Psychosocial Risk Factors and Outcomes Among Men and Women With Alcohol‐Associated Liver Disease Undergoing Transplant Evaluation

ABSTRACT

Background

Women with alcohol‐associated liver disease (ALD) have been shown to be listed for transplant less frequently and have more psychiatric comorbidities as compared to men. This study aimed to analyze differences in psychosocial risk factors between men and women with ALD using the Stanford Integrated Psychosocial Assessment for Transplant (SIPAT), a pre‐transplantation tool that assesses a candidate's psychosocial risk factors. The secondary aim was to determine if SIPAT scores are associated with transplant‐related outcomes.

Methods

A single‐center, retrospective cohort study was conducted. All outpatients with ALD who were evaluated for liver transplant between 2015 and 2021 and who had at least one completed SIPAT evaluation were included. Fisher's exact and Mann–Whitney U tests were used to compare baseline characteristics and SIPAT scores. Multivariable logistic regression was used to analyze transplant outcomes. Survival differences were assessed using survival analysis, with comparisons evaluated using the log‐rank test.

Results

Total SIPAT scores were similar between men and women, but women were found to have significantly higher psychopathology domain scores, related to higher rates of depression and anxiety. Higher psychopathology domain scores and presence of anxiety were associated with lower odds of death among women only. There were no differences in waitlisting, transplantation, or death between men and women with ALD.

Conclusions

This study highlights gender differences in psychosocial risk factors among patients with ALD undergoing transplant evaluation. The mechanism by which anxiety is associated with lower odds of death among women is unclear, and it remains to be seen whether this finding would be reproduced in a larger, prospective study or would generalize to women with other etiologies of cirrhosis. Overall, these findings underscore the importance of incorporating gender‐sensitive, nuanced psychosocial assessment in the liver transplant evaluation process.

This study evaluated gender differences in psychosocial risk factors among patients with alcohol‐associated liver disease. Using the Stanford Integrated Psychosocial Assessment for Transplant (SIPAT), women demonstrated higher rates of depression and anxiety, yet total SIPAT scores were similar. Notably, anxiety among women was associated with lower odds of death.

1. Introduction

The prevalence of alcohol use disorder (AUD) in the United States is steadily rising, particularly among women. Accordingly, rates of alcohol‐related emergency department visits, hospitalizations, and deaths in the United States have increased over the last several decades. While men continue to outnumber women in all of these categories, the rates have surged for women over this time period, narrowing the gender gap (White 2020). The COVID‐19 pandemic has only served to exacerbate the disproportionate rise in alcohol use and its devastating consequences (Barbosa et al. 2021; Grossman et al. 2020).

Concurrently, a rise in the prevalence of alcohol‐associated liver disease (ALD) has been observed in the United States, again with steeper increases for women (Mellinger et al. 2018, 2019). Among general cirrhosis patients, multiple studies have demonstrated that men are more likely to receive a liver transplant than women (Allen et al. 2018; Nephew and Serper 2021). Women who are actively listed for transplant also have higher mortality rates than men (Allen et al. 2018). There are several reasons for this disparity, including differences in height and Model for End‐Stage Liver Disease (MELD) exception point distribution, as well as an underestimation of cirrhosis severity related to underestimation of kidney dysfunction among women (due to lower muscle mass and consequently lower creatinine values) (Allen et al. 2018; Shimizu et al. 2012).

The implementation of MELD 3.0 has sought to address the gender gap by explicitly incorporating female sex as a variable in the score calculation, as well as by adjusting the weighting of serum creatinine and adding serum albumin (Hudson et al. 2024; Kim et al. 2021). Early post‐implementation data in the United States have demonstrated a reduction in sex‐based disparities, with an increase in the proportion of female candidates and recipients and a reduction in differences in waitlist mortality (Akabane et al. 2025). However, residual disparity persists, particularly related to factors such as organ size matching and exception point allocation (Kwong et al. 2025).

Fewer studies have been conducted specifically comparing transplantation of men and women with ALD. A recent retrospective analysis of liver transplantation for ALD at a large, tertiary liver transplant center revealed that men are nearly twice as likely to be listed for transplant and, once listed, are significantly more likely to undergo transplantation than women (McElroy et al. 2020). Women notably had more psychiatric comorbidities and were less likely to be listed due to active alcohol and opioid use, leading the authors to conclude that psychosocial risk factors likely influenced listing and transplantation rates (McElroy et al. 2020).

