Abstract
Background:
In this study, we evaluate the clinical impact of psychiatric illnesses (PI) on the hospital outcomes of patients admitted with alcoholic liver disease (ALD).
Methods:
From the National Inpatient Sample from 2012–2017, patients with alcoholic cirrhosis or alcoholic hepatitis were selected and stratified using the presence/absence of PI (which was a composite of psychiatric conditions). The cases were propensity score-matched to PI-absent controls and were compared to the following endpoints: mortality, death due to suicide, length of stay (LOS), hospitalization charges, and hepatic complications.
Results:
After matching, there were 122907 PI with and 122907 without PI. Those with PI were younger (51.8 vs. 51.9 years p=0.02) and more likely to be female (39.2 vs. 38.7% p=0.01); however, there was no difference in race. Patients with PI had lower rates of alcoholic cirrhosis but higher rates of alcoholic hepatitis/alcoholic hepatic steatosis. In multivariate, patients with PI had lower rates of all-cause mortality (aOR 0.51 95%CI 0.49–0.54); however, they experienced higher rates of deaths due to suicide (aOR 3.00 95%CI 1.56–5.78) and had longer LOS (aOR 1.02 95%CI 1.01–1.02).
Conclusion:
Presence of PI in ALD patients is associated with prolonged hospital stay and higher rates deaths due to suicide.
Keywords: Alcoholic hepatitis, PI, depression, anxiety, schizophrenia, bipolar disorder
1. Introduction
Patients with alcoholic liver disease (ALD) have a high prevalence of coexisting PI that could potentially make appropriate care during hospitalization challenging. Previous studies have found a high prevalence of co-occurring psychiatric conditions in patients with ALD, precluding these patients from establishing healthy rapport and cohesive relationship with healthcare providers (1–3). Furthermore, these studies have found an increased incidence of psychiatric disorders becoming more accentuated (4), as the incidence of drug use and other illicit substance use becomes more prevalent (5). This rise in psychiatric illnesses (PI) amongst those with ALD leads to significant healthcare burden in the US medical system (6), with significant associated costs accruing in the past several years per analysis trends (7). Thus, from a national scale, it is important to elucidate the clinical characteristics and implications of PI on hospital outcomes of patients with ALD, given that these patients may be at higher risk of admissions due to self-harm or suicidal ideations and suffer prolonged hospitalization with extended discharge plans that require of complex medical and psychiatric care (8).
In order to study the specific relationship between ALD and PI (9), we use a nationwide US hospital database to cross-sectionally evaluate the impact of major PI on hospital outcomes and prognosis of patients with ALD.
2. Methods
2.1. Database
We used the National Inpatient Sample (NIS) database from years 2012–2017. Each NIS compiles data from statewide inpatient databases (SID), composed of the claims data collected from hospitals in predesignated states. The database itself is funded by the Healthcare Cost and Utilization Project (HCUP) from the Agency for Healthcare Research and Quality (10, 11). We sampled data from the years 2012–2017 from the SID databases. (12). Our sampling methods of data from SID minimized sampling heterogeneity (13). International Classification of Diseases (ICD) 9 or ICD 10 codes were used to encode the diagnoses on discharge depending on whether the discharge was in the pre- or post-3rd fiscal cycle in 2015 respectively (14, 15). As this database contains discharge information that predates/includes 2015, variable selection process was homogenized through the use of a cross-referencing search program developed using the GEMS-base, which is an official converter of two ICD-systems developed by CMS (16, 17). This cross-linking tool was used to convert ICD codes between the two ICD dictionaries and thus capture homogeneous definitions from the two ICD systems (18–20).
2.2. Missing information
For missing data, multiple imputation using multiple imputation with chained equations (MICE) was used to replace data that was missing. This method has been validated in previous studies to be effective for imputing missing data in large-database driven studies (21–23).
