Influence of Aspirin on Hospital and Clinical Outcomes in Hepatocellular Carcinoma: Insights from National Data

Abstract

Background: Hepatocellular carcinoma (HCC) is a major global health burden and a leading cause of cancer-related deaths. While aspirin has shown potential chemopreventive effects in chronic liver disease, its impact on clinical outcomes among patients hospitalized with HCC remains under-investigated. Methods: Using the National Inpatient Sample (NIS) from 2016 to 2022, we conducted a retrospective cohort study to evaluate the association between aspirin use and clinical outcomes in adult HCC hospitalizations. Patients were stratified based on documented aspirin use, and propensity score matching with inverse probability of treatment weighting (IPTW) was applied to minimize confounding. The primary outcome was in-hospital mortality; secondary outcomes included morbidity-related complications, hospital length of stay, and total charges. Results: Among 337,730 hospitalizations with HCC, 8.37% involved aspirin use. Aspirin users demonstrated significantly lower in-hospital mortality (5.2% vs. 10.09%), with an adjusted odds ratio (OR) of 0.58 (95% CI: 0.50–0.68; p < 0.001). Aspirin use was also associated with shorter hospital stays (5.42 vs. 6.39 days), lower total charges ($80,310 vs. $95,098), and reduced incidence of complications, including acute liver failure, hepatic encephalopathy, ascites, spontaneous bacterial peritonitis, sepsis, ICU admission, and acute kidney injury. Importantly, no statistically significant increase in gastrointestinal or variceal bleeding was observed among aspirin users. Conclusions: These findings suggest that aspirin use may reduce mortality, morbidity, and healthcare burden in patients hospitalized with HCC.

1. Introduction

Hepatocellular carcinoma (HCC) is the most common primary liver malignancy, predominantly arising from hepatocytes, the liver’s main functional cells. It ranks as the sixth most common cancer globally and the third leading cause of cancer-related deaths, following lung and stomach cancers [1]. HCC carries a particularly poor prognosis due to its aggressive nature, late-stage detection, and limited therapeutic options in advanced stages. Its development is closely associated with chronic liver conditions, including chronic hepatitis B and C virus (HBV and HCV) infections, cirrhosis, and non-alcoholic fatty liver disease (NAFLD). Chronic inflammation and ongoing hepatocellular injury caused by these diseases lead to fibrosis and ultimately cirrhosis, forming the biological substrate for hepatocarcinogenesis. HCC prevalence is notably high in regions with elevated viral hepatitis rates, such as East Asia and Sub-Saharan Africa [2].

Although HCC incidence is declining in high-endemic regions due to improved antiviral therapies, it paradoxically continues to rise in other areas, including India, Europe, the United States, and Oceania. This trend is largely attributed to the growing global burden of metabolic syndrome, type 2 diabetes mellitus, and obesity, all of which predispose individuals to NAFLD, non-alcoholic steatohepatitis (NASH), and eventually HCC [3]. This epidemiological shift highlights the need for preventive and adjunctive therapeutic approaches that can mitigate the increasing morbidity and mortality associated with HCC in non-viral populations.

Emerging evidence suggests aspirin, a commonly used anti-inflammatory and antiplatelet medication, may offer protective effects against liver disease progression and HCC development. This protective role is thought to involve multiple biological pathways, including inhibition of cyclooxygenase-2 (COX-2), attenuation of liver fibrosis, suppression of pro-inflammatory lipid mediators, reduction in platelet-derived growth factor signaling, and modulation of immune and platelet activity [4,5]. In preclinical models, aspirin has demonstrated inhibition of hepatic stellate cell activation and reduced fibrogenesis—key processes implicated in the progression to cirrhosis and HCC. Several observational studies have reported a duration-dependent relationship between low-dose aspirin use and decreased HCC incidence among patients with chronic liver disease [4,5,6,7,8,9]. Notably, these protective benefits appear more pronounced in patients with viral hepatitis, though emerging data also support a beneficial role in non-viral etiologies.

While numerous studies have explored aspirin’s chemopreventive effects in reducing HCC incidence, fewer have assessed its impact on clinical outcomes in patients already diagnosed with HCC. A limited number of studies suggest aspirin use may be associated with reduced all-cause mortality in patients with established HCC [10,11,12,13,14,15].

