incidence of tumors over 4 centimeters rose from 0.6 to 0.9 per 100,000, a real but small increase. Distant-stage disease at diagnosis stayed near 0.4 per 100,000 throughout. Mortality did not move. One prediction of this reading is partly met and the other three fail.

Reading Two: More Looking

What it would predict. The rise concentrated in small tumors, in early stage, following the availability of imaging and biopsy. No change in late-stage disease. No change in mortality. What the data show. Papillary carcinomas under 2 centimeters account for 81 percent of the increase in this series. Localized stage accounts for 88 percent of it. Neck ultrasound volume in the registry's catchment rose roughly sixfold over the period, and fine-needle aspiration rates rose with it. Three quarters of the extra cases are small papillary tumors found in the neck of somebody who came in for something else.

Why the Comparison of the Two Readings Is the Analysis

Neither reading is tested by the incidence curve alone, which is why a surveillance report that stops at the trend line has done no analysis. The published national series makes the same comparison and reaches the same conclusion: the rise is overwhelmingly in small papillary tumors with stable mortality, which is the signature of detection rather than of occurrence (Davies & Welch, 2014). Where screening has been widespread, the pattern is sharper still: a national program of neck ultrasound produced a fifteenfold rise in incidence with no change in thyroid cancer deaths (Ahn et al., 2014). The international comparison confirms that the pattern is not confined to one health system (Vaccarella et al., 2016).

What this part is doingEach reading is given its own predictions before the data are consulted, so the conclusion follows from a test rather than from the author's starting view.

What the Small Real Increase Means

The analysis does not claim the entire rise is detection. The increase in tumors over 4 centimeters, 0.3 per 100,000, is small and probably real, and a reading that dismisses it to tell a cleaner story is as wrong as one that treats the whole rise as disease. Candidate explanations for a modest true increase include obesity prevalence and, in some populations, radiation exposure history. The honest summary is that roughly nine tenths of the rise is attributable to detection and that a small real increase sits underneath it.

Four Things This Analysis Changes for the Registry's Reporting

1. Report incidence by tumor size band, not only in total. A single curve invites the wrong reading.

2. Report incidence and mortality on the same figure, with the same age standard, since their divergence is the finding.

3. Stop describing the trend as an increase in thyroid cancer in press material, and describe it as an increase in diagnosed thyroid cancer.

4. Add imaging volume to the registry's annual report as context, since a diagnostic trend cannot be interpreted without it.

What this part is doingThe recommendation to print incidence and mortality on one figure is the smallest change in the list and the one most likely to stop the trend being misread.

How the Age Pattern Supports the Reading

One further check belongs in the analysis because it is cheap and discriminating. If the rise were occurrence, it should be concentrated in the age groups whose exposure changed. If it is detection, it should follow who gets imaged, which means adults in their forties and fifties with employer coverage and a reason to visit a clinic. In the composite series, incidence among women aged 35 to 64 rose from 7.1 to 24.8 per 100,000, while incidence among adults over 75, who are imaged less and die of other things, rose from 6.2 to 9.4. Mortality in neither band moved. An exposure acting on the thyroid would have no reason to spare the oldest group, and a diagnostic cascade has every reason to. The age pattern therefore agrees with the size and stage pattern, and three independent checks pointing the same way is a stronger argument than any one of them.

What Surveillance Data Cannot Settle Here

Surveillance shows the pattern and cannot demonstrate the mechanism in any individual case, so the analysis cannot say which of these patients had a tumor that would never have surfaced. It cannot quantify harm from treatment, which is where the consequence of overdiagnosis lies and which requires treatment and outcome data the registry does not hold. And the composite series ends in 2009, while national practice has since shifted toward observation for small tumors, so a current analysis would need the later years before advising anybody on what to do next.

