Healthcare Fragmentation and the Frequency of Radiology and Other Diagnostic Tests: A Cross-Sectional Study

Lisa M. Kern, Joanna K. Seirup, Lawrence P. Casalino, Monika M. Safford

Fragmented ambulatory care has been associated with high rates of emergency department visits and hospitalizations, but effects on other types of utilization are unclear. To determine whether more fragmented care is associated with more radiology and other diagnostic tests, compared to less fragmented care. We conducted a cross-sectional study using claims from five commercial payers for 2010. The study took place in the Hudson Valley, a seven-county region in New York State. We included adult patients who were insured through the participating payers and were attributed to a primary care physician in the region. We restricted the cohort to those with ≥4 ambulatory visits, as measures of fragmentation are not reliable if based on ≤3 visits (N = 126,801). For each patient, we calculated fragmentation using a reversed Bice-Boxerman Index, which we divided into seven categories. We used negative binomial regression to determine the association between fragmentation category and rates of radiology and other diagnostic tests, stratified by number of chronic conditions and adjusting for patient age, gender, and number of visits. Patients with the most fragmented care had approximately twice as many radiology and other diagnostic tests as patients with the least fragmented care, across all groups stratified by number of chronic conditions (each adjusted p < 0.0001). For example, among patients with ≥5 chronic conditions, those with the least fragmented care had 258 tests per 100 patients, and those with the most fragmented care had 542 tests per 100 patients (+284 tests per 100 patients, or +110 %, adjusted p < 0.0001). More fragmented care was independently associated with higher rates of radiology and other diagnostic tests than less fragmented care.