Abstract
Background
The delivery of treatment for hepatocellular carcinoma (HCC) could be influenced by place of HCC diagnosis (hospitalization vs. outpatient), subspecialty referral following diagnosis, as well as physician and facility factors. We conducted a study to examine the effect of patient and non-patient factors on the place of HCC diagnosis, referral, and treatment in Veterans Administration (VA) hospitals in the United States.
Methods
Using the VA Hepatitis C Clinical Case Registry, we identified HCV-infected patients who developed HCC during 1998–2006. All cases were verified and staged according to Barcelona Clinic Liver Cancer (BCLC) criteria. The main outcomes were place of HCC diagnosis, being seen by a surgeon or oncologist, and treatment. We examined factors related to these outcomes using hierarchical logistic regression. These factors included HCC stage, HCC surveillance, physician specialty, and facility factors, in addition to risk factors, co-morbidity, and liver disease indicators.
Results
Approximately 37.2% of the 1,296 patients with HCC were diagnosed during hospitalization, 31.0% were seen by a surgeon or oncologist, and 34.3% received treatment. Being seen by a surgeon or oncologist was associated with surveillance (adjusted odds ratio (aOR)=1.47;95%CI:1.20–1.80) and varied by geography (1.74;1.09–2.77). Seeing a surgeon or oncologist was predictive of treatment (aOR=1.43;95%CI:1.24–1.66). There was a significant increase in treatment among patients who received surveillance (aOR=1.37; 95%CI:1.02–1.71), were seen by gastroenterology (1.65;1.21–2.24) or were diagnosed at a transplant facility (1.48;1.15–1.90).
Conclusions
Approximately 40% of patients were diagnosed during hospitalization. Most patients were not seen by a surgeon or oncologist for treatment evaluation and only 34% received treatment. Only receipt of HCC surveillance was associated with increased likelihood of outpatient diagnosis, being seen by a surgeon or oncologist, and treatment.
Keywords: Hepatocellular carcinoma, surveillance, diagnosis, treatment
INTRODUCTION
Hepatocellular carcinoma (HCC) is currently the fastest rising cause of cancer-related death in the United States, where the incidence of HCC has tripled during the past 20 years 1. The overall 5-year survival is less than 10% in patients who do not receive treatment 2,3, but can be as high as 70% in patients with early stage HCC who received potentially curative treatment (transplant, resection, ablation) 4–14. In addition, palliative HCC-specific treatments, including transarterial chemoembolization (TACE) and systemic chemotherapy may also offer a modest survival benefit among advanced stage HCC patients with compensated liver disease and little co-morbidity.
Population-based studies using Surveillance, Epidemiology, and End Results (SEER)-Medicare linked data reported low rates of HCC-specific treatments 6,15,16. Potentially curative treatments were received by less than 20% of all HCC patients 6, and among patients who did not receive potentially curative therapy, less than 25% received TACE or systemic chemotherapy 15. While highly suggestive of underutilization, these data apply only to elderly patients ages 65 years and older. Moreover, reasons for the low utilization of HCC treatments are unclear.
We hypothesize that the place of initial diagnosis (outpatient setting or hospitalization) for patients with HCC, being seen by a cancer treatment specialist (surgeon or oncologist), and receiving any HCC specific treatment are measurable indicators of HCC quality of care that are likely to influence the outcomes of patients with HCC. Place of initial diagnosis is both a function of pre-diagnosis care as well as a determinant of subsequent HCC management. Variations in HCC quality of care as measured by these three variables may be influenced by patient related clinical factors, pre-HCC diagnosis care such as HCC surveillance, and by provider specialty and facility factors 17–19.
We conducted a retrospective cohort study among HCV-infected patients diagnosed with HCC during 1998–2006 in the Veterans Administration (VA) Hepatitis C Clinical Case Registry (CCR) to evaluate the effect of patient factors, physician specialty, and facility characteristics on the place of initial HCC diagnosis, being seen by a surgeon or oncologist, and treatment receipt.
METHODS
Study population and data source
The study cohort was previously described 20. Briefly, HCV-infected patients with incident HCC between October 1, 1998 and December 31, 2006 were identified using a previously validated algorithm based on the presence of ICD-9 CM code 155.0 and the absence of ICD-9 CM code 155.1 21. Data for this study were obtained from the VA Hepatitis C CCR, which contains patient information collected from 128 VA healthcare facilities nationwide 22. Data elements in the CCR include demographics, laboratory test results and inpatient and outpatient diagnoses and procedure codes. We manually reviewed all potential HCC cases and systematically abstracted information to verify the diagnosis and to determine the Barcelona Clinic Liver Cancer (BCLC) stage at diagnosis from VA electronic medical records. Facility characteristics were obtained from the VA Peer Facility database 23. This database captures facility-level information during 2003 to 2005.
Indicators of HCC management
We examined the following indicators of HCC management: 1) Place of HCC diagnosis based on whether the first reported HCC diagnosis occurred during hospitalization or in an outpatient clinic; 2) Being seen by a surgeon or oncologist around the time of HCC diagnosis as defined by at least one general surgery (401) clinic, oncology clinic (316), chemotherapy procedure unit (330) or chemotherapy support unit (431) stop code within 90-days following the HCC diagnosis date. 3) Receipt of any HCC-specific treatment (liver transplant, surgical resection, local ablation, TACE, or systemic chemotherapy) was identified by International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) or Current Procedure Terminology (CPT) codes within 12-months following HCC diagnosis (see Appendix).
Patient demographic and clinical factors
We ascertained age at HCC diagnosis, gender, and race (white, black, Hispanic, Asian missing). Propensity to use the VA healthcare system was measured by the number of inpatient and outpatient encounters in the 1-year prior to the HCC index date. Clinical variables included year of HCC diagnosis, BCLC HCC stage at diagnosis (A–D), Model for End Stage Liver Disease (MELD) score and indicators of advanced liver disease (ascites, encephalopathy, varices). We also identified medical co-morbidities (diabetes, coronary artery disease, chronic obstructive pulmonary disease, respiratory failure, congestive heart failure, hypertension and end stage renal disease), mental health disorders (anxiety, depression, bipolar disorder, psychosis), and substance use based on ICD-9 codes recorded in the one year prior to and following the HCC index date.