2.3. Comparative statistics
We used the encoded ICDs corresponding to the composite ALD variable to isolate the ALD population (which included of alcoholic cirrhosis and alcoholic hepatitis). From this, those under 18 years of age were excluded. The isolated population as stratified using the composite exposure variable PI, which included the psychiatric conditions enlisted in Supplementary Table 2 (depression, anxiety, adjustment disorder, schizophrenia, psychosis, hallucination, bipolar disorder, OCD, panic disorder, phobia, PTSD). The endpoints of the study were segregated into twofold outcomes; the primary: mortality, length of stay, and charges; the secondary: outcomes related to hepatic decompensation including ascites, varices, sepsis, variceal bleeding, spontaneous bacterial peritonitis, hepatorenal syndrome, encephalopathy, and acute liver failure.
Prior to the comparative analytics, the study population was 1:1 matched to the PI-absent controls using nearest neighbor propensity score matching method. The covariates that were incorporated to the propensity score-generating model included: age, gender, race, diabetes, hyperlipidemia, hypertension, COPD, coronary artery disease, chronic kidney disease, congestive heart failure, coagulopathy, and smoking. As the match employed a 1:1 match, the post-match sample sizes were equivalent between the stratified populations. Following the match, individual parametricity of the variables were determined using the Jarque–Bera test (24, 25). Depending on the parametricity of the variable distribution, the appropriate nominal (Chi-square or Fisher’s Exact test) or non-nominal testing (Student’s t-test or Whitney-Mann U test) was employed to compare the variables between the cases versus the controls. In Multivariate analysis, regression was performed using the match-exclusionary variables for the heterogenic hospital characteristics. For all statistical iterations, p-value less than 0.05 denoted significance. Crude and adjusted odds ratios were derived for nominal comparisons with corresponding confidence intervals.
3. Results
3.1. Patient Selection
Figure 1 demonstrates the patient selection process. Patients with ALD were included in this study, and were stratified by the presence of PI. Post-matching, there were a total of 245814 patients with ALD, and this population was stratified into 122907 with PI and 122907 patients without PI.
Figure 1.
Figure 1 denotes the patient selection procedure of the study.
3.2. Demographics and covariates
Table 1 demonstrates demographic and covariate-associated data. After matching and weighting, the cohort with psychiatric disorders was found to be younger on average (51.8 vs. 51.9 years, p=0.02) and have a higher proportion of females (39.2 vs. 38.7% p=0.01) than the cohort without psychiatric disorders. There were no significant differences in racial distribution between the two cohorts. The incidence of obesity (5.82 vs. 4.74%; p<0.001) were higher in the psychiatric group, but there were no differences in the incidence of other comorbidities. In terms of liver associated etiologies, the psychiatric cohort had lower rates of alcoholic cirrhosis (69.70 vs. 77.80%, p<0.001) and hepatocellular carcinoma (1.72 vs. 1.43%, p<0.001). However, they had higher rates of alcoholic hepatitis (39.60 vs. 32.50%, p<0.001), alcoholic hepatic steatosis (2.29 vs.1.89%, p<0.001), hepatitis C (7.70 vs. 6.37%, p<0.001), and nonalcoholic fatty liver disease (3.59 vs. 2.89%, p<0.001).