However, many previous studies investigating aspirin’s impact on HCC outcomes were limited by methodological constraints that restricted their broader applicability and clinical relevance. These studies typically focused only on cirrhotic populations, did not evaluate healthcare resource utilization, and inadequately controlled for confounding factors such as concurrent non-aspirin NSAID use [13]. Although our study similarly lacks data regarding concurrent use of other chemopreventive medications such as statins or metformin, it explicitly addresses other significant gaps by including both cirrhotic and non-cirrhotic HCC patients, systematically excluding non-aspirin NSAID users, and utilizing robust statistical techniques like propensity matching and inverse probability weighting to effectively control for measurable confounders. Additionally, we assessed both clinical outcomes and healthcare burden, providing a more comprehensive view of aspirin’s potential therapeutic utility.

Given aspirin’s potential as an inexpensive and widely accessible adjunctive therapy, its role beyond prevention of HCC incidence warrants further investigation. Aspirin’s potential as an additional therapeutic option for HCC is gaining increasing attention, especially considering its antithrombotic and anti-inflammatory properties.

The present study evaluates the impact of aspirin use on in-hospital mortality, morbidity-related complications, healthcare resource utilization, and total hospitalization costs among patients admitted with HCC at a national level.

2. Materials and Methods

2.1. Database

We conducted a retrospective analysis using data from the National Inpatient Sample (NIS) for the years 2016 to 2022. The NIS is the largest publicly available all-payer inpatient healthcare database in the United States. It provides nationally representative data from approximately 20% of hospital discharges, allowing for weighted estimates of inpatient healthcare utilization, cost, demographics, insurance coverage, and clinical outcomes. The NIS is maintained by the Agency for Healthcare Research and Quality (AHRQ) as part of the Healthcare Cost and Utilization Project (HCUP). The database captures a wide range of diagnoses and procedures using standardized ICD-10-CM codes, facilitating large-scale epidemiologic and outcome-based studies across a diverse patient population. Each year’s dataset contains roughly 7 million unweighted hospitalizations, which can be weighted to generate national estimates exceeding 35 million discharges annually.

2.2. Study Population and Variables

Adult hospitalizations (age ≥ 18 years) with a primary or secondary diagnosis of hepatocellular carcinoma were identified using the International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) codes. These admissions were then stratified based on documented aspirin use. Patients were included regardless of the indication for aspirin, but were excluded if they were concurrently using non-aspirin NSAIDs to minimize confounding from other anti-inflammatory agents.

The primary outcome was in-hospital mortality. Secondary outcomes included hospital resource utilization (length of stay and total hospital charges), as well as HCC-related complications such as acute liver failure, hepatic encephalopathy, ascites, spontaneous bacterial peritonitis, variceal bleeding, gastrointestinal bleeding, portal vein thrombosis, and obstructive jaundice. These complications were identified using validated ICD-10 codes. Patients with missing demographic or mortality data were excluded from the final analytic sample to ensure robustness in statistical modeling and outcome assessment. Demographic variables included age, sex, race/ethnicity, income quartile by ZIP code, insurance status, and hospital region.

2.3. Sensitivity Analysis

We conducted multiple sensitivity analyses to assess the robustness of the observed associations. E-values were calculated to estimate the minimum strength of unmeasured confounding that would be required to explain away the observed effects of aspirin on each clinical outcome. For instance, the observed association between aspirin use and in-hospital mortality (OR: 0.58, 95% CI: 0.50–0.68) had an E-value of 2.83, suggesting that substantial unmeasured confounding would be needed to nullify this effect. Similar robustness was observed across other key outcomes, including acute liver failure (E-value: 2.44), hepatic encephalopathy (E-value: 2.80), and others. These results support the conclusion that the observed associations are unlikely to be fully explained by residual confounding.

2.4. Statistical Analysis

Descriptive statistics were used to summarize patient and hospital characteristics. Categorical variables were compared using the chi-square test, and continuous variables were compared using the t-test. To reduce confounding bias, we employed 1:1 nearest-neighbor propensity score matching (PSM) with a caliper of 0.2 standard deviations to ensure close matches between aspirin and non-aspirin users. Propensity scores were generated using logistic regression models based on clinically relevant demographic and clinical covariates (Figure A1).

In addition to matching, inverse probability of treatment weighting (IPTW) was applied to account for residual differences in baseline characteristics. The covariates used for propensity scoring and weighting included obesity, substance use (cannabis, alcohol, opioids, smoking), dyslipidemia, hypertension, diabetes mellitus, chronic kidney disease, liver disease severity, congestive heart failure, coronary artery disease, peripheral vascular disease, NSAID use, and key hospital characteristics (bed size, teaching status, urban vs. rural location, and geographic region).