HCC surveillance was defined as receipt of ultrasound or AFP tests for surveillance purposes, as defined by a previously developed and validated algorithm that utilizes laboratory data and ICD-9-CM diagnostic codes available in the VA HCV CCR database 24. Receipt of HCC surveillance was recorded within 0–6 months prior to HCC diagnosis only, 7–24 months only, and both 0–6 months and 7–24 months.
Facility and physician specialty factors
Facility characteristics were based on the VA facility where HCC was diagnosed, and included geographic region (northeast, Midwest, south, west, or Puerto Rico/Virgin Islands), transplant center status (only 5 VA centers: Houston, Nashville, Pittsburgh, Portland, and Richmond), HCC experience (median number of patients diagnosed with HCC between 2003 and 2005), facility size (median number of hospital beds), and academic involvement (ratio of medical resident slots to 10,000 unique patients). Facility size and academic involvement were obtained from the VA Peer Facility database 23. We determined whether patients were seen by primary care only, gastroenterology only, both, or neither within the 90-days prior to the HCC diagnosis date.
Data Analysis
Patient factors and physician specialty listed above were examined as potential determinants of indicators of HCC management, namely location in which the HCC diagnosis was made, being seen by a surgeon or oncologist, and receipt of any HCC-specific treatment. Facility characteristics were also examined as potential determinants for being seen by a surgeon or oncologist and receipt of any HCC-specific treatment. Hierarchical multiple logistic regression models were used to examine the association between potential determinants and each of the three primary HCC management variables. Patients were clustered within VA facilities. Wald chi-square tests were used to assess the significance of each predictor variable. Variables with p<0.10 in univariate logistic regression models were further evaluated in a stepwise multivariable regression analysis. Variables with a known a priori clinical significance were forced into the final models. These variables included year of HCC diagnosis, age, sex, race, MELD score, HCC stage at diagnosis, and indicators of severe liver disease. Other variables that had a p-value<0.10 also were retained in the final model.
A multiple imputation approach was used to analyze missing data for HCC stage at diagnosis and MELD score. Imputations were based on the predicted probabilities of HCC stage and MELD score from logistic regression models. Factors that predicted HCC stage at diagnosis were age, year of HCC diagnosis, HCC surveillance, transplant center status, HCC experience, and facility size. Factors that predicted MELD score were the presence of ascites or encephalopathy. We used 500 imputation iterations for the three main logistic regression models examining location in which the HCC diagnosis was made, being seen by a surgeon or oncologist, and receipt of any HCC-specific treatment. Multiple-imputation adjusted odds ratios, 95% confidence intervals, and Wald test p-values were calculated using SAS proc mianalyze. All statistical analyses were performed using SAS 9.2 (SAS Institute Inc., Cary, NC).
The study protocol was approved by the Institutional Review Board of Baylor College of Medicine and the office of Human Subjects Research of the National Institutes of Health.
Results
The study cohort consisted of 1,296 patients with diagnostically verified HCC during 1998–2006. The mean age of HCC patients was 58.1 years (standard deviation=8.6) and the vast majority were men (99.4%). The greatest proportions of patients were white (56.3%), followed by blacks (27.2%) and Hispanics (10.0%). Approximately 8.6% has BCLC stage A, 20.4% stage B, 35.8% stage C, and 23.9% stage D; staging information was not available for 11.2% of the cohort. Patients were diagnosed with HCC at 116 VA facilities nationwide. Approximately 12% of patients were diagnosed at a VA facility with liver transplant capability. Most patients were diagnosed at VA facilities in the south (37.8%), followed by the west (27.3%), Midwest (19.0%), northeast (13.6%), and Puerto Rico/Virgin Islands (2.3%). The median number of HCC patients diagnosed at each facility between 2003 and 2005 was 74 (IQR: 45–103). The median number of hospital beds was 189 (IQR: 124–295), and the total number of residents training at each facility ranged from 0 to 47.
Place of HCC diagnosis
Approximately 37.2% of patients had their HCC diagnosis first recorded during hospitalization, while approximately 62.8% were first recorded in an outpatient setting (Figure 1). Diagnosis with HCC during hospitalization was significantly more likely among patients with stage D HCC at diagnosis, higher MELD score, clinical indicators of severe liver disease, or co-morbidities (chronic obstructive pulmonary disease, diabetes, psychosis) and substance use (alcohol and cocaine) (Table 1). HCC surveillance tests in the 24 months prior to HCC diagnosis and being seen by gastroenterology with or without primary care or prior to HCC diagnosis were associated with a significant decrease in the likelihood of HCC diagnosis during hospitalization.
Figure 1.
Tree diagram showing pathways of care following HCC diagnosis.
Table 1.
