Predictors of resolution in navigated patients with abnormal cancer screening tests

Patient navigation has been effective in improving cancer care, yet little is known about what predicts timely outcomes in navigated patients.

To identify predictors of resolution of abnormal cancer screening tests in patients who received navigation.

We examined data on patients with abnormal breast (n = 256) or cervical (n = 150) screening tests or symptoms who received navigation as part of the Ohio Patient Navigator Research Program during 2007-2010. We used multivariable Cox proportional hazards regression models to identify predictors of time to resolution (ie, when a patient's clinical abnormality or abnormal screening test was determined to be a benign condition or a cancer diagnosis).

The median time to resolution was 183 days for navigated patients with breast abnormalities and 172 days for navigated patients with cervical abnormalities. In patients with breast abnormalities, those who reported at least 1 barrier to care during navigation (HR, 0.66; 95% CI, 0.51-0.86) or higher perceived stress (HR, 0.90; 95% CI, 0.82-0.98) had slower resolution. Among patients with cervical abnormalities, those who reported at least 1 barrier to care during navigation had slower resolution (HR, 0.62; 95% CI, 0.42-0.91). Patients with cervical abnormalities had faster resolution if they had private health insurance, but this effect was present only in younger women (interaction 𝑃 = .003).

Unknown generalizability of results because patients were female and from clinics in central Ohio.

Several variables predicted whether patient navigation led to faster resolution, and predictors differed somewhat by disease site. Results will be useful in improving current patient navigation programs and designing future programs.