Covariate balance after matching and weighting was assessed using standardized mean differences (SMD), with all covariates demonstrating adequate balance (SMD < 0.1). The final matched cohort included 7910 patients.

Multivariate logistic regression was used to estimate adjusted odds ratios (ORs) for the association between aspirin use and the specified outcomes in the matched cohort, adjusting for all potential confounders. All statistical analyses were performed using STATA/MP version 17.0 (Stata Corp, College Station, TX, USA). A p-value < 0.05 was considered statistically significant.

3. Results

3.1. Patient and Hospital Characteristics

A total of 337,730 hospitalizations with a diagnosis of hepatocellular carcinoma (HCC) were included in the final analysis. Of these, 8.79% involved patients with documented aspirin use, while most hospitalizations 91.20% did not report aspirin use. The mean age of patients differed slightly between the two groups, with aspirin users being older on average (69.84 years) compared to non-users (65.30 years); however, this difference was not statistically significant.

Sex distribution showed a statistically significant difference between groups (p < 0.001). A higher proportion of males was observed among aspirin users (78.70%) compared to non-users (74.07%). Racial distribution also differed significantly across the two cohorts (p < 0.001). Aspirin users were more likely to be White (62.07% vs. 54.89%), while Hispanic patients (12.23% vs. 18.42%) and other races (8.28% vs. 11.89%) were more commonly represented among non-users. Black patients made up a similar, though slightly higher, proportion of aspirin users (17.42%) compared to non-users (14.80%).

Socioeconomic status, as assessed by the quartile distribution of median household income based on zip code, did not show any significant differences between the two groups (p = 0.892). The distribution across income quartiles was relatively uniform, with no apparent socioeconomic gradient in aspirin use among hospitalized HCC patients.

Significant disparities were observed in terms of insurance coverage (p < 0.001). Medicare was the predominant primary payer for aspirin users (68.28%) compared to non-users (53.62%), which may reflect the older average age in the aspirin group. Conversely, Medicaid coverage (11.21% vs. 19.07%) and private insurance (15.74% vs. 20.34%) were more prevalent among non-users, possibly indicating a younger and more economically diverse patient population.

Hospital-level characteristics revealed additional significant differences. A slightly higher proportion of aspirin users were admitted to urban teaching hospitals (84.10%) compared to non-users (82.56%), with a p-value of 0.004. This may suggest a trend toward more complex or academic care settings among aspirin users, though the absolute difference was modest. Hospital size did not significantly differ between groups (p = 0.464), with most patients in both cohorts admitted to large hospitals.

Regional distribution of hospitalizations showed a statistically significant variation (p < 0.001). Aspirin users were more frequently hospitalized in the Midwest (24.42% vs. 17.79%), while a higher proportion of non-users were admitted in the West (24.88% vs. 20.18%) and South (37.95% vs. 36.72%). The Northeast had comparable representation in both groups (18.68% among aspirin users vs. 19.39% in non-users). The demographics are represented in

Table 1. Socio-demographics and hospital characteristics of HCC hospitalizations with and without aspirin use.

Baseline Characteristics		HCC Hospitalizations with Aspirin Use (%)	HCC Hospitalizations Without Aspirin Use (%)	p Value
Age (in years)		69.84	65.30	-
Sex	Male	78.70	74.07	<0.001
	Female	21.30	25.93
Race	White	62.07	54.89	<0.001
	Black	17.42	14.80
	Hispanic	12.23	18.42
	Others	8.28	11.89
Quartile of median household income for zip code	0−25th	32.31	32.23	0.892
	26th−50th	24.87	25.24
	51st−75th	23.03	23.09
	76th−100th	19.79	19.43
Primary payer	Medicare	68.28	53.62	<0.001
	Medicaid	11.21	19.07
	Private	15.74	20.34
	Others	4.76	6.97
Hospital teaching status and location	Rural	3.90	3.78	0.004
	Urban non-teaching	12	13.66
	Urban teaching	84.10	82.56
Hospital bed-size	Small	15.70	15.08	0.464
	Medium	24.55	24.38
	Large	59.74	60.54
Hospital region	North-east	18.68	19.39	<0.001
	Mid-west	24.42	17.79
	South	36.72	37.95
	West	20.18	24.88

.