Comparison of patient factors, physician specialty, and facility characteristics between patients with a first recorded HCC diagnosis during a hospitalization versus in an outpatient clinic. (N=1,296)
| Diagnosed during hospitalization (n=482) | Diagnosed in an outpatient clinic (n=814) | P Value† | |
|---|---|---|---|
| Patient factors | |||
| Demographics | |||
| Age (mean, std. deviation) | 57.4 (8.4) | 58.6 (8.5) | 0.02 |
| Gender | 1.00 | ||
| Male | 479 (99.4) | 809 (99.4) | |
| Race | 0.22 | ||
| White | 285 (59.1) | 444 (54.6) | |
| Black | 122 (25.3) | 230 (28.3) | |
| Hispanic | 50 (10.4) | 80 (9.8) | |
| Asian | 3 (0.7) | 10 (1.2) | |
| Missing | 22 (4.6) | 50 (6.2) | |
| # of inpatient and outpatient encounters in the 1-year prior to HCC diagnosis (median, IQR) | 42.0 (23.0–81.0) | 37.0 (23.0–61.0) | <0.01 |
| Clinical variables | |||
| Year of HCC diagnosis | 0.33 | ||
| 1998–2002 | 105 (21.8) | 159 (19.5) | |
| 2003–2006 | 377 (78.2) | 655 (80.5) | |
| Barcelona-Clinic Liver Cancer Stage | <0.01 | ||
| A | 36 (7.5) | 76 (9.3) | |
| B | 75 (15.6) | 190 (23.4) | |
| C | 161 (33.4) | 303 (37.2) | |
| D | 177 (36.7) | 133 (16.3) | |
| Missing | 33 (6.9) | 112 (13.8) | |
| Model for End Stage Liver Disease score | <0.01 | ||
| <15 | 284 (58.9) | 507 (62.3) | |
| 15–20 | 73 (15.2) | 71 (8.7) | |
| >20 | 19 (3.9) | 11 (1.4) | |
| Missing | 106 (22.0) | 225 (27.6) | |
| Indicators of advanced liver disease | |||
| Ascites | 205 (42.5) | 175 (21.5) | <0.01 |
| Encephalopathy | 125 (25.9) | 110 (13.5) | <0.01 |
| Varices | 161 (33.4) | 182 (22.4) | <0.01 |
| Medical co-morbidities | |||
| Coronary artery Disease | 62 (12.9) | 87 (10.7) | 0.24 |
| Chronic heart failure | 27 (5.6) | 28 (3.4) | 0.06 |
| Chronic obstructive pulmonary disease | 102 (21.2) | 135 (16.6) | 0.04 |
| Diabetes | 171 (35.5) | 244 (30.0) | 0.04 |
| Hypertension | 264 (54.8) | 422 (51.8) | 0.31 |
| Mental health co-morbidities | |||
| Anxiety/PTSD | 116 (24.1) | 165 (20.3) | 0.11 |
| Bipolar/Depression | 73 (15.2) | 115 (14.1) | 0.62 |
| Psychosis | 46 (9.5) | 48 (5.9) | 0.01 |
| Substance use | |||
| Alcohol | 234 (48.6) | 259 (31.8) | <0.01 |
| Cocaine | 39 (8.1) | 32 (3.9) | <0.01 |
| HCC surveillance | <0.01 | ||
| 0–6 months only | 91 (18.9) | 234 (28.8) | |
| 7–24 months only | 80 (16.6) | 88 (10.8) | |
| Both | 146 (30.3) | 382 (46.9) | |
| None | 165 (34.2) | 110 (13.5) | |
| Physician specialty prior to HCC diagnosis | <0.01 | ||
| Primary care only | 197 (40.9) | 201 (24.7) | |
| Gastroenterology only | 48 (10.0) | 120 (14.7) | |
| Both | 167 (34.7) | 456 (56.0) | |
| Neither | 70 (14.5) | 37 (4.6) | |
P-values derived from Pearson chi-square tests based on contingency tables
In the multivariable model that adjusted for patient and clinical factors including HCC stage at diagnosis, patients who received HCC surveillance in both the 0–6 months and 7–24 months before HCC diagnosis were 20% less likely to be diagnosed during hospitalization compared to those without surveillance (Table 4). Patients seen by gastroenterology and primary care or by gastroenterology only were 51% and 42%, respectively, less likely to be diagnosed during hospitalization compared to patients seen by primary care only.
Table 4.
Results from three separate multiple logistic regression models examining physician specialty and facility factors associated with location of diagnosis, being seen by a surgeon or oncologist or receipt of HCC-specific treatment. Multiple imputations were used to account for missing values of HCC stage at diagnosis and MELD score.
| Diagnosis during hospitalization (vs. outpatient) | Seen by a surgeon or oncologist (vs. not seen) | Treatment (vs. no treatment) | |
|---|---|---|---|
| Adjusted Odds Ratio* (95% CI) | Adjusted Odds Ratio* (95% CI) | Adjusted Odds Ratio* (95% CI) | |
| Patient factors | |||
| AFP or US before HCC diagnosis | |||
| 0–6 months only | 0.73 (0.58–0.92) | 1.22 (0.97–1.53) | 1.27 (0.99–1.63) |
| 7–24 months only | 1.09 (0.82–1.44) | 0.77 (0.57–1.05) | 0.85 (0.61–1.17) |
| Both | 0.80 (0.65–0.99) | 1.47 (1.20–1.80) | 1.37 (1.02–1.71) |
| None | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| HCC stage at diagnosis | |||
| A | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| B | 0.75 (0.58–0.97) | 1.14 (0.90–1.45) | 1.42 (1.10–1.85) |
| C | 0.86 (0.69–1.08) | 0.96 (0.78–1.19) | 0.94 (0.74–1.18) |
| D | 1.56 (1.20–2.01) | 0.72 (0.55–0.94) | 0.42 (0.31–0.58) |
| Physician specialty at HCC diagnosis | |||
| Gastroenterology only | 0.58 (0.43–0.80) | 1.01 (0.76–1.35) | 1.65 (1.21–2.24) |
| Both | 0.49 (0.40–0.61) | 1.03 (0.83–1.27) | 1.11 (0.89–1.38) |
| Neither | 2.91 (2.04–4.15) | 0.98 (0.68–1.42) | 1.04 (0.72–1.52) |
| Primary care only | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| Place of initial HCC diagnosis | |||
| Outpatient | N/A | 1.14 (0.99–1.31) | 0.73 (0.63–0.86) |
| Inpatient | 1.00 (reference) | 1.00 (reference) | |
| Seen by a surgeon or oncologist | |||
| Yes | N/A | N/A | 1.43 (1.24–1.66) |
| No | 1.00 (reference) | ||
| Facility characteristics | |||
| Liver transplant center | |||
| Yes | N/A | 1.02 (0.80–1.28) | 1.48 (1.15–1.90) |
| No | 1.00 (reference) | 1.00 (reference) | |
| Geographic region | |||
| Midwest | N/A | 1.41 (0.85–2.33) | 1.23 (0.86–1.77) |
| South | 1.74 (1.09–2.77) | 0.87 (0.64–1.17) | |
| West | 1.67 (1.04–2.67) | 1.13 (0.83–1.54) | |
| Puerto Rico/Virgin Islands | 0.15 (0.03–0.82) | 1.01 (0.45–2.27) | |
| Northeast | 1.00 (reference) | 1.00 (reference) | |
| HCC experience | |||
| Median number of HCC Patients | N/A | 1.00 (0.99–1.01) | 1.01 (1.01–1.02) |
| Facility size | |||
| Median number of total beds | N/A | 1.00 (0.99–1.01) | 0.99 (0.99–1.00) |
| Academic affiliation (# of residents) | |||
| 11–17 | N/A | 1.13 (0.89–1.43) | 1.27 (0.97–1.66) |
| 18–22 | 1.05 (0.84–1.31) | 0.99 (0.78–1.26) | |
| >23 | 1.22 (0.99–1.51) | 1.13 (0.90–1.43) | |
| 0–10 | 1.00 (reference) | 1.00 (reference) | |
Adjusted for year of HCC diagnosis, age, sex, race, MELD score, indicators of severe liver disease, and number of inpatient and outpatient encounters in the 1-year prior to HCC diagnosis.