Table 1:
Pre- and Post-match Comparisons of Demographics and Medical Comorbidities in Alcoholic Liver Disease Patients With or Without Psychiatric Disorders
| Pre-Match Patients With Psychiatric Disorders vs Without Psychiatric Disorders |
Post-Match Patients With Psychiatric Disorders vs Without Psychiatric Disorders |
|||||
|---|---|---|---|---|---|---|
| Patients With Psychiatric Disorders |
Patients Without Psychiatric Disorders |
Patients With Psychiatric Disorders |
Patients Without Psychiatric Disorders |
|||
| Demographics | n = 122907 (29.21 %) | n = 297818 (70.79 %) | p-value | n = 122907 (50.00 %) | n = 122907 (50.00 %) | p-value |
|
| ||||||
| Age (year) | 51.80 year | 55.00 year | < 0.001 | 51.80 year | 51.90 year | 0.020 |
| Female (%) | 39.20 % | 25.00 % | < 0.001 | 39.20 % | 38.70 % | 0.011 |
| Race | < 0.001 | 0.500 | ||||
| White (%) | 76.50 % | 64.40 % | 76.50 % | 76.40 % | ||
| Black (%) | 8.19 % | 11.30 % | 8.19 % | 8.33 % | ||
| Hispanic (%) | 10.60 % | 17.70 % | 10.60 % | 10.40 % | ||
| Asian or Pacific Islander (%) | 0.78 % | 1.24 % | 0.78 % | 0.78 % | ||
| Native American (%) | 1.65 % | 2.34 % | 1.65 % | 1.69 % | ||
| Other (%) | 2.29 % | 2.98 % | 2.29 % | 2.32 % | ||
| Comorbidities | ||||||
|
| ||||||
| Diabetes (%) | 19.00 % | 21.80 % | < 0.001 | 19.00 % | 18.90 % | 0.790 |
| Hyperlipidemia (%) | 15.10 % | 12.30 % | < 0.001 | 15.10 % | 14.80 % | 0.061 |
| Hypertension (%) | 42.70 % | 34.20 % | < 0.001 | 42.70 % | 42.40 % | 0.091 |
| Chronic Obstructive Pulmonary Disease (%) | 17.10 % | 14.60 % | < 0.001 | 17.10 % | 17.20 % | 0.650 |
| Coronary Artery Disease (%) | 7.64 % | 8.88 % | < 0.001 | 7.64 % | 7.82 % | 0.086 |
| Chronic Kidney Disease (%) | 9.19 % | 14.60 % | < 0.001 | 9.19 % | 9.28 % | 0.460 |
| Congestive Heart Failure (%) | 7.94 % | 10.80 % | < 0.001 | 7.94 % | 8.07 % | 0.230 |
| Coagulopathies (%) | 5.99 % | 8.62 % | < 0.001 | 5.99 % | 6.11 % | 0.200 |
| Smoking (%) | 55.20 % | 47.10 % | < 0.001 | 55.20 % | 55.00 % | 0.280 |
| Obesity (%) | 5.82 % | 4.54 % | < 0.001 | 5.82 % | 4.74 % | < 0.001 |
| Alcohol-related Liver Diseases | ||||||
|
| ||||||
| Alcoholic Hepatitis (%) | 39.60 % | 26.60 % | < 0.001 | 39.60 % | 32.50 % | < 0.001 |
| Alcoholic Cirrhosis (%) | 69.70 % | 82.60 % | < 0.001 | 69.70 % | 77.80 % | < 0.001 |
| Alcoholic Hepatic Steatosis (%) | 2.29 % | 1.45 % | < 0.001 | 2.29 % | 1.89 % | < 0.001 |
| Other Alcohol-Related Liver Diseases (%) | 4.59 % | 4.75 % | 0.024 | 4.59 % | 4.77 % | 0.038 |
| Other Liver etiologies | ||||||
|
| ||||||
| Cirrhosis (%) | 70.50 % | 83.20 % | < 0.001 | 70.50 % | 78.50 % | < 0.001 |
| Hepatocellular Carcinoma (%) | 1.72 % | 3.37 % | < 0.001 | 1.72 % | 2.43 % | < 0.001 |
| Cholangiocarcinoma (%) | 0.05 % | 0.10 % | < 0.001 | 0.05 % | 0.07 % | 0.092 |
| Hepatitis B (%) | 0.56 % | 0.58 % | 0.510 | 0.56 % | 0.53 % | 0.380 |
| Hepatitis C (%) | 7.70 % | 6.89 % | < 0.001 | 7.70 % | 6.37 % | < 0.001 |
| Non-alcoholic Fatty Liver Disease (%) | 3.59 % | 2.33 % | < 0.001 | 3.59 % | 2.89 % | < 0.001 |
3.3. Comparison of socioeconomic status and hospital characteristics
Table 2 demonstrates the comparison of socioeconomic status and hospital characteristics. In terms of hospital characteristics, patients in the psychiatric disorders group were more likely to be in large sized beds and to be admitted into Midwest urban teaching hospitals. In terms of insurance, there was a greater proportion of patients in the psychiatric disorders group that had Medicare or Medicaid. In addition, the income distribution differed between the two groups, with the psychiatric cohort having a higher proportion of patients in the third- and fourth-income quartiles.