3.2. In Hospital Mortality and Morbidity Outcomes

Among hospitalizations with HCC, aspirin use was associated with significantly improved clinical outcomes. Specifically, patients with documented aspirin use had a lower in-hospital mortality rate of 5.2%, compared to 10.09% among those who did not use aspirin. After adjusting for potential confounding variables through multivariate logistic regression and propensity score matching, aspirin use remained independently associated with reduced mortality, with an adjusted odds ratio (aOR) of 0.58 (95% CI: 0.50–0.68, p < 0.001). Kaplan–Meier survival curves further illustrated a significant survival benefit in aspirin users, suggesting a consistent association between aspirin use and favorable in-hospital survival in the HCC population (Figure 1).

In addition to lower mortality, aspirin use was also associated with reduced healthcare resource utilization. Patients who used aspirin had a shorter mean length of hospital stay (5.42 days vs. 6.39 days; β coefficient = −0.83, 95% CI: −0.98 to −0.68; p < 0.001) and incurred lower total hospital charges ($80,310 vs. $95,098; β coefficient = −$6330, 95% CI: −$9797 to −$2863; p < 0.001).

Patients with aspirin use also had a significantly lower incidence of morbidity complications of HCC, such as acute liver failure (4.0%) compared to non-users (7.39%), with an aOR of 0.65 [0.55–0.78, p < 0.001]. Similarly, the incidence of hepatic encephalopathy was lower among aspirin users (1.17%) than in non-users (2.61%), with an aOR of 0.59 [0.43–0.80, p = 0.003]. Aspirin use was also associated with a notable reduction in the incidence of ascites, 28.0% with aspirin use vs. 42.87% without aspirin use (aOR = 0.68, p < 0.001), as well as spontaneous bacterial peritonitis, which had an aOR of 0.58 [0.47–0.71, p < 0.001].

Additional favorable outcomes were observed in systemic complications. Rates of sepsis were significantly reduced among aspirin users (9.39% vs. 14.09%; aOR = 0.70, 95% CI: 0.62–0.79, p < 0.001). ICU admission rates were also lower in this group (3.55% vs. 6.45%; aOR = 0.62, 95% CI: 0.52–0.75, p < 0.001), as were occurrences of acute kidney injury (27.46% vs. 31.75%; aOR = 0.81, 95% CI: 0.74–0.88, p < 0.001), indicating a broader systemic benefit associated with aspirin use during hospitalization.

We also observed a decrease in the incidence of gastrointestinal bleeding among aspirin users (3.46%) compared to non-users (2.59%). While this result was statistically significant in the multivariate logistic regression, with an aOR 1.09 (CI 1.01–1.19, p = 0.045), the significance was lost in our propensity matched model. [adjusted OR of 1.04 (0.84–1.24, p = 0.569)].

Similarly, no statistically significant difference was observed in the incidence of variceal bleeding (OR = 0.82, p = 0.075), obstructive jaundice, (OR = 1.09, p = 0.468) and portal vein thrombosis (OR of 0.89, p = 0.066) among HCC hospitalizations with and without aspirin use.

Overall, these findings suggest that aspirin use is associated with reduced mortality, lower healthcare utilization, and a decreased risk of multiple complications in patients hospitalized with HCC, and the associations were statistically significant (

Table 2. Multivariate logistic regression analysis showing the odds ratio of mortality outcomes of HCC hospitalizations stratified by aspirin use.

	HCC Hospitalizations with Aspirin Use	HCC Hospitalizations Without Aspirin Use	Multivariate Adjusted Odds Ratio [OR] */Coefficient	95% Confidence Interval	p Value
Mortality	5.2%	10.09%	0.58 **	0.50–0.68	<0.001
Length of stay	5.42 days	6.39 days	−0.83	(−0.98)–(−0.68)	<0.001
Total hospital charge	80,310 $	95,098 $	−6330	(−9797)–(−2863)	<0.001

* OR < 1 means aspirin users had lower odds of the outcome. Statistically significant protective associations indicated by p-value < 0.05. ** Propensity matching done between confounders.

and

Table 3. Comparison of the influence of Aspirin use on outcomes of hospitalized patients with Hepatocellular carcinoma.

