PHLT 8076 Week 9 Spatial Question Memo Example

Memorandum

To: County board member, District 4

From: Epidemiologist, cancer screening program

Date: March 9

Re: What a map can and cannot tell the board about the mammography unit closing at the southern hospital

The Request

You forwarded a letter from residents of the county's southern tracts, who are worried that the planned closure of the mammography unit at the southern community hospital in June will lead to more women being diagnosed with breast cancer at a late stage. Your note asked a single question: can we map this? The answer is yes, but only if we first decide what question the map should answer. This memo proposes that question, explains what the map would show and sets out what no map could prove. All figures cited are composite.

Turning the Concern Into a Question

The residents' concern contains two claims: that the closure will make screening harder to reach, and that harder access will lead to later diagnoses. A map can address the first directly and the second only indirectly. The proposed question has two parts:

1. How many women aged 40 and older live in tracts where the drive to the nearest open mammography facility will exceed 30 minutes after the closure, compared with before it?

2. In the five years before the closure, was the percentage of breast cancers diagnosed at a late stage higher in tracts with longer drive times, once tract poverty is taken into account?

The first part describes a change in access. The second asks whether access, as measured by drive time, has been associated with late diagnosis in our county in the past.

What the Map Would Show

A map of drive time from each tract to the nearest facility, before and after the closure, would show where access changes. Our preliminary estimate is that about 2,300 women aged 40 and older in seven southern tracts would move from under 30 minutes to over 30 minutes. A second map, of the percentage of diagnoses found late by tract, with drive time shown alongside, would show whether the two patterns line up. Facilities across the county line, including the regional medical center four miles south, would be included, since women use them.

Why Tracts

Tracts are small enough to separate the town around the hospital from the farmland beyond it, and the census publishes the population counts we need. Measuring drive time from each woman's home would be more precise, but the registry's address data cannot be released for that purpose, and tract-level analysis protects confidentiality.

What the Map Could Never Prove

Three limits deserve to be stated plainly before the map is made.

It Cannot Show the Closure Caused Anything

The closure has not yet happened, and late-stage diagnoses take years to change. Even after it happens, a rise in late-stage percentages could reflect other changes, such as insurance coverage or the loss of a primary care practice. A map shows patterns, not causes.

It Cannot Speak for Individual Women

A tract with longer drive times and higher late-stage percentages does not show that the women who were diagnosed late were the ones who lived far away. Robinson (1950) demonstrated long ago that associations among areas can differ from associations among individuals, the ecological fallacy, and that caution applies here.

Distance May Not Be the Main Barrier

A registry study spanning ten states found no link between late-stage diagnosis and travel time to the nearest mammography facility, while poverty and lack of insurance did show associations (Henry et al., 2011). Our map may find the same. If it does, the board should hear that the residents' concern about access is real but that the remedy for late diagnosis may lie in cost and coverage rather than distance alone.

What Results Would Look Like

If the maps show that late-stage percentages track drive time after accounting for poverty, the board would have evidence that access matters locally and a reason to press for replacement screening. If they show that late-stage percentages track poverty but not drive time, the board would still face the closure's access problem but should not expect transportation alone to change stage at diagnosis. Either result would be reported with the limits above. Both maps would go to the board with a single page explaining them in everyday words, and a draft can be ready six weeks after approval. No new data collection is needed.

Recommendation

Commission the two maps described above, using facility locations rebuilt for each year and tract poverty as context, as the access literature recommends for measuring the services a population can actually reach (Wang & Luo, 2005). Present the results with these limits stated on the page. Separately, ask the hospital and the regional medical center what transportation or scheduling support might offset the longer drive for the seven affected tracts, since that question does not need a map to begin answering.

References

Henry, K. A., Boscoe, F. P., Johnson, C. J., Goldberg, D. W., Sherman, R., & Cockburn, M. (2011). Breast cancer stage at diagnosis: Is travel time important? Journal of Community Health, 36(6), 933-942. https://doi.org/10.1007/s10900-011-9392-4

Robinson, W. S. (1950). Ecological correlations and the behavior of individuals. American Sociological Review, 15(3), 351-357. https://doi.org/10.2307/2087176

Wang, F., & Luo, W. (2005). Assessing spatial and nonspatial factors for healthcare access: Towards an integrated approach to defining health professional shortage areas. Health & Place, 11(2), 131-146. https://doi.org/10.1016/j.healthplace.2004.02.003