Table 2:
Pre- and Post-match Comparisons of Socioeconomic Status and Hospital Characteristics in Alcoholic Liver Disease Patients With or Without Psychiatric Disorders
| Pre-Match Patients With Psychiatric Disorders vs Without Psychiatric Disorders |
Post-Match Patients With Psychiatric Disorders vs Without Psychiatric Disorders |
|||||
|---|---|---|---|---|---|---|
| Patients With Psychiatric Disorders |
Patients Without Psychiatric Disorders |
Patients With Psychiatric Disorders |
Patients Without Psychiatric Disorders |
|||
| Socioeconomic Status/ Hospital Characteristics | n = 122907 (29.21 %) | n = 297818 (70.79 %) | p-value | n = 122907 (50.00 %) | n = 122907 (50.00 %) | p-value |
|
| ||||||
| Median Household Income | < 0.001 | < 0.001 | ||||
| Quartile 1 (Lowest) (%) | 30.20 % | 34.20 % | 30.20 % | 31.80 % | ||
| Quartile 2 (%) | 25.90 % | 25.80 % | 25.90 % | 26.20 % | ||
| Quartile 3 (%) | 24.30 % | 22.70 % | 24.30 % | 23.60 % | ||
| Quartile 4 (Highest) (%) | 19.60 % | 17.30 % | 19.60 % | 18.40 % | ||
| Hospital Bed Size | < 0.001 | 0.020 | ||||
| Small (%) | 16.90 % | 16.10 % | 16.90 % | 17.00 % | ||
| Medium (%) | 28.70 % | 28.60 % | 28.70 % | 29.10 % | ||
| Large (%) | 54.40 % | 55.20 % | 54.40 % | 53.90 % | ||
| Hospital Location/Teaching Status | < 0.001 | < 0.001 | ||||
| Rural (%) | 7.72 % | 7.63 % | 7.72 % | 8.12 % | ||
| Urban Nonteaching (%) | 29.30 % | 29.90 % | 29.30 % | 30.30 % | ||
| Urban teaching (%) | 62.90 % | 62.40 % | 62.90 % | 61.50 % | ||
| Hospital Region | < 0.001 | < 0.001 | ||||
| Northeast (%) | 20.00 % | 17.60 % | 20.00 % | 20.30 % | ||
| Midwest (%) | 24.40 % | 18.90 % | 24.40 % | 22.00 % | ||
| South (%) | 32.60 % | 36.90 % | 32.60 % | 36.10 % | ||
| West (%) | 23.00 % | 26.60 % | 23.00 % | 21.50 % | ||
| Insurance Type | < 0.001 | < 0.001 | ||||
| Medicare (%) | 30.00 % | 30.90 % | 30.00 % | 25.20 % | ||
| Medicaid (%) | 33.10 % | 30.70 % | 33.10 % | 32.80 % | ||
| Private Insurance (%) | 23.00 % | 21.60 % | 23.00 % | 24.00 % | ||
| Self-Pay (%) | 8.96 % | 11.30 % | 8.96 % | 12.50 % | ||
| No Charge (%) | 0.94 % | 1.12 % | 0.94 % | 1.32 % | ||
| Other (%) | 4.04 % | 4.31 % | 4.04 % | 4.21 % | ||
3.4. Comparison of hospital outcomes