Outcomes of Hepatocellular Carcinoma	With Aspirin Use (%)	Without Aspirin Use (%)	Multivariate Adjusted Odds Ratio [aOR] *	p-Value
Acute liver failure	4.0	7.39	0.65 [0.55–0.78]	<0.001
Hepatic encephalopathy	1.17	2.61	0.59 [0.43–0.80]	0.003
Variceal Bleeding	2.77	4.90	0.82 [0.66–1.02]	0.075
Portal vein thrombosis	10.51	14.45	0.89 [0.79–1.01]	0.066
Ascites	28.0	42.87	0.68 [0.64–0.72]	<0.001
Spontaneous bacterial peritonitis	1.81	4.27	0.58 [0.47–0.71]	<0.001
Gastrointestinal Bleeding	3.46	2.59	1.04 [0.84–1.24]	0.569
Obstructive Jaundice	2.94	3.01	1.09 [0.87–1.36]	0.468
Sepsis	9.39	14.09	0.70 [0.62–0.79]	<0.001
ICU Admission	3.55	6.45	0.62 [0.52–0.75]	<0.001
Acute Kidney Injury	27.46	31.75	0.81 [0.74–0.88]	<0.001

* Interpretation: aOR < 1 means aspirin users had lower odds of the outcome. Statistically significant protective associations (p < 0.05) are bolded. No outcome showed a statistically significant increase in risk.

).

4. Discussion

The effect of aspirin use on reducing the incidence of HCC has been established through prior retrospective analyses, particularly among patients with chronic liver disease [5]. In this nationwide population analysis, we examined hospitalizations for HCC to assess the outcomes associated with aspirin use. Our findings demonstrated that aspirin use was associated with a lower risk of all-cause mortality, reduced healthcare resource utilization, including shorter hospital stays, and lower total hospitalization costs among patients with HCC. These observations align with recent meta-analytic evidence indicating that aspirin use is associated with reduced risks of both HCC incidence and mortality. For instance, a systematic review and meta-analysis reported a pooled hazard ratio (HR) of 0.72 (95% CI 0.60–0.87) for HCC mortality among aspirin users [10]. Additionally, a retrospective cohort study from Sweden found that low-dose aspirin was associated with a 31% reduction in the risk of HCC (adjusted HR 0.69, 95% CI 0.62–0.76) [5]. Similarly, a recent prospective study by Aktan et al. demonstrated that aspirin use independently improved overall survival in HCC patients, reinforcing its potential prognostic benefit beyond the inpatient setting [16].

Beyond mortality and cost outcomes, we investigated the impact of aspirin on various morbidity-related complications associated with HCC. Aspirin use was linked to reduced risks of multiple complications, including acute liver failure, hepatic encephalopathy, variceal bleeding, portal vein thrombosis, and ascites. To our knowledge, studies directly linking aspirin use to reductions in these specific complications, particularly among hospitalized populations, remain limited. Nonetheless, our findings support the hypothesis that aspirin’s anti-inflammatory and antithrombotic properties may confer broader hepatic protective effects extending beyond tumor suppression alone.

Interestingly, our analysis also showed that aspirin use among hospitalized HCC patients was associated with a lower incidence of gastrointestinal (GI) bleeding, although this difference did not achieve statistical significance in the propensity-matched model. This observation is particularly noteworthy because aspirin, as an antiplatelet agent, is typically expected to increase GI bleeding risks. However, several observational studies have found no significant increase in bleeding risk among chronic liver disease patients using aspirin. For instance, a large Swedish nationwide cohort study reported no significant difference in the 10-year cumulative incidence of GI bleeding between aspirin users and nonusers (7.8% vs. 6.9%; difference, 0.9 percentage points; 95% CI, −0.6 to 2.4) [5]. Similarly, Lee et al. observed no statistically significant increase in GI bleeding in aspirin users with chronic hepatitis B (adjusted subhazard ratio [sHR], 1.11; 95% CI, 0.87–1.42; p = 0.391) [8].

Conversely, some studies have reported modest increases in GI bleeding risks associated with aspirin. A meta-analysis by Li et al. identified an elevated but statistically nonsignificant risk of major bleeding in HCC patients receiving aspirin (HR, 1.14; 95% CI, 0.99–1.31) [11]. Nevertheless, the overall balance of evidence suggests that in carefully selected patients—particularly those without active varices or severe coagulopathy—the benefits of aspirin therapy likely outweigh its bleeding risks. Our findings reinforce this perspective and advocate for careful patient stratification when initiating aspirin therapy.