Seen by a specialist after HCC diagnosis
Almost one-third (30.9%) of all patients were seen by a surgeon or oncologist following HCC diagnosis. Among those seen by a surgeon or oncologist, 69.8% had been diagnosed with HCC in an outpatient clinic and only 30.2% were diagnosed during hospitalization (Figure 2). Patient demographic features and the presence of medical co-morbidities (except stage at HCC diagnosis and presence of varices) were not significantly associated with being seen by a surgeon or oncologist (Table 2). Greater proportions of patients who were seen by a surgeon or oncologist had received HCC surveillance or received care from both gastroenterology and primary care prior to HCC diagnosis. Among facility factors, greater HCC experience, larger facility size, and academic involvement were also associated with being seen by a surgeon or oncologist.
Table 2.
Comparison of patient factors, physician specialty, and facility characteristics among patients being seen by a surgeon or oncologist following HCC diagnosis and those not seen by a surgeon or oncologist. (N=1,296)
| Seen by a surgeon or oncologist (%) (n=401) |
Not seen by a surgeon or oncologist (%) (n=895) |
P Value† | |
|---|---|---|---|
| Patient factors | |||
| Demographics | |||
| Age (mean, std. deviation) | 57.9 (8.4) | 58.8 (8.7) | 0.10 |
| Gender | 0.72 | ||
| Male | 399 (99.5) | 889 (99.3) | |
| Race | 0.19 | ||
| White | 234 (58.4) | 495 (55.3) | |
| Black | 112 (27.9) | 240 (26.8) | |
| Hispanic | 28 (7.0) | 102 (11.4) | |
| Asian | 4 (1.0) | 9 (1.0) | |
| Missing | 23 (5.7) | 49 (5.5) | |
| # of inpatient and outpatient encounters in the 1-year prior to HCC diagnosis (median, IQR) | 43.0 (26.0–73.0) | 37.0 (22.0–64.0) | 0.08 |
| Clinical variables | |||
| Year of HCC diagnosis | 0.22 | ||
| 1998–2002 | 90 (22.4) | 174 (19.4) | |
| 2003–2006 | 311 (77.6) | 721 (80.6) | |
| Barcelona-Clinic Liver Cancer Stage | |||
| A | 46 (11.5) | 66 (7.4) | |
| B | 97 (24.2) | 168 (18.8) | |
| C | 140 (34.9) | 324 (36.2) | |
| D | 70 (17.5) | 240 (26.8) | |
| Missing | 48 (12.0) | 97 (10.8) | |
| Model for End Stage Liver Disease score | 0.11 | ||
| <15 | 259 (64.6) | 532 (59.4) | |
| 15–20 | 46 (11.5) | 98 (10.9) | |
| >20 | 5 (1.3) | 25 (2.8) | |
| Missing | 91 (22.7) | 240 (26.8) | |
| Indicators of advanced liver Disease | |||
| Ascites | 270 (30.2) | 110 (27.4) | 0.32 |
| Encephalopathy | 170 (19.0) | 65 (16.2) | 0.23 |
| Varices | 252 (28.2) | 91 (22.7) | 0.04 |
| Medical co-morbidities | |||
| Coronary artery Disease | 99 (11.1) | 50 (12.5) | 0.46 |
| Chronic heart failure | 21 (5.2) | 34 (3.8) | 0.24 |
| Chronic obstructive pulmonary disease | 71 (17.7) | 166 (18.6) | 0.72 |
| Diabetes | 121 (30.2) | 294 (32.9) | 0.34 |
| Hepatitis B virus | 5 (1.3) | 12 (1.3) | 0.89 |
| Hypertension | 490 (54.8) | 196 (48.9) | 0.05 |
| Mental health co-morbidities | |||
| Anxiety/Post traumatic stress disorder | 199 (22.2) | 82 (20.5) | 0.47 |
| Bipolar/Depression | 130 (14.5) | 58 (14.5) | 0.98 |
| Psychosis | 72 (8.0) | 22 (5.5) | 0.10 |
| Substance use | |||
| Alcohol | 140 (34.9) | 353 (39.4) | 0.12 |
| Cocaine | 47 (5.3) | 24 (6.0) | 0.59 |
| HCC surveillance | <0.01 | ||
| 0–6 months only | 107 (26.7) | 218 (24.4) | |
| 7–24 months only | 37 (9.2) | 131 (14.6) | |
| Both | 326 (36.4) | 202 (50.4) | |
| None | 202 (24.6) | 326 (36.4) | |
| Physician specialty prior to HCC diagnosis | 0.02 | ||
| Primary care only | 104 (25.9) | 294 (32.9) | |
| Gastroenterology only | 56 (14.0) | 112 (12.5) | |
| Both | 214 (53.4) | 409 (45.7) | |
| Neither | 27 (6.7) | 80 (8.9) | |
| Place of initial HCC diagnosis | <0.01 | ||
| Hospitalization | 121 (30.2) | 361 (40.3) | |
| Outpatient clinic | 280 (69.8) | 534 (59.7) | |
| Facility characteristics | |||
| Geographic region | 0.02 | ||
| Midwest | 172 (19.2) | 74 (18.5) | |
| Northeast | 121 (13.5) | 55 (13.7) | |
| South | 336 (37.5) | 154 (38.4) | |
| West | 237 (26.5) | 117 (29.2) | |
| Puerto Rico/Virgin Islands | 1 (0.3) | 29 (3.2) | |
| Liver transplant facility | 0.43 | ||
| Yes | 49 (12.2) | 96 (10.7) | |
| No | 352 (87.8) | 799 (89.3) | |
| HCC experience | |||
| Median number of HCC patients (IQR) | 76.0 (51.0–103.0) | 68.0 (40.0–102.0) | 0.05 |
| Facility size | |||
| Median number of total hospital beds (IQR) | 227.0 (133.0–295.0) | 169.0 (123.0–290.0) | <0.01 |
| Academic involvement (based on # of residents in quartiles) | <0.01 | ||
| 0–10 | 74 (18.5) | 236 (26.4) | |
| 11–17 | 97 (24.2) | 203 (22.7) | |
| 18–22 | 106 (26.4) | 247 (27.6) | |
| >23 | 124 (30.9) | 209 (23.4) | |
P-values derived from Pearson chi-square tests based on contingency tables
In the multivariable model (Table 4), patients who received HCC surveillance in both the 0 to 6 months and 7 to 24 months prior to HCC diagnosis were associated with a 47% increased likelihood of being seen by a surgeon or oncologist. Those who received care at a facility in the south or west regions of the United States were 74% and 67%, respectively, more likely to be seen by a surgeon or oncologist. Physician specialty, diagnosis at a liver transplant facility, HCC experience, and facility size were not associated with being seen by a surgeon or oncologist in the adjusted model.