Table 3 compares the hospital outcomes. Mortality was lower in the psychiatric cohort compared to the non-psychiatric cohort (3.13% vs. 5.89% p<0.001, OR 0.52 95% CI 0.50–0.54). The number of patients, however, that died from suicide was significantly higher in the psychiatric group compared to the group without psychiatric diagnoses (0.03% vs. 0.01% p<0.001, OR 3.00, 95% CI 1.56–5.77). Patients with PI also had longer hospital length of stay (6.03 days vs. 5.92 days; p<0.001, aOR 1.017 95% CI 1.014–1.021), but lower hospitalization charges ($48, 655 vs. $56, 344, p<0.001, aOR 0.857 95% CI 0.857–0.857). In addition, most liver-related complications, including sepsis (7.15 vs. 10.40%; p<0.001), spontaneous bacterial peritonitis (2.20 vs. 3.30%; p<0.001), ascites (14.20 vs. 20.90%; p<0.001), encephalopathy (15.10 vs. 17.50%; p<0.001), variceal bleeding (3.94 vs. 6.24%; p<0.001), varices (15.50 vs. 20.00%; p<0.001), and hepatorenal syndrome (3.00% vs. 4.69%; p<0.001) were lower in the psychiatric group. Multivariate analysis showed patients in the psychiatric cohort had significantly lower mortality (p<0.001, aOR 0.514, 95%,CI 0.494–0.535), higher deaths due to suicide (p<0.001, aOR 3.003, 95% CI 1.562–5.775), longer length of stay (p<0.001, aOR 1.017, 95% CI 1.014–1.021), and lower hospitalization charges (p<0.001, aOR 0.857, 95% CI 0.857–0.857). Figures 2 and 3 shows the forest plot representations of the multivariate models with all-cause inpatient deaths and suicide-related deaths as the outcome variables.
Table 3:
Pre- and Post-match Comparisons of Hospital Outcomes in Alcoholic Liver Disease Patients With or Without Psychiatric Disorders
|
Pre-Match Patients With Psychiatric Disorders vs Without Psychiatric Disorders
|
||||||||
| Patients With Psychiatric Disorders | Patients Without Psychiatric Disorders | Univariate Analysis | ||||||
|
|
||||||||
| Hospital Outcomes | n = 122907 (29.21 %) | n = 297818 (70.79 %) | p-value | OR | 95% CI | |||
|
|
||||||||
| Mortality (%) | 3.13 % | 6.91 % | < 0.001 | 0.43 | (0.42 – 0.45) | |||
| Death Due to Suicide (%) | 0.03 % | 0.01 % | < 0.001 | 3.23 | (1.96 – 5.32) | |||
| Length of Stay (days) | 6.03 days | 6.15 days | < 0.001 | |||||
| Hospitalization Cost ($) | 48,655 $ | 61,117 $ | < 0.001 | |||||
| Disposition at Discharge | < 0.001 | |||||||
| Routine (%) | 62.30 % | 60.30 % | ||||||
| Short-term Hospital (%) | 2.82 % | 3.28 % | ||||||
| SNF or other facility (%) | 17.40 % | 14.70 % | ||||||
| Home Health Care (%) | 10.20 % | 11.10 % | ||||||
| AMA (%) | 4.17 % | 3.65 % | ||||||
| Died (%) | 3.13 % | 6.91 % | ||||||
| Unknown (%) | 0.03 % | 0.06 % | ||||||
| Liver complications | ||||||||
|
| ||||||||
| Ascites | 14.20 % | 22.30 % | < 0.001 | 0.58 | (0.57 – 0.59) | |||
| Varices | 15.50 % | 21.60 % | < 0.001 | 0.67 | (0.66 – 0.68) | |||
| Sepsis | 7.15 % | 11.50 % | < 0.001 | 0.60 | (0.58 – 0.61) | |||
| Variceal Bleeding | 3.94 % | 6.85 % | < 0.001 | 0.56 | (0.54 – 0.58) | |||