Aspirin’s role in reducing liver-related complications and slowing disease progression is increasingly attributed to its diverse mechanisms of action. Primarily, aspirin inhibits cyclooxygenase-2 (COX-2), an enzyme highly expressed in inflammatory cells and hepatic tumor tissues [10,17,18,19,20]. COX-2 facilitates angiogenesis and tumor progression and is associated with poorer prognosis in HCC patients [21]. By suppressing COX-2 and related prostaglandin pathways, aspirin can hinder tumor vascularization and limit growth. Additionally, aspirin’s inhibition of platelet activation and thromboxane A2 production further reduces systemic and hepatic inflammation and fibrosis—central processes in hepatocarcinogenesis, particularly in HBV-related liver disease [10,22,23]. In summary, aspirin’s biological plausibility in protective effects is strongly supported by its anti-inflammatory, antiplatelet, and antifibrotic mechanisms (Figure 2).

Moreover, aspirin may reduce the incidence and severity of cirrhosis-related complications by targeting the hypercoagulable state observed in chronic liver disease. Although traditionally associated with coagulopathy, cirrhosis involves complex hemostatic imbalances predisposing patients to thrombosis. Increased thrombin generation, platelet activation (evidenced by elevated 11-dehydro-thromboxane B2, p-selectin expression, and von Willebrand factor interactions), and microvascular thrombosis contribute significantly to portal hypertension and variceal bleeding [12,24,25,26,27,28,29]. By inhibiting these pathways, aspirin may mitigate portal vein thrombosis and other vascular complications, ultimately contributing to improved prognosis and reduced disease decompensation in HCC patients [30].

Overall, aspirin’s combined anti-inflammatory, antiplatelet, and antifibrotic properties provide substantial protective benefits in preventing and managing HCC, particularly in patients with chronic liver diseases such as HBV infection and cirrhosis [10,17,30].

It is important to recognize that, given the observational nature and inherent limitations of the NIS database, our findings should be interpreted as associations rather than causal relationships. Specifically, the database lacks details of pre-admission medication use and the precise sequence of clinical events during hospitalization. Therefore, while aspirin use was associated with improved outcomes, definitive conclusions regarding temporal or causal relationships between aspirin exposure and reduced complications cannot be drawn.

Despite these limitations, this study strongly suggests aspirin use may offer significant clinical benefits for patients with HCC, including reduced in-hospital mortality, fewer hepatic complications, and decreased healthcare resource utilization. These findings highlight aspirin’s potential as a valuable adjunctive therapy in HCC management, particularly among patients with concurrent cardiovascular conditions already warranting aspirin therapy. Given aspirin’s low cost and widespread availability, it could represent a feasible intervention to enhance patient outcomes, especially in resource-limited settings. However, prospective randomized controlled trials are necessary to confirm these findings and clarify optimal dosing, duration, and patient selection. Moving forward, such trials would provide more definitive evidence and help refine clinical guidelines for aspirin use in HCC.

5. Conclusions

Aspirin use in patients hospitalized with hepatocellular carcinoma is associated with reduced in-hospital mortality, fewer liver-related complications, and lower healthcare burden. These findings highlight its potential role as an adjunctive therapy in HCC management. Further prospective studies are needed to confirm these associations and guide clinical decision-making regarding aspirin use in this population.

6. Limitations

We acknowledge several limitations to our study. The retrospective nature of our study, conducted using the NIS coding database, limits the availability of detailed data such as medication usage and laboratory results, which could influence the outcomes studied. As a result, it was not feasible to assess the severity of the HCC outcomes. Additionally, because the NIS database primarily includes data from hospitalized patients, it may not fully represent the entire HCC patient population, as many cases are managed in outpatient settings unless hospitalization is required due to complications. Consequently, generalizing our findings to the broader population may be challenging. Furthermore, as aspirin is not prescribed specifically for the indication of HCC, its use in these patients is usually secondary to primary or secondary prevention for cardiovascular comorbidities. Although we adjusted for these confounders, this could still affect the study results. Another limitation is that we did not distinguish the dose of aspirin used among these hospitalizations and were unable to completely capture over-the-counter use or adherence to aspirin.

Moreover, the lack of granular data on aspirin dosage, frequency, and duration introduces the possibility of exposure misclassification. Since the NIS only captures documented in-hospital medication use, we could not differentiate between chronic, intermittent, or incidental aspirin exposure. As a result, patients classified as aspirin users may have heterogeneous patterns of prior use, which may attenuate or obscure the true effect estimates. This limitation should be considered when interpreting the strength and consistency of associations observed.

Nevertheless, our use of propensity score matching and inverse probability weighting substantially mitigates the impact of measured confounders, enhancing confidence in our associative findings.

Despite these limitations, our study’s robust findings raise significant questions regarding the chemoprotective effect of aspirin on mortality and morbidity outcomes in HCC hospitalizations.