Ana M. Progovac
Ana M. Progovac is first author of the sharpest counter-result in the optimism and health literature, and it sits inside the same cohort and uses the same instrument as the headline longevity paper. Linking 48,291 Women's Health Initiative participants to Medicare claims, her team found that women with the lowest optimism scores received about 5 per cent more frequent screening mammograms after complete covariate adjustment, which they state in print was contrary to their hypotheses.
Who they are
Progovac is a health services researcher. Her 2019 paper in Medicine is a study of behaviour rather than of outcomes, and that is what makes it awkward for the standard explanation of why optimists appear to live longer.
The design uses Women's Health Initiative observational cohort survey data linked to Medicare claims: 48,291 women with no history of breast cancer, enrolled in Medicare Parts A and B for at least two years between 2005 and 2010, with optimism measured at baseline on the revised Life Orientation Test and screening frequency modelled with repeated-events survival models. Claims data rather than self-report is the right measure for an attendance question, and the sample is large.
The result, verbatim from the abstract: contrary to our hypotheses, repeated events survival models showed that women with the lowest optimism scores, meaning more pessimistic tendencies, received 5 per cent more frequent screenings after complete covariate adjustment, p below .01, compared to the most optimistic group, and showed no association between cynical hostility and frequency of screening mammograms. Sociodemographic factors did not appear to mediate the relationship between optimism and screenings, while higher education and higher income were each associated with more frequent screening.
Her team's own qualification travels with it and should not be dropped: the results for optimism were primarily driven by women who were aged 75 or older after January 2009, when changes to clinical guidelines led to uncertainty about the risks and benefits of screening in that age group. That is a real limit on how far the finding generalises, and it does not rescue the behavioural story it cuts against.
Why this result matters more than its modest size suggests is that it is not an exotic outcome measured in an unrelated sample. It is the same cohort as the 2022 longevity paper and the same questionnaire. And attendance at a physical examination for screening purposes is one of the behavioural covariates in Lee and colleagues' main model, the panel of behaviours whose addition takes the women's life-span estimate from 14.9 per cent to 8.7. So one of the variables doing the attenuation in the headline paper points the wrong way in the cohort next door. The mechanism that is supposed to explain the association predicts that optimists attend screening more. Here they attended less.
There is a smaller and older literature on unrealistic optimism about personal breast cancer risk and mammography attendance, pointing in the same direction. It has not been read by this publication, carries no figures here, and is mentioned only as existing. This result stands on its own and is better evidence.
Our read is the abstract, read in full at an institutional repository that prints it entire. An erratum exists, reported as correcting two authors' names and affiliations without changing the findings; the erratum itself was not read, so its existence is cited and not its content. Her affiliation is left blank here rather than guessed.
Key papers
Progovac is a working researcher whose relevant output is journal articles rather than books.
Progovac, A. M., Pettinger, M., Donohue, J. M., Chang, C.-C. H., Matthews, K. A., Habermann, E. B., Kuller, L. H., Rosal, M., Li, W., Garcia, L., & Tindle, H. A. (2019). "Optimism may moderate screening mammogram frequency in Medicare: A longitudinal study." Medicine 98(24), e15869, doi:10.1097/MD.0000000000015869. Abstract read in full for this article; full text not read.
Key ideas
In 48,291 Women's Health Initiative participants linked to Medicare claims, the least optimistic women received about 5 per cent more frequent screening mammograms after full adjustment.
The finding ran against the authors' own stated hypothesis, which they say in print. That is worth more than a confirmatory result of the same size.
Education and income were each associated with more frequent screening, while sociodemographic factors did not appear to mediate the optimism relationship.
The effect was concentrated in women aged 75 or older after the 2009 guideline change, a period of genuine clinical uncertainty for that group.
Screening attendance is one of the behavioural covariates in the headline longevity model, so this is not a peripheral behaviour.
Sources
- Progovac, A. M., Pettinger, M., Donohue, J. M., and colleagues (2019). Medicine 98(24), e15869, doi:10.1097/MD.0000000000015869. Abstract only. An erratum exists and was not read.
- Lee, L. O., James, P., Zevon, E. S., and colleagues (2019). Proceedings of the National Academy of Sciences 116(37), 18357-18362, doi:10.1073/pnas.1900712116. Read in full. Cited here for screening attendance as a behavioural covariate.
- No affiliation is stated for her because none is recorded in the material read. No other biographical detail is asserted here.