| Spontaneous Bacterial Peritonitis | 2.20 % | 3.65 % | < 0.001 | 0.59 | (0.57 – 0.62) | |||
| Hepatorenal syndrome | 3.00 % | 5.44 % | < 0.001 | 0.54 | (0.52 – 0.56) | |||
| Encephalopathy | 15.10 % | 19.10 % | < 0.001 | 0.75 | (0.74 – 0.76) | |||
| Acute Liver failure | 2.61 % | 3.70 % | < 0.001 | 0.70 | (0.67 – 0.72) | |||
|
Post-Match Patients With Psychiatric Disorders vs Without Psychiatric Disorders
|
||||||||
| Patients With Psychiatric Disorders | Patients Without Psychiatric Disorders | Univariate Analysis | Multivariate Analysis | |||||
|
|
|
|||||||
| Hospital Outcomes | n = 122907 (50.00 %) | n = 122907 (50.00 %) | p-value | OR | 95% CI | aOR | 95% CI | p-value |
|
| ||||||||
| Mortality (%) | 3.13 % | 5.89 % | < 0.001 | 0.52 | (0.50 – 0.54) | 0.514 | (0.494 – 0.535) | < 0.001 |
| Death Due to Suicide (%) | 0.03 % | 0.01 % | < 0.001 | 3.00 | (1.56 – 5.77) | 3.003 | (1.562 – 5.775) | < 0.001 |
| Length of Stay (days) | 6.03 days | 5.92 days | < 0.001 | 1.017 | (1.014 – 1.021) | < 0.001 † | ||
| Hospitalization Cost ($) | 48,655 $ | 56,344 $ | < 0.001 | 0.857 | (0.857 – 0.857) | < 0.001 † | ||
| Disposition at Discharge | < 0.001 | |||||||
| Routine (%) | 62.30 % | 63.20 % | ||||||
| Short-term Hospital (%) | 2.82 % | 3.24 % | ||||||
| SNF or other facility (%) | 17.40 % | 13.20 % | ||||||
| Home Health Care (%) | 10.20 % | 10.20 % | ||||||
| AMA (%) | 4.17 % | 4.16 % | ||||||
| Died (%) | 3.13 % | 5.89 % | ||||||
| Unknown (%) | 0.03 % | 0.05 % | ||||||
| Liver complications | ||||||||
|
| ||||||||
| Ascites | 14.20 % | 20.90 % | < 0.001 | 0.63 | (0.62 – 0.64) | |||
| Varices | 15.50 % | 20.00 % | < 0.001 | 0.73 | (0.72 – 0.75) | |||
| Sepsis | 7.15 % | 10.40 % | < 0.001 | 0.66 | (0.64 – 0.68) | |||
| Variceal Bleeding | 3.94 % | 6.24 % | < 0.001 | 0.62 | (0.59 – 0.64) | |||
| Spontaneous Bacterial Peritonitis | 2.20 % | 3.30 % | < 0.001 | 0.66 | (0.63 – 0.69) | |||
| Hepatorenal syndrome | 3.00 % | 4.69 % | < 0.001 | 0.63 | (0.60 – 0.66) | |||
| Encephalopathy | 15.10 % | 17.50 % | < 0.001 | 0.84 | (0.82 – 0.86) | |||
| Acute Liver failure | 2.61 % | 3.65 % | < 0.001 | 0.71 | (0.67 – 0.74) | |||
Used Poisson regression analysis
Figure 2.
Figure 2 represents the multivariate forest plot using all-cause mortality as the final endpoint, with covariates including psychiatric disorders as independent variables.
Figure 3.
Figure 3 represents the multivariate forest plot using death due to suicide as the final endpoint, with covariates including psychiatric disorders as independent variables.