HCC Treatment
Only 34.3% of patients received any HCC-specific treatment. Most patients received TACE (15.3%), followed by ablation (9.1%), systemic chemotherapy (2.3%), resection (4.4%) and transplant (3.2%). Among patients who were most likely to be eligible for HCC-specific treatment (stages A–C), only 40% received therapy. Overall, approximately 58% of patients with HCC stage A (11.6% transplant, 9.8% resection, 17.0% ablation, 23.2% TACE, 8.0% systemic chemotherapy) 42% of stage B (3.0% transplant, 8.7% resection, 11.7% ablation, 24.5% TACE, 6.0% systemic chemotherapy), 35% of stage C, and 16% of stage D received any of those HCC-specific treatments.
Patients who received any HCC-specific treatment were predominantly white or Hispanic compared to those who did not receive any HCC-specific treatment (Table 3). Those who received care at a larger VA facility, designated liver transplant center, with more HCC experience, or greater academic involvement were more likely to receive HCC-specific treatment compared to those who did not receive treatment.
Table 3.
Comparison of patient factors, physician specialty, and facility characteristics among patients who received any HCC-specific treatment compared to those who did not receive HCC-specific treatment. (N=1,296)
| Any HCC-specific treatment (%) (n=445) |
No HCC-specific treatment (%) (n=851) |
P Value† | |
|---|---|---|---|
| Patient factors | |||
| Demographics | |||
| Age (mean, std. deviation) | 57.3 (7.9) | 58.6 (8.8) | <0.01 |
| Gender | 0.85 | ||
| Male | 442 (99.3) | 846 (99.4) | |
| Race | 0.03 | ||
| White | 255 (57.3) | 474 (55.7) | |
| Black | 113 (25.4) | 239 (28.1) | |
| Hispanic | 54 (12.1) | 76 (8.9) | |
| Asian | 7 (1.6) | 6 (0.7) | |
| Missing | 16 (3.6) | 56 (6.6) | |
| # of inpatient and outpatient encounters in the 1-year prior to HCC diagnosis (median, IQR) | 39.0 (25.0–67.0) | 39.0 (22.0–68.0) | 0.59 |
| Clinical variables | |||
| Year of HCC diagnosis | 0.41 | ||
| 1998–2002 | 179 (21.0) | 85 (19.1) | |
| 2003–2006 | 672 (79.0) | 360 (80.9) | |
| Barcelona-Clinic Liver Cancer Stage | <0.01 | ||
| A | 65 (14.6) | 47 (5.5) | |
| B | 111 (24.9) | 154 (18.1) | |
| C | 163 (36.6) | 301 (35.4) | |
| D | 50 (11.2) | 260 (30.6) | |
| Missing | 56 (12.6) | 89 (10.5) | |
| Model for End Stage Liver Disease score | <0.01 | ||
| <15 | 324 (72.8) | 467 (54.9) | |
| 15–20 | 27 (6.1) | 117 (13.8) | |
| >20 | 2 (0.5) | 28 (3.3) | |
| Missing | 92 (20.7) | 239 (28.1) | |
| Indicators of advanced liver Disease | |||
| Ascites | 122 (27.4) | 258 (30.3) | 0.28 |
| Encephalopathy | 73 (16.4) | 162 (19.0) | 0.24 |
| Varices | 127 (28.5) | 216 (25.4) | 0.22 |
| Medical co-morbidities | |||
| Coronary artery Disease | 57 (12.8) | 92 (10.8) | 0.28 |
| Chronic heart failure | 19 (4.3) | 36 (4.2) | 0.97 |
| Chronic obstructive pulmonary disease | 78 (17.5) | 159 (18.7) | 0.61 |
| Diabetes | 153 (34.4) | 262 (30.8) | 0.19 |
| Hypertension | 231 (51.9) | 455 (53.5) | 0.59 |
| Mental health co-morbidities | |||
| Anxiety/PTSD | 103 (23.2) | 178 (20.9) | 0.36 |
| Bipolar/Depression | 81 (18.2) | 107 (12.6) | <0.01 |
| Psychosis | 26 (5.8) | 68 (8.0) | 0.16 |
| Substance use | |||
| Alcohol | 173 (38.9) | 320 (37.6) | 0.65 |
| Cocaine | 23 (5.2) | 48 (5.6) | 0.72 |
| HCC surveillance | <0.01 | ||
| 0–6 months only | 117 (26.3) | 208 (24.4) | |
| 7–24 months only | 46 (10.3) | 122 (14.3) | |
| Both | 302 (35.5) | 226 (50.8) | |
| None | 56 (12.9) | 219 (25.7) | |
| Physician specialty prior to HCC diagnosis | <0.01 | ||
| Primary care only | 81 (18.2) | 317 (37.3) | |
| Gastroenterology only | 81 (18.2) | 87 (10.2) | |
| Both | 376 (44.2) | 247 (55.5) | |
| Neither | 36 (8.1) | 71 (8.3) | |
| Place of initial HCC diagnosis | 0.95 | ||
| Hospitalization | 166 (37.3) | 316 (37.1) | |
| Outpatient clinic | 279 (62.7) | 535 (62.9) | |
| Seen by a surgeon or oncologist | <0.01 | ||
| Yes | 188 (42.3) | 213 (25.0) | |
| No | 257 (57.8) | 638 (75.0) | |
| Facility characteristics | |||
| Geographic region | 0.05 | ||
| Midwest | 87 (19.6) | 159 (18.7) | |
| Northeast | 60 (13.5) | 116 (13.6) | |
| South | 147 (33.0) | 343 (40.3) | |
| West | 137 (30.8) | 217 (25.5) | |
| Puerto Rico/Virgin Islands | 14 (3.2) | 16 (1.9) | |
| Liver transplant center | <0.01 | ||
| Yes | 83 (18.7) | 62 (7.3) | |
| No | 362 (81.4) | 789 (92.7) | |
| HCC experience | |||
| Median number of HCC patients (IQR) | 89.0 (60.0–110.0) | 66.0 (41.0–96.0) | <0.01 |
| Facility size | |||
| Median number of total hospital beds (IQR) | 227.0 (133.0–291.0) | 189.0 (124.0–295.0) | 0.05 |
| Academic involvement (# of resident slots) | <0.01 | ||
| 0–10 | 61 (13.7) | 249 (29.3) | |
| 11–17 | 116 (26.1) | 184 (21.6) | |
| 18–22 | 140 (31.5) | 213 (25.1) | |
| >23 | 132 (28.8) | 205 (24.1) | |
P-values derived from Pearson chi-square tests based on contingency tables
In the multivariable model, patients who received HCC surveillance in both the 0 to 6 months and 7 to 24 months were 37% more likely to receive HCC treatment (Table 4). Receipt of treatment was also 65% higher in patients seen by gastroenterology prior to HCC diagnosis. Patients who were seen by a surgeon or oncologist following HCC diagnosis were 43% more likely to receive treatment compared to patients who did not see a surgeon or oncologist, however only 46.9% of patients who were seen by a surgeon or oncologist received any HCC-specific treatment (Figure 1). Those diagnosed with HCC at a liver transplant center were 48% more likely to receive any HCC-specific treatment compared to those not seen at a liver transplant center (Table 4).