Psychosocial assessment is an important component of the overall transplant evaluation process. A standardized approach to evaluate a patient's psychosocial factors is important to avoid subjectivity and bias. The Stanford Integrated Psychosocial Assessment for Transplant (SIPAT) is a pre‐transplantation assessment tool that was developed by Maldonado et al. (2012) to address the need for a more objective approach in identifying a candidate's areas of psychosocial vulnerability. With this tool, scores range from 0 to 110, with higher scores indicating a higher psychosocial risk. The ranges are categorized as follows: 0–6 (excellent), 7–20 (good), 21–39 (minimally acceptable), 40–69 (poor), and ≥ 70 (high risk). Scores are associated with both psychosocial and medical outcomes after transplantation (Maldonado et al. 2015). Important components of the evaluation include an assessment of social support, substance use history, comorbid psychiatric illnesses, the patient's understanding of the transplant process, and readiness to undergo transplant—all of which impact patient well‐being and graft survival after transplant (Matthews and Lucey 2022). However, it is important to note that our center does not have a “cutoff” for SIPAT score. This is one data point in the overall assessment of the patient's candidacy, and it is ultimately at the discretion of each transplant center and their multidisciplinary committee whether a patient is a candidate for transplant based on the broader evaluation of the patient (including their medical co‐morbidities).

While not all transplant centers in the United States utilize the SIPAT tool as part of their evaluation process, many centers do use some form of standardized assessment of psychosocial candidacy. Maldonado et al. (2012) demonstrated that the SIPAT compares favorably with the Psychosocial Assessment of Candidates for Transplantation (PACT), with high inter‐rater reliability and a strong association between SIPAT scores and post‐transplant psychosocial difficulties. In one study that evaluated a large population of patients being assessed for transplant, higher SIPAT scores were independently associated with decreased incidence of transplant listing as well as increased likelihood of immunosuppression nonadherence (Deutsch‐Link et al. 2021). It was demonstrated that patients with ALD had higher mean SIPAT scores compared with their non‐ALD counterparts, and composite SIPAT scores were lower in the patients selected for transplant versus non‐selected patients in both the ALD and non‐ALD groups (Daniel et al. 2022).

Overall, there are limited data regarding psychosocial risk factors among women with ALD. If women with ALD are generally more at risk for psychiatric comorbidities compared to their male counterparts, it would follow that this could affect their transplant candidacy. This, in turn, may drive some of the disparities that have been noted in transplant evaluation within this population. The goals of this study were to: (1) utilize SIPAT scoring to determine if there are differences in psychosocial risk factors between men and women with ALD presenting for liver transplant evaluation and (2) analyze if SIPAT scores are associated with transplant outcomes in these populations.

2. Materials and Methods

2.1. Study Population

A retrospective cohort study was conducted at a single transplant center. Approval was obtained from the center's institutional review board. All adult patients with ALD who were seen for an initial outpatient evaluation in the transplant hepatology clinic from January 2015 to October 2021 and had at least one SIPAT score were included.

Patients were excluded if they: (1) were evaluated urgently while in the hospital, (2) had another, separate outpatient transplant evaluation (with a SIPAT assessment) during the same period (Jan 2015 to Oct 2021), or (3) had a prior inpatient transplant evaluation. A flow diagram outlining patient selection methods is shown in Figure 1. Ultimately, a total of 447 patients were included in this study.

FIGURE 1.

Flowchart of patient inclusion and exclusion criteria.

Information regarding patient demographics, lab values, comorbid diseases, cirrhosis decompensations, evaluation dates, and clinical outcomes was obtained from the electronic medical record (EMR) and entered into our institutional REDcap database.

2.2. Definition of ALD

The determination of ALD as one of the etiologies of chronic liver disease was made based upon documentation in the transplant hepatology notes, not the ICD codes. Providers either noted that patients had “Alcoholic cirrhosis,” “Alcohol‐related cirrhosis,” “Alcoholic hepatitis,” or some variation on this terminology in the assessment section of their note. All patients with discrepancies regarding the etiology of liver disease were reviewed by one author (LC) for consistency. Patients with etiologies of liver disease in addition to alcohol were included, and the additional etiology was recorded.

2.3. Psychosocial Assessment

Since 2015, our institution has included the SIPAT tool in the multidisciplinary assessment of potential liver transplant candidates. This assessment was first introduced in 2012 by Maldonado et al. as a means to more objectively evaluate patients' psychosocial risks, which can significantly impact outcomes after transplant. The SIPAT is comprised of 4 domains: “Patient's Readiness Level,” “Social Support System,” “Psychological Stability & Psychopathology,” “Lifestyle & Effect of Substance Use.” Each of these domains includes multiple scored items. A score is calculated for each of the four domains, which are then added together to give the total SIPAT score.

The patient's score and identified areas of psychosocial risk are meant to serve as a guide in the overall evaluation of the patient. Our institution does not use a “cutoff” score above which a patient is declined for transplant. The psychosocial evaluation overall helps to identify areas of improvement that can potentially be addressed by the patient prior to a decision being made regarding transplant candidacy. In these cases, some patients at our institution undergo additional SIPAT evaluations.

For the purposes of this study, only the first SIPAT evaluation was used. The date of this evaluation, the 4 domain scores, and the total SIPAT score were recorded.