3.5. Supplementary tables
Supplementary Table 2 delineates the comprising elements of PI laid out in a list format, underscoring the individual incidence found in the study cohort. Supplementary Table 3 denotes the hospital outcomes of ALD patients when stratified using depression-only as primary exposure, since this variable represented the leading etiology among psychiatric illnesses. Likewise, Supplementary Tables 4–5 summarize the respective outcomes when using either anxiety or bipolar disorder as primary exposures.
4. Discussion
4.1. Findings and observations
This study is a propensity-score matched comparison of patients with ALD with and without PI. The study showed that those with underlying psychiatric conditions have increased risk of death secondary to suicide, although general, all-cause mortality rates were lower. Furthermore, those with psychiatric conditions were more likely to experience longer lengths of stay during hospitalization.
These study findings are concerning as psychiatric conditions independently confer an escalated risk toward suicide-related mortality. While this is expected, given that there is three times the risk ratio associated with suicide-related deaths among those with PI versus those without, the finding accentuates the critical need to evaluate the prevalence of psychiatric conditions in ALD patients and to develop ways to curtail these conditions and attenuate factors such as suicide risks by early detection and treatment. Prior studies demonstrated that excessive alcohol use and chronic presentation of ALD were associated with major depression, as well as suicidal ideation and attempts (26, 27). Moreover, from a physiological level, there appears to be an intricate connection between heavy alcohol use, ALD, and major depression (28, 29), with concurrent presence of ALD in alcohol users potentiating the risk of mental health disorders and depressive episodes (30). The neuromodulating physiologic effect of alcohol acts to alter the mood circuitry in the central nervous system (31) and previous studies have alluded to ALD directly affecting the mood-controlling systems in the amygdala (32), which may exacerbate the already deranged and impaired mood regulatory cycles and result in an exacerbated clinical response (33). Furthermore, excessive amounts of ingested alcohol may not be properly metabolized by the diseased liver, leading to toxic and undigested metabolites affecting the brain, as well as exacerbated hepatic encephalopathy (34, 35), which can likewise alter cognition and mood regulation (36). In addition, the ingestion of psychoactive substances and medications may accentuate the dysphoric response in ALD patients, as hepatic impairment may result in increased non-metabolized toxins and drug substances that disrupt mood regulation (34, 37). Nevertheless, molecular and psychologically-grounded studies are needed to precisely delineate the neuromodulatory imbalances in ALD patients, and to understand the mechanistic relationship between hepatic disorders and mood dysregulation in ALD adults.
4.2. Clinical implications
As this study shows that PI in patients with ALD increases the incidence of death due to suicide ideation, methods to improve outcomes in these patients should be discussed. As such, PI may be an important prognostic indicator as it is more common in patients with ALD compared to those with chronic liver diseases not caused by alcohol (9, 27). Interventions focused on helping patients understand ALD and their PI as well as providing strategies to manage their symptoms may help facilitate patient empowerment and a sense of control. This may help reduce symptoms of anxiety and depression while improving quality of life (38). To provide more individualized care, physicians should understand their patients’ individual coping strategies and provide support when they are undergoing or at risk of alcohol withdrawal. Both physicians and relatives of these patients can strengthen and acknowledge these patients’ coping skills through interaction and continued support. (39).
While most prior studies regarding integrative care were for those with hepatitis C and PI or substance use disorders, some of these strategies can be applied as well (40, 41). Motivational enhancement therapy, directed towards decreasing substance use, has been previously shown to increase the percentage of days of sobriety in patients with chronic hepatitis C and ongoing alcohol use (42). Similarly, using motivational enhancement therapy could help decrease alcohol consumption of patients with ALD and psychiatric disorders, which may ultimately lead to prevention of further damage to the liver and improve their psychiatric illnesses.
