PUBH 6036 Week 2 Study Critique Example

What a Single Caregiver Survey Can and Cannot Settle: A Critique of a Composite Study Claiming That Building-Wide Bed Bug Control Cut Children's Missed School Days

Student Name

Walden University

PUBH 6036: Public Health Research Methods

Instructor Name

Month Day, Year

What a Single Caregiver Survey Can and Cannot Settle: A Critique of a Composite Study Claiming That Building-Wide Bed Bug Control Cut Children's Missed School Days

The Article and Its Claim

The article critiqued here is a composite, built in the shape of a report that appears often in housing and pest management newsletters. Its title states that "Building-wide bed bug control reduced children's missed school days in public housing." The authors describe a housing agency that hired a contractor to inspect and treat every apartment in four buildings over one spring. In the following fall, the authors mailed a survey to caregivers in those buildings. Of 412 households with school-age children, 138 returned the survey (33.5%). Seventy-one percent of respondents said their child missed "fewer days" of school after the program than before it, and the authors conclude that the program improved attendance.

The question this critique asks is simple: what question could this design actually answer, and how far is that from the question its title says it settled? The design is a single survey, given once, after the program, to the people who chose to answer it. Grimes and Schulz (2002a) describe such studies as useful for describing a group and generating hypotheses, but unable to test whether an exposure changed an outcome, because they lack a comparison.

What the Design Can Answer

A one-time survey after a program can answer a descriptive question: how did caregivers who responded perceive their child's attendance after the program, compared with their memory of attendance before it? That is a real question and worth knowing. Caregivers' perceptions shape whether they report pests, allow inspectors inside and stay in the building. The 71% figure is a fair estimate of satisfaction among respondents, within the limits of a 33.5% response rate.

What the design cannot answer is whether attendance changed, or whether the program caused any change. Three missing pieces explain the gap.

Missing Piece 1: A Measure of Absence

The outcome is a caregiver's memory, not an attendance record. Caregivers were asked in the fall to compare absences across two school years. Memory for counts of ordinary events over many months is poor, and it bends toward what a person expects. A caregiver who has just lived through a treatment program, with its notices, bagged clothes and returned furniture, is likely to expect improvement. School districts keep daily attendance, and with written permission those records could have replaced recall entirely. The authors do not say whether they tried.

Missing Piece 2: A Comparison

There is no group that did not receive the program and no measurement before it. Any change in absence between the two school years could reflect many things that move attendance at once: a mild or severe flu season, a new school bus route, a change in district truancy policy, children growing a year older. Grimes and Schulz (2002b) note that a cohort design gains its strength by following people with and without an exposure forward from a starting point. This study has neither the second group nor the starting point.

Missing Piece 3: An Exposure Measure

The article does not report which apartments actually had bed bugs before treatment, or which were cleared. The gap is larger than it looks. In a careful trial in one low-income high-rise building, Wang et al. (2009) eradicated bed bugs from only half of the treated apartments after 10 weeks, even with steam, mattress encasements and repeated visits. A program that treats every apartment does not mean every child stopped living with bed bugs, and many children in treated buildings may never have lived with them at all. Without knowing who was exposed and who was cleared, even a true drop in absence could not be tied to the insects.

Who Answered

Two thirds of eligible households did not return the survey. The authors do not compare respondents with nonrespondents, so the reader cannot judge the direction of bias. Several directions are plausible. Families whose infestations were cleared may have been more willing to answer a survey about the program. Families who moved out during treatment, possibly those with the worst infestations, may never have received it. Families who feared that admitting bed bugs could affect their lease may have skipped it. Each would push the 71% in a different direction, and the article gives no way to tell which dominated.

The Title Against the Data

The title makes three claims: that the program reduced absences, that the reduction belonged to children and that it happened in public housing generally. The data support a narrower statement: most caregivers who answered believed their child missed fewer days. Wording carries weight here. "Reduced" is a causal verb, and a single post-program survey cannot carry it. A fair title would read something like "Caregivers' views of school attendance after a building-wide bed bug program: a survey of four public housing buildings."

What a Stronger Study Would Need

A study able to approach the title's claim would need four things. First, district attendance files covering a full school year on each side of the program, linked with consent. Second, confirmation of infestation at baseline for each apartment, using inspection with interceptor cups, which found far more infestations than residents themselves reported in a 43-building survey (Wang et al., 2016). Third, a comparison group: buildings scheduled for the program later, followed over the same period. Fourth, follow-up of families who move out, so that the most affected households are not lost.

Even then, the study would estimate an association in an observational setting unless treatment order could be assigned. That trade-off returns later in this course, when experimental designs are weighed.

A Second Look at the 71%

Even read as a description, the headline figure needs care. Seventy-one percent of 138 respondents is 98 caregivers. Measured against the 412 eligible households, those 98 are under a quarter of the families the program served. If every nonrespondent had seen no change, the share of all households reporting fewer absences would be about 24%, not 71%. The truth is almost certainly somewhere between those two numbers, and the article's readers deserve to see both bounds. Presenting a figure with its denominator stated, and with the best and worst cases beside it, costs a sentence and protects a report from being quoted as if it described every family in the four buildings.

Conclusion

The composite article answers a descriptive question about caregiver perception and claims to answer a causal question about attendance. The distance between those two questions is the critique's main finding. The study is not useless. It shows that caregivers noticed the program and believed it helped, and it suggests a hypothesis worth testing. Its fault is a title that promises more than one survey of one third of families can deliver.

References

Grimes, D. A., & Schulz, K. F. (2002a). Descriptive studies: What they can and cannot do. The Lancet, 359(9301), 145-149. https://doi.org/10.1016/S0140-6736(02)07373-7

Grimes, D. A., & Schulz, K. F. (2002b). Cohort studies: Marching towards outcomes. The Lancet, 359(9303), 341-345. https://doi.org/10.1016/S0140-6736(02)07500-1

Wang, C., Gibb, T., & Bennett, G. W. (2009). Evaluation of two least toxic integrated pest management programs for managing bed bugs (Heteroptera: Cimicidae) with discussion of a bed bug intercepting device. Journal of Medical Entomology, 46(3), 566-571. https://doi.org/10.1603/033.046.0322

Wang, C., Singh, N., Zha, C., & Cooper, R. (2016). Bed bugs: Prevalence in low-income communities, resident's reactions, and implementation of a low-cost inspection protocol. Journal of Medical Entomology, 53(3), 639-646. https://doi.org/10.1093/jme/tjw018