References

Ahn, H. S., Kim, H. J., & Welch, H. G. (2014). Korea's thyroid-cancer epidemic: Screening and overdiagnosis. The New England Journal of Medicine, 371(19), 1765-1767. https://doi.org/10.1056/NEJMp1409841

Davies, L., & Welch, H. G. (2014). Current thyroid cancer trends in the United States. JAMA Otolaryngology: Head and Neck Surgery, 140(4), 317-322. https://doi.org/10.1001/jamaoto.2014.1

Vaccarella, S., Franceschi, S., Bray, F., Wild, C. P., Plummer, M., & Dal Maso, L. (2016). Worldwide thyroid-cancer epidemic? The increasing impact of overdiagnosis. The New England Journal of Medicine, 375(7), 614-617. https://doi.org/10.1056/NEJMp1604412

Reading the PU 570 Unit 7 instructions

Unit 7 usually gives a surveillance trend and asks what it means. Put the incidence and mortality curves side by side first, because their divergence is the evidence. Then write out two competing readings, a real increase and increased detection, each with the predictions it makes about size, stage, age and mortality, before consulting the data. Test both against the figures and report which predictions fail. Resist a clean story: if part of the rise looks real, say how much and name candidate causes. Add an independent check, such as the age pattern or imaging volume. Finish with concrete changes to how the registry reports the trend, and with what surveillance cannot settle.

Inside the PU570 Unit 7 example

The analysis on this page gives the two curves, then sets out each reading with its predictions and tests them. A highlighted line states that three quarters of the extra cases are small papillary tumors found incidentally. A margin note explains why flat mortality is treated as the central evidence rather than as reassurance. Published national and international comparisons follow, including a screening program that produced a fifteenfold rise with no change in deaths. The small real increase is conceded with candidate causes. An age-pattern check appears, with a note, and the analysis closes with four reporting changes and three questions surveillance cannot answer. A fourth reporting change is proposed in the same section, naming imaging volume as standing context for any diagnostic trend.

PU570 Unit 7 rubric: what earns full points

Scoring turns on whether the trend was tested. The heaviest credit goes to two readings written with predictions in advance and then checked against size, stage, age and mortality. Marks go to treating flat mortality as evidence rather than as a footnote. Conceding a small real increase, with candidate explanations, scores better than a single clean conclusion. An independent check, such as the age gradient or imaging volume, is rewarded. Reporting recommendations are weighted, especially putting incidence and mortality on one figure and renaming the trend as diagnosed disease. Points also go to naming what surveillance cannot do, including quantifying treatment harm. APA 7 citation of the published trend literature is required.

PU570 Unit 7 help: mistakes that cost points

The common failure is a report that describes the rise and stops, which performs no analysis. Write the competing readings. The second error is ignoring mortality, which is the single most informative series on the page. Third, do not claim the whole increase is detection, because a modest real rise in large tumors is visible and dismissing it is as wrong as inflating it. Avoid describing the trend in press material as an increase in thyroid cancer rather than in diagnosed thyroid cancer. And remember that surveillance cannot tell you which individual tumor would never have surfaced. An analysis of your own registry trend takes two days.

Get a PU570 Unit 7 example written to your instructions

Name the condition and the surveillance system the PU570 Unit 7 analysis uses, whether SEER, BRFSS or a state registry. Include the data extract or report, with the prompt and rubric. Within 24-48 hours a free first custom sample tests that series for artifacts before interpreting any trend, then says which part of the change it would defend. The paper above is an original model document written by our desk, not a submitted student paper and not an official Purdue University Global document.

More PU570 papers and MPH samples

PU570 Unit 7 questions, answered

Where can I find a free PU570 Unit 7 sample paper?

The whole surveillance analysis, both readings and all the checks, is published on this page.

What does flat mortality with rising incidence mean?

That most of the additional cases are not the kind that kills, which is the signature of detection rather than of occurrence.

How do you test for overdiagnosis?

Check where the rise sits by tumor size, stage and age, and compare it against imaging volume. A real increase would not spare the oldest adults.

Is any of the rise real?

A small part. Tumors over 4 centimeters rose from 0.6 to 0.9 per 100,000, which is modest and probably genuine.

How should a registry report this?

Incidence by size band, incidence and mortality on one figure, imaging volume as context, and the phrase diagnosed thyroid cancer.