Exploring women's health issues (by Tanya Kenkre and Bernard Goldstein)

The Pittsburgh region does not compare well with benchmark regions in indicators of women’s health, both for our white and black populations.  We evaluated a variety of measures that are used to measure adult women’s health, other than those related to pregnancy.  Data from the two most recent nationwide CDC telephone surveys show that in 2006 we were last and 2008 next to last among 15 comparable regions in mammography rates for black women.  For white women we were 14th in 2006, and for 2008 we are still below average, but not as dramatically.  When we looked at another test of proven value for preventing cancer, the pap test for early cervical cancer, white women in our region also do very poorly. In 2006 we ranked 14th and in 2008 we ranked last among the 15 benchmark regions.  Pap test rates for black women are not included in this indicator because we are not satisfied that the sample size of the most recent survey leads to reliable results.  Earlier survey results, however, on Pap smears for black Pittsburghers was as negative as it is for white Pittsburghers.  

Nationwide there is no difference between white and black women over the age of 40 in mammography rates in the past two years for which data are available, both being at 75.7%, while white women in benchmark regions report somewhat higher mammography rates than black women, 77.1% vs. 74.5%.  Both white and black women in Pittsburgh are well below these averages.  Again, it is black women who have the far lower rate (57.9% vs. 72.5% in white women).  

Why do we do so poorlyin these two important measures of disease prevention?  One possible partial explanation is that women in our area have relatively low rates of health care coverage.   For white women we are third from the bottom and for black women we rank last.  This is unlikely to be the full explanation – less than 10% of white women are not covered by a health plan while close to 20% have not had a pap test within the past three years and 27.5% over the age of 40 have not had a mammogram in the past two years.  20% of black women report not having health care coverage – twice the rate of white women, but again not likely to beenough to account for the difference observed in mammography and pap smear rates.

We also compared the percent of women who reported that they did not see a doctor because of cost in the previous year.  Although indirect, this might tell us something about whether cost issues played a role in the failure to obtain mammography or a pap test.    Again, black women in the Pittsburgh MSA have a higher rate of reporting that cost was a deterrent to see a doctor than did white women (15.2% vs 12.0%); but for this indicator black women in Pittsburgh did better than the national or benchmark averages for black women, while Pittsburgh white women did somewhat worse than the benchmark average and were equal to the national average for white women. 

How good are these data?   There is always the potential for faulty sampling or misreporting on telephone survey data.  Data problems in terms of consistency across some of the possible explanatory metrics are present, so we caution against overinterpreting the findings.  The limitations of these data sources are magnified when the population size becomes too small.  For example, the pap smear rates reported for African-American women were among the lowest in comparison with benchmark areas in 2006, but in 2008 were among the highest.  This most likely represents statistical variation among surveys due to the small sample of randomly chosen women.   Mortality, which is based on all death certificates, is a much more stable measure.  Accordingly, we place more reliance on CDC’s telephone survey data for white women in the Pittsburgh area as the sample size is larger; and more reliance on mortality data than on telephone survey responses. 

One approach to check on the validity of the benchmark approach is to evaluate whether the data conforms to expectations.  In addition to data on black and white women, the CDC also reports on Hispanic women.  Because we have created few low end jobs in recent years, the Pittsburgh MSA is thought to have had a low rate of Hispanic immigrants.  The Hispanic community, although small, is on average believed to be relatively more affluent and educated than Hispanic communities in the rest of the country.  As expected, and providing some support to the overall validity of the indicator data, for both 2006 and 2008 CDC reported a much higher rate of mammography, pap smears and health plan enrollment than for Hispanic communities in other benchmark cities, and a lower rate of having not seen a doctor because of cost considerations.   Hispanic women do better than white or black women in our MSA when examining Pap tests and mammography rates; they also do better than black women when examining rates of health care coverage and not seeing a doctor due to cost.  With respect to these last two measures, they do about as well as white women in our MSA.  Nationally they do worse than white or black women in all but mammography rates which are about the same. 

Finally, and most importantly, what difference does it make that white and black women in the Pittsburgh region have lower rates for preventive activities?  One way to check is to look at breast cancer rates in our areas as compared to the benchmark areas and the rest of the country.  (We could not look at cervical cancer rates because the incidence for black women is lower than acceptable for benchmark comparisons).  Unfortunately, the breast cancer mortality rates in the Pittsburgh region are higher for both white and black women than for the rest of the US or for the benchmark areas.  In our region, the breast cancer mortality for white women in 2004, the last year for which data are available, was 27.3 per 100,000, as compared to the benchmark average of 24.9 and the national average of 23.8.  For black women, the Pittsburgh MSA mortality rate for breast cancer was 34.2 as compared to the benchmark average of 31.7 and the national average of 32.3.   In comparison to other cities we are ranked 9th for black women; but for white women only Philadelphia had a higher breast cancer mortality in 2004.  We emphasize that the data can be expected to move back and forth through the years, but the findings are consistent.   Mammography is a proven preventive measure for the early detection of breast cancer when it is far more likely to be treatable.   Low mammography rates predict that more women in our area will die of breast cancer – and that is what is happening.