After the initial analysis, LC and MJ went back to the initial SIPAT evaluation to record individual item scores within the “Psychological Stability & Psychopathology” and the “Lifestyle & Effect of Substance Use” domains. The patient's age at onset of alcohol use, length of time since last alcohol use, prior history of alcohol treatment and alcohol relapse, use of other substances, history of treatment with psychiatric medication, and prior history of verbal/physical/sexual abuse inflicted upon them were all recorded from the social worker's SIPAT assessment as well.

2.4. Study Variables

2.4.1. Exposure Variables

For SIPAT outcomes, patient gender (man or woman) was the exposure variable that was assessed in relation to total and domain scores. Of note, no patients were identified in our study who had a gender identity that was different from their sex assigned at birth.

For transplant outcomes (waitlisted, transplanted, biopsy‐proven rejection, and death), the SIPAT total and domain scores were the exposure variables. For the purposes of this study, scores were recorded continuously (they were not grouped into “high” and “low” risk groups, for instance).

2.4.2. SIPAT Outcomes

One of the primary outcomes evaluated in this study was the total SIPAT score and the four SIPAT domain scores of men and women with ALD. These scores had been entered into the EMR by the social worker administering the test and were recorded by the authors in the institutional REDcap database.

2.4.3. Transplant Outcomes

Information regarding patients' listing status (listed, declined, still under evaluation, or listed then later removed), transplantation status, episodes of biopsy‐proven rejection, and the associated dates were recorded. The reasons for patients being declined for listing or being removed from the waitlist were also noted. Additionally, patients' statuses (alive, dead, or lost to follow‐up) at the time of data collection were recorded.

2.4.4. Covariates

Patient demographics (age, height, weight, etc.), lab values, calculated MELD‐Na scores, comorbid diseases, and presence of HCC were included as covariates.

2.4.5. Statistical Analysis

All primary analyses were conducted using SAS software version 9.4 (Cary, NC).

Fisher's exact and Mann–Whitney U tests were used to compare baseline characteristics and SIPAT scores. Multivariable logistic regression was used to analyze transplant outcomes.

Survival differences between men and women were assessed through survival analysis, with comparisons evaluated using the Log‐Rank test. This data analysis was conducted using R and RStudio. For this analysis, death was defined as all‐cause mortality. Cohort entry was defined as the date of the initial outpatient liver transplant evaluation. Censoring occurred at the time of the last‐known direct contact with the patient documented in the electronic medical record at the study institution or the defined study end date, whichever came first.

3. Results

3.1. Demographics

3.1.1. Baseline Clinical Characteristics

A total of 447 patients (338 men and 109 women) met the inclusion criteria for this study (Figure 1). Baseline characteristics are displayed in Table 1.

TABLE 1.

Baseline characteristics of men and women with ALD presenting for outpatient transplant evaluation.

	Men (n = 338)	Women (n = 109)	p
Age at evaluation, years	54 (48–60)	54 (49–60)	0.813
BMI at evaluation	28.7 (25.2–32.9)	25.5 (22.3–29.8)	< 0.0001***
Ethnicity/race
White	301 (89.1%)	101 (92.7%)	0.36
All other races/ethnicities	37 (11.0%)	8 (7.3%)
Highest level of education completed
Less than high school degree	69 (20.5%)	15 (13.8%)	0.161
High school degree	190 (56.4%)	61 (56.0%)
College degree or higher	78 (23.1%)	33 (30.3%)
Employment status
Employed	70 (20.7%)	11 (10.1%)	0.005***
Unemployed	55 (16.3%)	32 (29.4%)
Retired	40 (11.8%)	15 (13.8%)
Disabled	173 (51.2%)	51 (46.8%)
Marital status
Married or partnered	168 (49.7%)	47 (43.1%)	0.27
Not married/unpartnered	170 (50.3%)	62 (56.9%)
Charlson Comorbidity Index	5.0 ± 1.7	4.7 ± 1.3	0.139
MELD‐Na at baseline	18 (13–23)	20 (16–24)	0.028***
Other etiologies	115 (34.0%)	20 (18.4%)	0.002***
Known HCC	58 (17.2%)	9 (8.3%)	0.030***

Note: Data presented as number (percentage), median (interquartile range), or mean ± SD. *** indicates significance at the p < 0.05 level.

Men and women had several significant baseline differences, including: BMI (median of 28.7 for men vs. 25.5 for women, p < 0.0001), employment status (20.7% of men employed vs. 10.1% of women, p = 0.005), MELD‐Na (median of 18 for men vs. 20 for women, p = 0.028) contributing etiologies of cirrhosis (34% of men vs. 18.4% of women, p = 0.002), and HCC prevalence (17.2% of men vs. 8.3% of women, p = 0.030).

3.1.2. Baseline Alcohol Use and Treatment History

Women were significantly more likely to have used alcohol within 6 months prior to their transplant evaluation (Table 2). Of the patients that reported their age at first use of alcohol (n = 363), men were significantly more likely to report having started drinking before age 20. More men reported being told by a doctor to stop drinking alcohol prior to their transplant evaluation, although this was not statistically significant (76.1% vs. 66.3%, p = 0.068). There were no significant differences in the percentage of men and women who reported prior treatment/counseling for AUD.