DISCUSSION
In this study of the management of HCC within the VA healthcare system, we found that approximately 37% of all HCC patients were initially diagnosed during hospitalization, less than one-third were seen by a surgeon or oncologist for HCC treatment evaluation, and only one-third received any HCC specific treatment. Among patients who were most likely to receive therapy (BCLC stages A–C), only 40% received treatment (58% stage A, 42% stage B, 35% stage C). Among patient and non-patient factors examined, only receipt of HCC surveillance was significantly associated with diagnosis in an outpatient clinic, increased likelihood of being seen by a surgeon or oncologist, and receipt of treatment. This study provides further evidence that HCC surveillance influences HCC care regardless of patient characteristics, HCC stage at diagnosis, geographic region, or facility factors.
This is the first study to report the utilization of HCC treatment in the national VA system. Several previous studies have examined the utilization of curative and palliative treatments for HCC in non-VA populations 6,15,16. Findings from this study are consistent with these reports and support the low rates of HCC treatment in the United States. This study provides new insights into non-patient factors that contribute to the low treatment rates for HCC, including place of diagnosis, physician specialty at diagnosis and facility-level experience in managing patient with HCC.
We hypothesized that an outpatient diagnosis of HCC would be an indicator of a process of care that is associated with both better pre-diagnosis care (e.g. HCC surveillance) as well as outcomes at the time of diagnosis (e.g. early HCC stage) and after diagnosis (referral and treatment). Only part of this hypothesis was confirmed. Patients diagnosed with HCC in an outpatient location were indeed more likely to receive HCC surveillance, to be seen by a gastroenterologist prior to diagnosis and to be diagnosed with an early HCC stage. Approximately one-third of HCC patients were first diagnosed during hospitalization; this group included those with more advanced HCC, MELD scores greater than 15, indicators of severe liver disease, presence of medical co-morbidity, and substance abuse. However, surprisingly an outpatient location of diagnosis was associated with less likelihood of receiving HCC treatment than diagnosis during hospitalization. We believe that this finding indicates a breakdown in the timely identification of HCC diagnosis, patient recall, and subsequent referral; these processes are considerably easier to achieve in a hospitalized patient.
We found that only 31% of all HCC patients were seen by a surgeon or oncologist for treatment evaluation. There are potential explanations for the low rates of referral including physician bias regarding the perceived ineffectiveness of HCC treatment or lack of knowledge about currently available treatment options. Given the increasing number of effective HCC-specific treatments now available for patients with HCC, all patients irrespective of stage should be seen by a specialist for formal treatment evaluation. In this study, receiving HCC surveillance was predictive of a greater likelihood of being seen by a surgeon or oncologist.
Only HCC surveillance was significantly associated with diagnosis in an outpatient clinic, as well as increased likelihood of being seen by a surgeon or oncologist and receipt of treatment. This finding corroborates results from several other studies that demonstrated that HCC surveillance is associated with earlier stage at HCC diagnosis, receipt of treatment, and improved survival 26–32. Findings from this study, while clearly limited by their observation retrospective design, provide further evidence to support the use of HCC surveillance in patients at risk of developing HCC. Furthermore, after adjusting for surveillance and stage of HCC at the time of diagnosis, physician and facility related factors were independent and important predictors of receiving HCC treatment.
Patients seen by a gastroenterologist were more likely to be diagnosed in an outpatient setting and receive treatment than patients not seen by these physicians. Physician specialty has been reported as a significant predictor in other areas of liver disease management, including HCC surveillance. In our previous study 33, receiving pre-HCC diagnosis care from both a primary care physician and a gastroenterologist was associated with a 3-fold increase in receipt of surveillance.
Our study suggests an underutilization of treatment in VA patients with early stage HCC. However, it would be premature to equate these rates with performance or quality measures. It is possible that appropriate reasons for including or excluding patients from HCC treatment were present, such as patient preferences or co-morbidity. Detailed structured record review to identify these reasons, which was not done in this study, is required. We are currently conducting such a study.
