Date of Award

January 2020

Document Type


Degree Name

Medical Doctor (MD)



First Advisor

Tamar H. Taddei


Introduction: Hepatocellular carcinoma (HCC) requires complex care coordination that can cause delays at many points in the care process. Patient safety may be compromised with untimely follow-up of abnormal liver imaging results. At many tertiary care institutions, optimal treatment is planned at a multidisciplinary liver tumor board (MDLTB). At many tertiary care centers, patients suspected for HCC would immediately be referred to the internal MDLTB for workup and management (internal patients). However, many community hospitals lack the resources to conduct effective MDLTBs. Thus, HCC patients discovered by community providers are often referred externally to nearby tertiary care centers for continued care (external patients). As the timing of external referrals is at individual provider's discretion, the external referral process can create delays and result in different quality of care received for external patients compared to internal patients. This project consists of two studies that examine factors that can reduce quality of care for patients with HCC, such as delayed timeliness of diagnosis and treatment, and explore potential interventions that can improve this care process. The first study evaluated whether the implementation of an electronic medical record-linked abnormal imaging identification and tracking system at a Veterans Hospital improved timeliness and quality of HCC care. The tracking system studied reviewed all liver radiology reports daily via diagnostic codes and natural language processing of reports, generated a queue of abnormal cases for care coordinators to review, and provided a tracking feature to monitor each patient’s care progress. The second study compares internal vs. external patients discussed at a tertiary care center MDLTB to determine whether external patients received lower quality of care compared to internal patients.

Method: For the first study, we designed a retrospective pre-/post-test study to evaluate whether implementation of this tracking system reduced time between HCC diagnosis and treatment, as well as time between first suspicious image and first specialty-care appointment, diagnosis, and treatment. We compared patients diagnosed with HCC in the 36-months before tracking system implementation (pre-intervention) to patients diagnosed with HCC in the 72-months (less a 12-month grace period) after tracking system implementation (post-intervention). We used multivariable regression models, unpaired t-test, Wilcoxon rank-sum test, product limit estimator, and Fisher’s exact test to calculate statistical significance. For the second study, we designed a retrospective comparative cohort study to compare internal and external patients to see whether external patients received a lower quality of care for HCC management. We reviewed incident HCC cases presented over three years at an American College of Surgeons accredited MDLTB. Internal patients were defined as patients whose care originated within the tertiary care institution of the MDLTB; external patients were defined as patients who received initial care from outside institutions and referred to the tertiary care MDLTB for consultation. Differences in care quality were measured by [1] receiving delayed timeliness of care, [2] undergoing unnecessary diagnostic studies, [3] receiving different first treatments, and [4] showing reduced survival.

Results: For the first study, 60 patients were included in the pre-intervention period and 130 in the post-intervention period. Compared to the pre-intervention cohort, time-elapsed for the post-intervention cohort was reduced from 85.0 to 56.2 days from diagnosis-to-treatment

(p<0.05), 86.4 to 39.4 days from imaging-to-diagnosis (p<0.01), and 171.3 to 95.6 days from imaging-to-treatment (p<0.01). For the second study, 122 internal cases and 170 external cases were included. Compared to internal cases, external cases took significantly more time to be diagnosed with HCC (4.5 vs. 38 days, p<0.001) and to be discussed at MDLTB (20 vs. 56 days, p<0.001). Internal cases were more often diagnosed by imaging studies (78%), while external cases were more often diagnosed by biopsy (42%) or by consensus of experts at MDLTB (26%).

Conclusion: This project identified data-driven interventions that quality of care for HCC could be improved. The electronic cancer-care tracking system improved timeliness of HCC diagnosis and treatment, which suggest that this system may be useful for improving HCC care coordination and delivery. We also found that external patients discussed at MDLTB showed delayed timeliness of care and underwent more unnecessary (and often invasive) diagnostic procedures than internal patients. Our results suggest that these differences may be improved for external patients if community providers would immediately refer patients for MDLTB discussion and care upon first suspicion of HCC. A fluid external referral process, similar to that of internal patients, may limit unnecessary diagnostic procedures and result in improved timeliness of care and decreased healthcare costs for patients and the healthcare community.


This thesis is restricted to Yale network users only. This thesis is permanently embargoed from public release.