TABLE 2.

Baseline alcohol use and treatment history among men and women.

	Men, n (%)	Women, n (%)	p
Reported last alcohol use prior to evaluation a
≤ 6 months ago	86 (25.6)	41 (37.6)	0.02 b
Age at first alcohol use c
< 20 years old	231 (84.3)	60 (67.4)	< 0.0001 b
20–29 years old	37 (13.5)	15 (16.9)
30–39 years old	2 (0.7)	8 (9.0)
40–49 years old	4 (1.5)	6 (6.7)
Doctor recommended abstinence d	235 (76.1)	67 (66.3)	0.068 e
Prior alcohol treatment f	157 (47.7)	47 (44.8)	0.654

^a Data missing for 2 patients in all (2 men and 0 women).

^b Indicates significant at the p < 0.05 level.

^c Data missing for 84 patients in all (64 men and 20 women).

^d Data missing for 37 patients in all (29 men and 8 women).

^e Indicates a non‐significant difference, p < 0.10.

^f Data missing for 13 patients in all (9 men and 4 women).

3.1.3. Baseline Substance Use History

There was no significant difference in the smoking status of men and women at the time of transplant evaluation (current, former, or never smoker. See Table S1). There were also no significant differences between men and women regarding their use of cannabis, cocaine, or unhealthy use of other substances (including stimulants, heroin, hallucinogens, and prescription drugs).

3.2. SIPAT Scores

There was no significant difference in the total SIPAT scores for men and women (median of 32 for both, p = 0.333). There were also no significant differences in the readiness, social support, and lifestyle domain scores. However, women had a significantly higher psychopathology domain score, with a median of 5 compared to 3 for men (Table 3).

TABLE 3.

Median total and domain SIPAT scores of men and women.

	Men (n = 338)	Women (n = 109)	p
Total SIPAT score	32 (24–39)	32 (24–41)	0.333
Readiness domain score	8 (6–12)	9 (6–12)	0.797
Social support domain score	5 (4–7)	5 (4–7)	0.442
Psychopathology domain score	3 (0–7)	5 (3–10)	< 0.0001***
Lifestyle domain score	12 (9–16)	12 (8–15)	0.559

Note: Data provided as median (interquartile range). *** indicates significant at the p < 0.05 level.

3.2.1. Lifestyle Domain

Further analysis was done to determine if there were differences in unhealthy substance use patterns, as reflected by the SIPAT evaluation, despite a similar overall lifestyle domain score. No significant differences in scores were found between men and women on the five items included in the lifestyle domain (Table 4). There were more women that were assessed to be at higher risk of alcohol relapse, though this did not reach statistical significance (p = 0.051). Likewise, more men were deemed to be at higher risk of substance use relapse, but this also did not reach statistical significance (p = 0.076).

TABLE 4.

Scores for men and women on individual questions in the lifestyle domain of the SIPAT evaluation.

	Men (n = 338)	Women (n = 109)	p
Alcohol use history	6 (4–6)	6 (4–6)	0.210
Risk of alcohol relapse	2 (1–3)	2 (2–3)	0.051
Substance use history	2 (0–2)	2 (0–2)	0.143
Risk of substance use relapse	1 (0–1)	0 (0–1)	0.076
Nicotine use history	1 (0–5)	1 (0–5)	0.439

Note: Results displayed as median (IQR).

3.2.2. Psychopathology Domain

Within the psychopathology domain of the SIPAT, 75% or more of the answers to several of the individual items were “0” for both men and women (See Table S2), meaning that these items did not often contribute to the overall psychopathology domain score. Among the items assessing history of anxiety and depression, women had significantly higher prevalence and severity of both, as compared to men (Table 5).

TABLE 5.

Comparison of prevalence and severity of anxiety and depression among patients with ALD undergoing liver transplant workup.

	Men (n = 338)	Women (n = 109)	p
Depression (SIPAT IXa)
None	219 (64.8)	51 (46.8)	0.002***
Mild	97 (28.7)	43 (39.5)
Moderate	22 (6.5)	15 (13.8)
Anxiety (SIPAT IXb)
None	255 (75.4)	53 (48.6)	< 0.0001***
Mild	60 (17.8)	36 (33.0)
Moderate	21 (6.2)	20 (18.4)
Severe	2 (0.6)	0 (0)

Note: Results displayed as number (percentage). *** indicates significance at the p < 0.05 level.

3.3. Transplant‐Related Outcomes

3.3.1. Listing

There were a total of 194 patients (43.4%) listed for transplant, 218 (48.8%) declined for listing, and 35 (7.8%) still had open evaluations at the end of the study period. There were no significant differences in these transplant evaluation outcomes when comparing men and women (Table 6).

TABLE 6.

Transplant listing outcomes among men and women, as well as outcomes after listing for transplant.