The findings from the study must be interpreted within their limitations. The HCV CCR only contains patients with HCV-related HCC and therefore our results may not be generalizable to populations with other underlying HCC risk factors. In addition, these findings are limited to a VA practice setting. Yet, previous studies have shown that VA and non-VA facilities within the same geographic area exhibit similar practice patterns of healthcare use. 17 Lastly, the availability of HCC treatments for patients diagnosed with all stages of HCC has increased since the time period of this study. Physician awareness about HCC treatment may have also improved. We suspect that the observed utilization of HCC treatment during this study time period is low compared to current treatment practices.
In conclusion, receipt of any HCC-specific treatment is relatively low among newly diagnosed patients with any stage HCC. HCC surveillance was associated with increased likelihood of being seen by a surgeon or oncologist and receipt of treatment. However, irrespective of stage of HCC at the time diagnosis, there are several important potentially modifiable factors related to process of care such as physician specialty, referral process related to place of diagnosis, and experience of the facility that affect the receipt of HCC treatment.
Abbreviations
- HCC
Hepatocellular carcinoma, abdominal ultrasound
- VA
Veterans Administration
Appendix: ICD-9 and CPT codes used to determine the receipt of HCC-specific treatment
| Codes | |
|---|---|
| Liver transplantation | ICD-9-CM: 50.5, 50.59, 50.51, V42.7, 50.4 CPT: 47135, 47136, 47140, 47141, 47142 |
| Surgical resection | ICD-9-CM: 50.21, 50.22, 50.3 CPT: 47120, 47122, 47125, 47130 |
| Local ablation | ICD-9-CM: 50.29 CPT: 47370, 76490, 76362, 47380, 47382 |
| Transarterial chemoembolization |
Embolization ICD-9-CM: 38.80, 38.86 CPT: 37204, 75894 Chemotherapy within 30-days of embolization ICD-9-CM: 99.25 CPT: J9000, J9280, J9060, 96405, 96408, 96420, 96422, 96423, 96425, 96440, 96445, 96545, 96549, 0331, 0335 |
| Systemic chemotherapy | ICD-9: 99.25 CPT: J9000, J9010, J9190, J9200, J9201, J9217, J9265, J9060, J9062, J9170, J9178, J9181, J9182, J9280, J9293, J9370, J9015, J9017, J9035, J9202, J9055, 90782, 96400, 96405, 96408, 96410, 96412, 96414, 96420, 96422, 96423, 96425, 96440, 96445, 96545, 96549, 0331, 0332, 0335 |
Footnotes
Conflicts of Interest: No conflicts of interest exist for Drs. Davila, Kramer, Duan, Richardson, Tyson, Sada, Kanwal, or El-Serag.
References
- 1.El-Serag HB. Hepatocellular carcinoma. N Engl J Med. 2011;365:1118–1127. doi: 10.1056/NEJMra1001683. [DOI] [PubMed] [Google Scholar]
- 2.Arii S, Yamaoka Y, Futagawa S, et al. Results of surgical and nonsurgical treatment for small-sized hepatocellular carcinomas: a retrospective and nationwide survey in Japan. The Liver Cancer Study Group of Japan. Hepatology. 2000;32:1224–1229. doi: 10.1053/jhep.2000.20456. [DOI] [PubMed] [Google Scholar]
- 3.Llovet JM, Burroughs A, Bruix J. Hepatocellular carcinoma. Lancet. 2003;362:1907–1917. doi: 10.1016/S0140-6736(03)14964-1. [DOI] [PubMed] [Google Scholar]
- 4.El-Serag HB, Giordano TP, Kramer J, Richardson P, Souchek J. Survival in hepatitis C and HIV co-infection: a cohort study of hospitalized veterans. Clin Gastroenterol Hepatol. 2005;3:175–183. doi: 10.1016/s1542-3565(04)00620-2. [DOI] [PubMed] [Google Scholar]
- 5.El Serag HB. Hepatocellular carcinoma: recent trends in the United States. Gastroenterology. 2004;127:S27–S34. doi: 10.1053/j.gastro.2004.09.013. [DOI] [PubMed] [Google Scholar]
- 6.El Serag HB, Siegel AB, Davila JA, et al. Treatment and outcomes of treating of hepatocellular carcinoma among Medicare recipients in the United States: a population-based study. J Hepatol. 2006;44:158–166. doi: 10.1016/j.jhep.2005.10.002. [DOI] [PubMed] [Google Scholar]
- 7.Franca AV, Elias JJ, Lima BL, Martinelli AL, Carrilho FJ. Diagnosis, staging and treatment of hepatocellular carcinoma. Braz J Med Biol Res. 2004;37:1689–1705. doi: 10.1590/s0100-879x2004001100015. [DOI] [PubMed] [Google Scholar]
- 8.Lencioni R. Loco-regional treatment of hepatocellular carcinoma. Hepatology. 2010;52:762–773. doi: 10.1002/hep.23725. [DOI] [PubMed] [Google Scholar]
- 9.Liu JH, Chen PW, Asch SM, Busuttil RW, Ko CY. Surgery for hepatocellular carcinoma: does it improve survival? Ann Surg Oncol. 2004;11:298–303. doi: 10.1245/aso.2004.03.042. [DOI] [PubMed] [Google Scholar]
- 10.Livraghi T, Bolondi L, Buscarini L, et al. No treatment, resection and ethanol injection in hepatocellular carcinoma: a retrospective analysis of survival in 391 patients with cirrhosis. Italian Cooperative HCC Study Group. J Hepatol. 1995;22:522–526. doi: 10.1016/0168-8278(95)80445-5. [DOI] [PubMed] [Google Scholar]
- 11.Massarweh NN, Park JO, Farjah F, et al. Trends in the utilization and impact of radiofrequency ablation for hepatocellular carcinoma. J Am Coll Surg. 2010;210:441–448. doi: 10.1016/j.jamcollsurg.2009.12.026. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Mazzaferro V, Regalia E, Doci R, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med. 1996;334:693–699. doi: 10.1056/NEJM199603143341104. [DOI] [PubMed] [Google Scholar]
