Changes in Diagnoses and Outcomes for Patients of Hematopathology Specialty and Other Laboratories

Advances in diagnostic testing, including molecular profiling, have improved diagnostic accuracy for hematologic cancers, but with increased need for integration and interpretation of data from multiple tests. Accurate and earlier diagnosis may be achieved with use of hematopathology specialty laboratories, supporting appropriate patient management. This study compares diagnostic changes, patterns of additional testing, treatment decisions, and health care costs for patients with suspected hematologic cancers whose diagnostic tests were managed by specialty hematology and other types of laboratories.

Patients with bone marrow procedure (biopsy/aspirate) and suspected hematologic cancer/disease were identified from claims data (2005 – 2011) from a large US health plan. Included patients had ≥6 pre- & ≥3-months post-biopsy health plan coverage. Lab tests in the 30 days post-biopsy were identified. Patient cohorts were based upon laboratories performing marrow morphology assessment/directing testing sequence: Genoptix (GX, a specialty hematology-testing laboratory); large commercial laboratories (LL), and other laboratories (OL) such as community hospital laboratories; academic labs were not included. Following the bone marrow testing, initial diagnoses were identified. As a proxy for diagnostic uncertainty, we determined whether patients' initial diagnoses remained stable and whether diagnostic changes were recorded in the 1 year following initial biopsy; codes indicative of disease progression or hematologic signs/symptoms were explicitly not counted as instability or change. Health care costs (per patient per month) for the 1 year period post-biopsy were examined. Chi-square and F-tests assessed unadjusted differences between cohorts. Logistic regression assessed differences between cohorts in stability, change in diagnosis and repeat biopsies. Generalized linear models assessed costs, adjusting for demographics, baseline clinical characteristics and initial and final diagnoses.

The study population included a total of 1,387 GX, 4,162 LL, and 19,115 OL patients with suspected hematologic malignancy/disease and bone marrow morphology assessment. OL patients were slightly younger (average age 58.19 OL vs. 59.88 GX, 59.39 LL, P<0.001), and slightly more likely to be enrolled in MedicareAdvantage plans (25.24% OL, vs. 23.00% GX and 21.19% LL, p<0.001). Stability of initial diagnosis varied across the cohorts; 6.16% GX, 8.04% LL, and 9.73% OL patients had unstable initial diagnoses (p<0.001); compared to OL, adjusted odds ratios were 0.864 for GX [95% CI: 0.678,1.1] and 1.07 for LL [95% CI: 0.93, 1.23]. Changes to hematologic diagnoses occurred for 7.88% of GX, 11.19% of LL, and 14.08% OL patients (p<0.001), with adjusted odds ratios of change vs. OL of 0.82 for GX [95% CI: 0.72, 0.93] and 0.93 for LL [95% CI: 0.86, 1.01]. Fewer GX patients underwent repeat marrow biopsies (9.59% GX, vs. 17.11% LL and 28.16% OL, p<0.001), with differences remaining after adjusting for types of cancer diagnoses and other characteristics (odds ratios vs. OL: GX 0.31 [0.25, 0.37]; LL 0.54 [0.49, 0.60]). Among patients who began chemotherapy, 4.58% GX, 6.68% LL, and 7.37% OL (p=0.91) changed treatment within 30 days of treatment initiation; an additional 1.78% GX, 3.68% LL, and 5.12% OL (p=0.001) patients changed treatment within 60 days. 1-year PPPM costs adjusted for differences in patient characteristics were $8,202 GX, $7,711 LL, and $10,302 OL p<0.05); unadjusted costs were $5,362 GX, $6,409 LL, and $10,061 OL (P<0.001). Adjusted costs PPPM excluding the testing period were $6,019 GX, $6,649 LL, and $7,801 OL (p<0.05); unadjusted costs for this period were $3,853 GX, $5,171 LL, and $7,653 OL (P<0.0001).

Stability and changes in hematologic diagnoses varied by the type of lab performing the initial testing on the bone marrow sample, with a trend for fewer changes observed for the hematology specialty lab. Repeat bone marrow biopsies, changes in chemotherapy, and costs in the period following initial diagnostic workup were lower for patients whose samples were assessed by the specialty laboratory versus other types of laboratories after adjusting for differences in patient populations. Further exploration of the impact of management by specialized (hematopathologist) compared to general pathology services on outcomes is warranted.

Engel-Nitz:Novartis Molecular Diagnostics: Research Funding to OptumInsight Other. Eckert:Novartis Molecular Diagnostics: Employment. Song:Novartis Molecular Diagnostics: Research funding to OptumInsight. Other. Hulbert:Novartis Molecular Diagnostics: Research Funding to OptumInsight. Other. McPheeters:Novartis Molecular Diagnostics: Research funding to OptumInsight Other. Teitelbaum:Novartis Molecular Diagnostics: Research funding to OptumInsight. Other.