	Total (n = 447)	Men (n = 338)	Women (n = 109)	p
Outcome of evaluation
Declined	218 (48.8)	165 (48.8)	53 (48.6)	0.595
Listed	194 (43.4)	144 (42.6)	50 (45.9)
Evaluation still open	35 (7.8)	29 (8.6)	6 (5.5)
	Total (n = 194)	Men (n = 144)	Women (n = 50)	p
Outcome after listing
Transplanted	118 (60.8)	86 (59.7) a	32 (64.0)	0.312
Removed from waitlist	60 (30.9)	48 (33.3)	12 (24.0)
Awaiting transplant	16 (8.2)	10 (6.9)	6 (12.0)

Note: Data presented as number (percentage).

^a Of the 86 men transplanted, 84 were transplanted at our center, and 2 were transplanted at another center.

3.3.2. Transplantation

Among the 194 listed patients, 118 patients (60.8%) were transplanted (Table 6). Among the 86 men that were transplanted, 2 patients were transplanted at an outside transplant center. Of the 76 patients not transplanted by the end of the study period, 60 (30.9% of listed patients) had been removed from the transplant waitlist (“delisted”) and 16 were still actively listed for transplant. There were no significant differences in the outcomes of listed men and women (p = 0.312).

3.3.3. Reasons for Declining Patients for Listing

Among the 218 patients who were declined for transplant listing, 35 patients (16.1%) had more than one reason for being declined. The most frequently given reason for men to be declined was that they were “Too well” (23.6%), while the most common reason for women to be declined was due to “Comorbidities and/or Too sick” (22.6%) (Figure 2). “Declined transplant and/or lost to follow‐up” was the only reason that was significantly different between men and women (20.0% of men vs. 7.6% of women, p = 0.04). When combining the categories of “Comorbidities and/or Too sick” and “Died” into a single category, women were more likely to fall into this category compared to men, although this did not reach statistical significance (43.4% vs. 30.9%, p = 0.10) (Figure S1).

FIGURE 2.

Comparison of the reasons that men and women were declined for transplant. *** indicates significance at the p < 0.05 level.

3.3.4. Reasons for Removal From Waitlist

There were a total of 48 men and 12 women who were removed from the transplant waitlist. There were no significant differences in the reasons that men and women were removed from the waitlist. The most frequently given reason for delisting (Figure S2) was the same for women and men: they were “Too well” for transplant (66.7% of delisted women and 35.4% of delisted men). Though not statistically significant, women were removed more frequently for this reason (p = 0.099).

3.3.5. Biopsy‐Proven Rejection

Of the 84 men transplanted at our center, 10 (11.9%) had at least one episode of biopsy‐proven rejection during the study period. Among the 32 transplanted women, there were five (15.6%) with at least one episode of biopsy proven rejection (p = 0.554). The median time between transplant and rejection among the men was 174 days (range 7–757). Among the women, the median was 578 days (8–837).

3.3.6. Death (All‐Cause Mortality)

Of the 338 total men evaluated for liver transplant, 70 (20.7%) had died by the end of the study period, compared to 20 of 109 women (18.3%). This difference was not statistically significant (p = 0.681). Using all‐cause mortality as the end point, survival analysis also demonstrated no difference in the overall survival probability at any time point in the study (Figure 3).

FIGURE 3.

Kaplan–Meier curve with risk table. There was no significant difference in the overall survival probability of men and women undergoing transplant evaluation.

3.4. Associations Between SIPAT Scores and Transplant Outcomes

3.4.1. Listing

Higher total SIPAT scores were significantly associated with lower odds of listing, with an adjusted odds ratio (aOR) of 0.920 for men (Table S3) and 0.905 for women (Table S4). Higher scores in each SIPAT domain were also significantly associated with lower odds of listing for both men and women (Tables S3 and S4), except for the psychopathology domain score among women. However, after controlling for age, marital status, presence of hepatocellular carcinoma, and the MELD‐Na score at the time of evaluation, higher psychopathology domain scores were associated with lower odds of listing among women as well (Table S4).

3.4.2. Transplantation

On univariable analysis, higher total SIPAT and lifestyle domain scores were significantly associated with lower odds of transplantation among listed men (Table S5). However, on multivariable analysis, these variables were no longer significant.

There were no significant associations between SIPAT scores and transplantation among the listed women.

3.4.3. Death (All‐Cause Mortality)

Among men presenting to the transplant clinic for evaluation, a higher readiness score (i.e., not being ready for transplant) was associated with higher odds of death (all‐cause mortality) on univariable analysis, though this was not significant in our model (OR 1.061, 95% CI: 0.998–1.128). This difference was statistically significant on multivariable analysis (aOR 1.079, 95% CI 1.005–1.158) (Table S7). There were no other significant associations between the other domain scores and the total SIPAT score and the odds of death among men.

Among women, only the psychopathology domain score was significantly associated with death (Table S8). This finding persisted on multivariable analysis: women with higher psychopathology scores had a lower odds of death overall (aOR 0.805, 95% CI: 0.685–0.945).

