Cost and value in liver disease guidelines: 2011–2022

Background: Chronic liver disease (CLD) is associated with rising health care utilization and cost. We aimed to describe the frequency of cost/value (C/V) statements in CLD-related clinical guidance documents (CGDs). Methods: CGD with a focus on CLD published between January 2011 and February 2022 from 3 US societies [Association for the Study of Liver Diseases (AASLD), American College of Gastroenterology (ACG), and American Gastroenterological Association (AGA)] were analyzed. Findings: Forty-five CGDs were identified. Eighty of 1334 guidance statements were C/V statements (6%). Only 1.1% reported patient-level costs and none reported out-of-pocket costs. Despite the increased importance of incorporating cost and value into care, the proportion of C/V statements in CGDs related to liver disease is low.


INTRODUCTION
The economic burden of liver disease in the United States is rising. In 2016 alone, health care spending for patients with chronic liver disease (CLD) exceeded $32 billion, or 1.2% of total US health care expenditures. [1] As costs of liver disease care rise, there is a growing need for more explicit and transparent assessments of the financial burden of liver disease on patients. [2] Professional society clinical guidance documents (CGD) are created to inform clinicians of best clinical practices and to encourage shared decision making with patients to promote the delivery of high-value care. While cost-effectiveness data are often considered in the   expert consensus documents (ECD), and 1 (2.2%) had both ECD and CPG components. All CGDs included in the analyses are described in Table 1. Cost and value considerations were classified based on protocols adapted from prior studies. [3,4] See Supplementary Material, http://links.lww.com/HC9/A18 for further description of the methodology. In the 45 CGD, a total of 1334 guidance statements were made, of which 80 (6.0%) were C/V statements [AASLD (n = 50), AGA (n = 23), and ACG (n = 7)]. Nine (20.0%) CGD contained no C/V statements. The median number of C/V statements was 1 (interquartile range: 1-2.3) in ECD and 2 (interquartile range: 1-2) in CPG. A similar proportion of the 80 C/V statements were found in ECD (n = 41, 51.3%) compared to CPG (n = 39, 48.7%). The AGA had the highest percentage of C/V statements out of the total number of guidance statements (16.8%) compared to the AASLD (5.3%) and the ACG (3.3%) (P < 0.001) (Supplemental Figure 1, http://links.lww.com/HC9/A16).
Of the 80 C/V statements, the majority (85%) used costs implicitly to support the recommendation, whereas 12.5% explicitly stated that cost was integrated into the recommendation. One statement intentionally excluded cost in its considerations. When comparing CGD from 2011 to 2016 versus 2017 to 2022, there was no difference in the proportion of C/V statements between these time periods (6.3% vs. 5.8%, p = 0.7).
Of the 80 C/V statements, 15 (18.8%) highlighted the economic impact of liver disease or its management (Supplemental Figure 2, http://links.lww.com/ HC9/A17). An example of this was in the AGA Clinical Update on the Interaction between oral direct-acting antiviral agents for chronic hepatitis C and HCC: "In patients with cirrhosis, compared with no surveillance, semi-annual HCC surveillance every 6 months for up to 15 years was cost-effective using willingness-to-pay threshold of $100,000/quality-adjusted life-year." [5] Five statements (6.3%) advocated for C/V issues in clinical care, such as in the ACG acute-o-chronic liver failure clinical guidelines: "Given the expense, logistic challenges of setting up infusions and potential for causing pulmonary edema, the effectiveness of IV albumin in conditions other than spontaneous bacterial peritonitis and postparacentesis circulatory dysfunction needs more study." [6] Eight statements (10.0%) reported gaps in C/V evidence in hepatology, such as AASLD's palliative care and symptom-based management in decompensated cirrhosis practice guidelines: "Assessing financial burden is also important for patients with decompensated cirrhosis, who often report high rates of cost-related nonadherence to medications and food insecurity." [7] Lastly, 52 (65.0%) C/V statements justified guidance recommendations, such as the AASLD primary biliary cholangitis practice guidelines: "The dose of UDCA is important. A study comparing three different doses of UDCA showed that a dose of 13 to 15 mg/kg/day appeared superior…in biochemical responses and cost." [8] Statement categorization did not differ by clinical CGD type (p = 0.3) or time period (p = 0.3).