- 13.N’Kontchou G, Mahamoudi A, Aout M, et al. Radiofrequency ablation of hepatocellular carcinoma: long-term results and prognostic factors in 235 Western patients with cirrhosis. Hepatology. 2009;50:1475–1483. doi: 10.1002/hep.23181. [DOI] [PubMed] [Google Scholar]
- 14.Ringe B, Pichlmayr R, Wittekind C, Tusch G. Surgical treatment of hepatocellular carcinoma: experience with liver resection and transplantation in 198 patients. World J Surg. 1991;15:270–285. doi: 10.1007/BF01659064. [DOI] [PubMed] [Google Scholar]
- 15.Davila JA, Duan Z, McGlynn KA, El-Serag HB. Utilization and outcomes of palliative therapy for hepatocellular carcinoma: A population-based study in the United States. J Clin Gastroenterol. 2012;46(1):71–7. doi: 10.1097/MCG.0b013e318224d669. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Shah SA, Smith JK, Li Y, Ng SC, Carroll JE, Tseng JF. Underutilization of therapy for hepatocellular carcinoma in the medicare population. Cancer. 2011;117:1019–1026. doi: 10.1002/cncr.25683. [DOI] [PubMed] [Google Scholar]
- 17.Ashton CM, Petersen NJ, Souchek J, et al. Geographic variations in utilization rates in Veterans Affairs hospitals and clinics. N Engl J Med. 1999;340:32–39. doi: 10.1056/NEJM199901073400106. [DOI] [PubMed] [Google Scholar]
- 18.Kanwal F, Hoang T, Spiegel BM, et al. Predictors of treatment in patients with chronic hepatitis C infection - role of patient versus nonpatient factors. Hepatology. 2007;46:1741–1749. doi: 10.1002/hep.21927. [DOI] [PubMed] [Google Scholar]
- 19.McGinn J, Davis C. Geographic variation, physician characteristics, and diabetes care disparities in a metropolitan area, 2003–2004. Diabetes Res Clin Pract. 2006;72:162–169. doi: 10.1016/j.diabres.2005.10.023. [DOI] [PubMed] [Google Scholar]
- 20.El-Serag HB, Kramer JR, Chen GJ, Duan Z, Richardson PA, Davila JA. Effectiveness of AFP and ultrasound tests on hepatocellular carcinoma mortality in HCV-infected patients in the USA. Gut. 2011;60:992–997. doi: 10.1136/gut.2010.230508. [DOI] [PubMed] [Google Scholar]
- 21.Davila JA, Weston A, Smalley W, El Serag HB. Utilization of screening for hepatocellular carcinoma in the United States. J Clin Gastroenterol. 2007;41:777–782. doi: 10.1097/MCG.0b013e3180381560. [DOI] [PubMed] [Google Scholar]
- 22.Backus LI, Gavrilov S, Loomis TP, et al. Clinical Case Registries: simultaneous local and national disease registries for population quality management. J Am Med Inform Assoc. 2009;16:775–783. doi: 10.1197/jamia.M3203. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Byrne MM, Daw CN, Nelson HA, Urech TH, Pietz K, Petersen LA. Method to develop health care peer groups for quality and financial comparisons across hospitals. Health Serv Res. 2009;44:577–592. doi: 10.1111/j.1475-6773.2008.00916.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Richardson PA, Henderson LM, Davila JA, Kramer JR, Chen G, El Serag HB. Surveillance for Hepatocellular Carcinoma: Development and Validation of an Algorithm to Classify Tests in Administrative and Laboratory Data. Dig Dis Sci. 2010 Sep 16; doi: 10.1007/s10620-010-1387-y. Epub ahead of print. [DOI] [PubMed] [Google Scholar]
- 25.VA Information Research Center. [on 14 February 2012.];VA/CMS Data for Research. Accessed at www.virec.research.va.gov/
- 26.Kuo YH, Lu SN, Chen CL, et al. Hepatocellular carcinoma surveillance and appropriate treatment options improve survival for patients with liver cirrhosis. Eur J Cancer. 2010;46:744–751. doi: 10.1016/j.ejca.2009.12.018. [DOI] [PubMed] [Google Scholar]
- 27.Tong MJ, Sun HE, Hsien C, Lu DS. Surveillance for hepatocellular carcinoma improves survival in Asian-American patients with hepatitis B: results from a community-based clinic. Dig Dis Sci. 2010;55:826–835. doi: 10.1007/s10620-009-1059-y. [DOI] [PubMed] [Google Scholar]
- 28.Trevisani F, De NS, Rapaccini G, et al. Semiannual and annual surveillance of cirrhotic patients for hepatocellular carcinoma: effects on cancer stage and patient survival (Italian experience) Am J Gastroenterol. 2002;97:734–744. doi: 10.1111/j.1572-0241.2002.05557.x. [DOI] [PubMed] [Google Scholar]
- 29.Trevisani F, Cantarini MC, Labate AM, et al. Surveillance for hepatocellular carcinoma in elderly Italian patients with cirrhosis: effects on cancer staging and patient survival. Am J Gastroenterol. 2004;99:1470–1476. doi: 10.1111/j.1572-0241.2004.30137.x. [DOI] [PubMed] [Google Scholar]
- 30.Trevisani F, Santi V, Gramenzi A, et al. Surveillance for early diagnosis of hepatocellular carcinoma: is it effective in intermediate/advanced cirrhosis? Am J Gastroenterol. 2007;102:2448–2457. doi: 10.1111/j.1572-0241.2007.01395.x. [DOI] [PubMed] [Google Scholar]
- 31.Yu EW, Chie WC, Chen TH. Does screening or surveillance for primary hepatocellular carcinoma with ultrasonography improve the prognosis of patients? Cancer J. 2004;10:317–325. doi: 10.1097/00130404-200409000-00009. [DOI] [PubMed] [Google Scholar]
- 32.Yuen MF, Cheng CC, Lauder IJ, Lam SK, Ooi CG, Lai CL. Early detection of hepatocellular carcinoma increases the chance of treatment: Hong Kong experience. Hepatology. 2000;31:330–335. doi: 10.1002/hep.510310211. [DOI] [PubMed] [Google Scholar]
- 33.Davila JA, Morgan RO, Richardson PA, Du XL, McGlynn KA, El-Serag HB. Use of surveillance for hepatocellular carcinoma among patients with cirrhosis in the United States. Hepatology. 2010;52:132–141. doi: 10.1002/hep.23615. [DOI] [PMC free article] [PubMed] [Google Scholar]