Sensitivity analysis was conducted, in which multiple covariates were included in the model, to see if any variables other than the psychopathology domain score could explain the decreased odds ratio. The results seen in Figure 4 illustrate that additional covariates do not significantly modify the impact of the psychopathology score, and they do not significantly improve the model. Models including one or two additional covariates do not substantially outperform the univariate model (AIC = 100.9). A reduction of ≥ 2 in AIC would indicate a significantly better model; no combination achieved this.

FIGURE 4.

Sensitivity analysis of univariate model (relationship between psychopathology domain score and odds of death among women).

On further analysis, presence of depression was not associated with death for either men or women (Tables S9 and S10). However, presence of anxiety was significantly associated with lower odds of death among women, with an odds ratio (OR) of 0.334 (95% CI: 0.118–0.950). After controlling for other variables, anxiety remained significantly associated with lower odds of death among women, with an adjusted OR of 0.231 (95% CI: 0.068–0.783).

4. Discussion

The overall purpose of this single‐center, retrospective study is to determine if men and women with ALD undergoing outpatient liver transplant evaluation have different psychosocial risk factors, as measured by the SIPAT. Total SIPAT scores and 3 of the 4 domain scores were similar between men and women, but women had significantly higher psychopathology domain scores, indicating more psychopathologic risk factors.

Additionally, this study aimed to identify differences in transplant‐related outcomes among men and women with ALD. There were no differences in transplant waitlisting, removal from the waitlist, transplantation, or all‐cause mortality. However, further analysis of the reasons for declining a patient for transplant revealed that women were more likely to be “too sick” or to die during the evaluation process (though this difference did not reach statistical significance). Quite the opposite was found, however, with regard to why patients who were waitlisted were subsequently removed from the waitlist; though this difference also did not reach statistical significance, women were more likely to be removed from the waitlist for being “too well” to warrant a transplant compared to men (66.7% vs. 35.4%, p = 0.099).

Perhaps the most notable finding is that a higher psychopathology domain score among women was associated with lower overall odds of death. This association was not observed in men. Upon further investigation, anxiety seemed to be the driver of this lower odds of death, not depression. This difference persisted in the sensitivity analysis.

Transplant psychosocial screening scales can identify psychosocial burden and may predict post‐transplant outcomes. For example, the Psychosocial Assessment of Candidates for Transplantation (PACT) consists of an 8‐item rating scale that can also capture the clinician's assessment of a patient's unhealthy substance use, compliance, and coping strategies. A lower score (between 0 and 2) can signal issues such as limited social support, psychological instability, and increased risk for psychopathology. For female liver recipients, lower PACT scores and ALD have been associated with a greater risk of post‐transplant mortality, with prior relapse, ongoing alcohol consumption, and lack of engagement in AUD treatment resulting in a higher likelihood of relapse after liver transplantation (Deutsch‐Link et al. 2020; Schneekloth et al. 2019). Additionally, depression and anxiety are noted to be significant risk factors for morbidity and mortality after transplant (Dew et al. 2015).

The findings in this study demonstrate that, when utilizing the SIPAT as an assessment tool, women with ALD have higher psychopathology domain scores compared to men with ALD (median of 5 vs. 3; p < 0.0001) (Table 3). On further analysis, this is secondary to higher levels of anxiety and depression in women (Table 5). This is consistent with prior studies that have assessed depression and anxiety among patients undergoing transplant evaluation for heart, kidney, and hematopoietic stem cell transplant (Courtillié et al. 2023; Loh et al. 2020; Tecchio et al. 2013).

This study also demonstrates that women with higher psychopathology scores have a lower odds of death (aOR 0.805, 95% CI: 0.685–0.945) (Table S8). Interestingly, it is anxiety, not depression, that is associated with a lower odds of death among women (aOR 0.231, 95% CI: 0.068–0.783) (Table S10). There is some evidence in the literature to suggest that anxiety could play a role in enhancing certain positive behaviors both pre‐ and post‐transplant and, when managed effectively, can contribute to increased vigilance and adherence to medical advice and medication adherence (Courtwright et al. 2016; Dew et al. 2018; Wessels‐Bakker et al. 2022). There is no prior data regarding anxiety and transplant outcomes specifically in the liver transplant literature.

Surprisingly, this study demonstrated that the lifestyle domain scores were not significantly different between men and women, with similar patterns of comorbid drug and tobacco use (Table S1). This runs counter to several studies demonstrating that men are significantly more likely to smoke or have coexisting substance use disorders compared to women (Martin et al. 2014; McElroy et al. 2020).

With regard to gender differences in transplant outcomes, there is overwhelming evidence in the literature that women are both less likely to get listed for and receive liver transplantation overall. In general, women often have lower MELD scores due to lower serum creatinine from less muscle mass for the same degree of renal dysfunction as their male counterparts (Listabarth et al. 2022; Singh et al. 2022). The most recent OPTN/SRTR 2023 Annual Data Report demonstrates that, despite the introduction of MELD 3.0, pretransplant mortality remains higher and transplant rates are lower for women compared to men (Kwong et al. 2025). Singh et al. (2022) proposes that differences in listing disparities among all‐comers with cirrhosis might be explained by the difficulty that women have with adopting the sick role and the lack of adequate support from their partners.