Of the 52 C/V statements that justified guidance recommendations, 37 (71.1%) focused on societal-level costs and 15 (28.9%) focused on patient-level costs. Of these 52 C/V statements justifying guidance recommendations, 34 (65.4%) recommended use of an intervention due to its incremental benefit justifying the additional cost (n = 20), equal effectiveness at lower cost (n = 8), or to reduce future costs (n = 6). For example, the AGA clinical practice update on screening and surveillance for HCC in patients with NAFLD justified screening individuals with NAFLD for HCC because the incidence of HCC in this population was estimated to be > 1.5%. [9] A total of 18 (34.6%) C/V statements discouraged the use of an intervention either due to an uncertain cost-tobenefit ratio (n = 9) or because its incremental benefit did not justify the additional cost (n = 9). For example, in the AASLD guidance statement on malnutrition, frailty, and sarcopenia, routine use of computed tomography was not recommended in the assessment of muscle mass in patients with cirrhosis due to cost as well as exposure to ionizing radiation. [10] In the AASLD guidance statement on the diagnosis and management of NAFLD, routine screening for nonalcoholic steatohepatitis in individuals with diabetes was not felt to be costeffective because of disutility associated with available treatment and thus was not recommended. [11] No C/V statements reported estimated out-of-pocket costs for recommended therapies.
Despite the increased recognition of the rising economic and financial burden of liver disease care, our analysis demonstrates that C/V assessments were present in only 6% of guidance statements in CGD from the AASLD, AGA, and ACG. The majority of available C/V assessments justified guidance recommendations based on cost-effectiveness analyses that supported the use of an intervention despite the additional cost. Notably, only 15 (1.1%) out of a total of 1334 guidance statements reported patient-level costs of care and none explicitly reported out-of-pocket costs.
With limited patient-level data available on costs of liver disease care, it becomes challenging for clinicians to support patients with liver disease in their medical decision making based on their capacity to pay for the available treatment options. Prior work has shown that while clinicians acknowledge the responsibility to have cost discussions with patients, very few report a firm understanding of the out-of-pocket costs of health care. This deficit has significant implications as diagnostic, surveillance, and therapeutic management options in liver disease care continue to expand, such as for patients with decompensated cirrhosis and HCC. [12] Nationally, over 1 in 4 patients with CLD experiences financial hardship from medical bills, which is associated with increased cost-related medication nonadherence and health care resource utilization. [13] In addition, costrelated nonadherence has been shown to be associated with maladaptive cost-reducing behaviors. [14] Given the prevalence of financial hardship for those with CLD, practitioners should consider screening for financial burden with questions such as, "are you having difficult with paying for your medical care?" [7] If concerns are expressed, early social work and pharmacy support services should be engaged.
The development of a systematic approach to grade patient-level costs may guide the incorporation of C/V assessments into CGD, such as the use of value frameworks by national cardiology and oncology professional societies. Including granular price transparency data in CGD may also help clinicians to guide patients when choosing options that are feasible and effective based on potential financial constraints. For example, the American Diabetes Association CGD on pharmacologic management for glycemic treatment reports the median monthly costs for glucose-lowering agents. [15] For physicians, it is important to keep in mind that the cost of medications can also exacerbate disparities. Within the hepatology literature, researchers have found that rifaximin, an expensive nongeneric medication, has been prescribed less frequently to Black and Hispanic patients. Being mindful about how cost is incorporated into individual patient decision making is important going forward. [16] The main limitations of this study are the focus on CGD from US-based societies and the qualitative approach to the analysis. We acknowledge the complexity of integrating C/V statements into CGD due to limited high-quality evidence regarding value-based care in hepatology, shifting reimbursement structures, distinct value assessments between organizations and payors, and variations in market prices.
In summary, the proportion of C/V statements in AASLD, AGA, and ACG CGD related to liver disease is low and even lower for patient-level costs. Given the rising health care costs for liver disease, future studies are needed to better understand the financial burden, cost, and value of liver-related diagnostics and therapeutics. Now is the time for gastroenterology and hepatology professional societies to take on the challenge of integrating patient-level costs into guidance statements to improve the financial burden of patients with liver disease.
AUTHOR CONTRIBUTIONS E.A.: concept/design, acquisition of data, data analysis/ interpretation, statistical analysis, drafting article, critical revision of article, approval of article. A.K.: acquisition of data, data analysis/interpretation, statistical analysis, drafting article, critical revision of article, approval of article. N.N.U.: study supervision, concept/design, data