Among patients with ALD, men are 95% more likely to be listed and 105% more likely to receive a liver transplant relative to women (McElroy et al. 2020). A study by Kaplan et al. (2023) used metrics including the listing‐to‐death (LDR) ratio, which divides the number of patients listed for transplant by those who died from end‐stage liver disease (ESLD), and the transplant‐to‐death ratio, which divides the number of patients who underwent liver transplant by the number of patients who died of ESLD, to explore racial and sex disparities in the liver transplantation process. They found that women with ALD demonstrated a lower LDR as compared to men and a lower transplant‐to‐death ratio as compared to men (Kaplan et al. 2023). These findings suggest that fewer women with ALD are being listed and receiving transplants relative to overall deaths from ALD. There are several theories to explain this discrepancy, including increased stigma preventing women from seeking transplant evaluation, as well as differences in alcohol metabolism affecting the timing and severity of presentation (Kaplan et al. 2023).

For patients with ALD who were included in this study, there were no significant differences in listing, transplant, and survival outcomes between men and women (see Table 6 and Figure 3). Our institution does not currently have specific strategies in place to mitigate or minimize gender disparities in transplant and survival. These results may be more reflective of an underpowered study to highlight those differences, with a total of 118 patients who were transplanted during the study period.

While there were no significant differences overall in the listing of men and women in this study, a pattern emerged related to reasons that men and women were declined for listing, with more women under transplant evaluation declined due to being “too sick” or dying (43.4% vs. 30.9%; p = 0.10). Women had significantly higher MELD scores compared to men when presenting for transplant evaluation (Table 1), and it is suspected that this difference in severity of illness might explain why women were more likely to be declined for being too sick or dying. It is possible that factors like increased stigma among women with ALD, as proposed by Kaplan et al. (2023), and delayed presentation explain why women in this study presented for evaluation with higher MELD scores. That being said, some of the difference in native MELD scores in this study are likely related to the higher percentage of men with HCC compared to women (17.2% of men vs. 8.3% of women, Table 1), which may contribute to the men looking “less sick” than the women at the time of evaluation.

Another possible explanation for these observed differences relates to the concept of “telescoping” in AUD. Telescoping refers to the accelerated trajectory from initiation of alcohol use to the development of dependence, medical complications, and adverse outcomes that has been consistently described in women compared to men. Prior studies have shown that women often begin drinking later in life but progress more quickly to problematic use and liver disease, with shorter intervals between first alcohol use, diagnosis of AUD, and cirrhosis (Johnson et al. 2005; Piazza et al. 1989; Randall et al. 1999; Towers et al. 2022). This accelerated course may explain why, in our cohort, women were more likely to report alcohol use within 6 months of evaluation compared to men. These findings suggest that telescoping may contribute to women presenting for transplant evaluation at a later stage of illness, with more recent alcohol use, thereby influencing candidacy decisions and potentially explaining why women were more likely to be declined for being “too sick” or to die during evaluation.

The present study also demonstrated a pattern of women being removed from the transplant waitlist due to being “too well” as compared to men, although this was not statistically significant (66.7% of delisted women, 35.4% of delisted men; p = 0.099) (Figure S2). Similarly, Loy et al. (2018) showed a significant difference between the removal from the transplant waitlist due to improvement in the health of women with metabolic dysfunction‐associated steatohepatitis (MASH) cirrhosis compared to men. Furthermore, Pose et al. (2021) found that female sex was independently associated with removal from the waitlist among patients with ALD due to clinical improvement, suggesting that less access to liver transplant (based on MELD and height discrepancies) might allow those with lower MELD scores more time to re‐compensate and subsequently be removed from the waitlist.

Another study by Giard et al. (2019) found that women had a 24% higher likelihood of being removed from the waitlist for being too well but also a 9% higher risk of being removed for death or being too sick. They proposed that unmeasured factors (such as obesity, diabetes, caffeine use, and adherence) might explain women being more likely to show clinical improvement. The present study additionally found that women were more likely to be evaluated for transplant with ≤ 6 months of sobriety compared to men, so it is possible that, as women remained on the waitlist and had a longer period of sobriety, they may have regained sufficient liver function to warrant removal from the waitlist.

In contrast, there are also many studies demonstrating that women are more likely to be removed from the waitlist due to death or being too sick. In a study by Locke et al. (2020) evaluating patients on the waitlist for liver transplantation, women with any form of cirrhosis were 8.6% more likely to die while on the waiting list compared to men. Another study demonstrated that after the introduction of the MELD score, women were more likely than men to die or become too ill while awaiting a liver transplant (Moylan et al. 2008). Specifically regarding ALD, Cullaro et al. (2018) demonstrated that women are 10% more likely to be delisted because of clinical deterioration. They suggest the perception of frailty, longer duration on the waitlist due to lower MELD scores, and higher likelihood of declining a transplant offer related to small stature as possible explanations for this discrepancy, overall calling for a more objective and systematic approach to determining sickness and subsequent removal from the transplant waitlist.

Limitations of this study include its retrospective nature and the associated risk of introducing bias into the selection of patients. This study was also conducted at one transplant center, and the number of patients, particularly women, was overall low. Therefore, the study was not powered to detect differences in some of the outcomes. We also acknowledge that given this is a single‐center, retrospective study, it may be more difficult to generalize listing outcomes as they can be subjective, center‐specific, and exhibit wide variability across centers. A multi‐center study of this patient population would be much more informative from this perspective.

Another limitation is the fact that the patients included in the study population were those presenting for initial evaluation between the years of 2015 and 2021. There was a significant shift in the approach to transplantation for patients with ALD in this time period and afterwards; however, much of this change has occurred with respect to the timing of transplant for patients with acute alcohol‐associated hepatitis (AH). By limiting the study's population to patients presenting in the outpatient setting, we believe the findings in this study are still relevant to current clinical practice. However, this in itself is also a limitation of the study since most patients with severe AH were not included in the analysis (because these patients generally present for urgent inpatient evaluations). As such, the results of this study are unlikely to apply to patients presenting with AH. Additionally, MELD scores in this study were lower than those reported in similar studies (Daniel et al. 2022). The psychosocial risk by gender is likely different depending on the severity of disease, which was not captured in this cohort.

Finally, the SIPAT assessment itself is not completely objective and, though the members of the transplant team administering the SIPAT were trained in how to administer it, the possibility of interobserver differences cannot be excluded. It is also noted that the items assessing anxiety and depression included within the SIPAT are not validated tools for the diagnosis of either condition. Unfortunately, most patients did not have concurrent, validated assessments of anxiety and depression (such as PHQ‐9 or GAD‐7). Future studies could be designed to collect this data in a prospective manner alongside SIPAT assessments for further evaluation of the findings of this study. Much of the data included in this study also relied on self‐reporting (including the quantity of alcohol use and time since last use) from the patients themselves, who may have felt that they could not tell the entire truth in the context of wanting to be seen in a positive light during the transplant selection process. Phosphatidylethanol (PEth) testing was not readily available and was not checked for most patients at the study institution until 2019, so the truthfulness of patient self‐reporting was not able to be corroborated with biomarkers, which may have affected results.

In conclusion, using the widely available pre‐transplant SIPAT tool, women with ALD were found to have significantly higher psychopathology domain scores (indicating more psychopathologic risk factors) compared to men with ALD undergoing transplant evaluation. Specifically, women were found to have higher prevalence and severity of both anxiety and depression. There were no overall differences between men and women with regard to transplant waitlisting, removal from the waitlist, transplantation, or death in this study. However, higher psychopathology domain scores and presence of anxiety were associated with lower odds of death among women only.

While the presence of more psychosocial risk factors has generally been thought to have a negative impact on transplant candidacy and outcomes, this study suggests that anxiety may be protective for women with ALD. If this is the case, adjustments to psychosocial evaluations may need to be made to account for this gender difference. Further studies are needed to corroborate whether anxiety among women with ALD is associated with better outcomes, whether this can be observed in other populations with cirrhosis, and to determine the mechanism by which anxiety and/or treatment of anxiety contributes to improved outcomes.

Funding

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Table S1: Comparison of the smoking status and history of marijuana, cocaine, and other substance use among men and women under evaluation for liver transplant.

Table S2: Scores for men and women on individual questions in the psychopathology domain of the SIPAT evaluation. *** indicates significance at the p < 0.05 level.

Table S3: Relationship between total and individual SIPAT domain scores and odds of transplant listing among men with ALD. *** indicates significance at the p < 0.05 level.

Table S4: Relationship between total and individual SIPAT domain scores and odds of transplant listing among women with ALD. *** indicates significance at the p < 0.05 level.

Table S5: Relationship between total and individual SIPAT domain scores and odds of transplantation among men with ALD. *** indicates significance at the p < 0.05 level.

Table S6: Relationship between total and individual SIPAT domain scores and odds of transplantation among women with ALD.

Table S7: Relationship between total and individual SIPAT domain scores and odds of death among men with ALD. *** indicates significance at the p < 0.05 level.

Table S8: Relationship between total and individual SIPAT domain scores and odds of death among women with ALD. *** indicates significance at the p < 0.05 level.

Table S9: Relationship between anxiety and depression and odds of death among men. *** indicates significance at the p < 0.05 level.

Table S10: Relationship between anxiety and depression and odds of death among women. *** indicates significance at the p < 0.05 level.

Figure S1: Comparison of men and women who were not listed due to severity of illness or death.

Figure S2: Comparison of the reasons that men and women were removed from the transplant waitlist.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.