Limitations
Given the retrospective nature of the study, the study is susceptible to selection bias during the patient selection procedure. Furthermore, while the psychiatric comorbidities can be ascertained, the duration and the severity of the comorbidities cannot be stratified, given that this data is unavailable in the database. Similarly, the severity of alcohol consumption and the periods of sobriety cannot be ascertained from the database. As the database is derived from an inpatient registry, further prospective studies are needed to correlate the risks of PI in ALD patients beyond what is observed in inpatient settings, in order to delineate the long-term effects of PI on post-discharge care of ALD patients. Nonetheless, the current study employs a propensity-score matched comparison of a large in-hospital registry in order to demonstrate the risks of preexisting psychiatric illnesses in inpatient care and the outcomes of patients with ALD.
5. Conclusion
The presence of PI in patients with ALD was associated with higher deaths due to suicide and prolonged hospital stay. Screening of PI and suicidal ideation, and increasing efforts in providing education, counselling and relapse prevention therapy to patients with ALD may decrease mortality due to suicide and improve outcomes.
Supplementary Material
Footnotes
Declaration of Interest Statement:
DU Lee was funded by NIH NIDDK T32 DK067872–17. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.
Reviewer Disclosures
Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.
References
- 1.Callahan EJ, Jaén CR, Crabtree BF, Zyzanski SJ, Goodwin MA, Stange KC. The impact of recent emotional distress and diagnosis of depression or anxiety on the physician-patient encounter in family practice. J Fam Pract. 1998;46(5):410–418. [PubMed] [Google Scholar]
- 2.Hahn SR. Physical symptoms and physician-experienced difficulty in the physician-patient relationship. Ann Intern Med. 2001;134(9 Pt 2):897–904. [DOI] [PubMed] [Google Scholar]
- 3.el-Serag HB, Kunik M, Richardson P, Rabeneck L. Psychiatric disorders among veterans with hepatitis C infection. Gastroenterology. 2002;123(2):476–482. [DOI] [PubMed] [Google Scholar]
- 4.Blader JC. Acute inpatient care for psychiatric disorders in the United States, 1996 through 2007. Arch Gen Psychiatry. 2011;68(12):1276–1283. [DOI] [PubMed] [Google Scholar]
- 5.Merikangas KR, McClair VL. Epidemiology of substance use disorders. Hum Genet. 2012;131(6):779–789. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Christensen MK, Lim CCW, Saha S, et al. The cost of mental disorders: a systematic review. Epidemiol Psychiatr Sci. 2020;29:e161. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Adejumo AC, Cholankeril G, Iqbal U, et al. Readmission Rates and Associated Outcomes for Alcoholic Hepatitis: A Nationwide Cohort Study. Dig Dis Sci. 2020;65(4):990–1002. [DOI] [PubMed] [Google Scholar]
- 8.Pages KP, Russo JE, Roy-Byrne PP, Ries RK, Cowley DS. Determinants of suicidal ideation: the role of substance use disorders. J Clin Psychiatry. 1997;58(11):510–515; quiz 516–517. [PubMed] [Google Scholar]
- 9.Sarin SK, Sachdev G, Jiloha RC, Bhatt A, Munjal GC. Pattern of psychiatric morbidity and alcohol dependence in patients with alcoholic liver disease. Dig Dis Sci. 1988;33(4):443–448. [DOI] [PubMed] [Google Scholar]
- 10.Lee DU, Fan GH, Ahern RR, Karagozian R. The effect of malnutrition on the infectious outcomes of hospitalized patients with cirrhosis: analysis of the 2011–2017 hospital data. Eur J Gastroenterol Hepatol. Published online November 30, 2020. [DOI] [PubMed] [Google Scholar]
- 11.Lee DU, Fan GH, Hastie DJ, et al. The clinical impact of cirrhosis on the hospital outcomes of patients admitted with influenza infection: propensity score matched analysis of 2011–2017 US hospital data. J Clin Exp Hepatol. Published online February 4, 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Healthcare Cost and Utilization Project (HCUP). 2017 Introduction to the HCUP National Inpatient Sample (NIS). U.S. Agency for Healthcare Research and Quality, Rockville, MD; 2019. [Google Scholar]
- 13.Healthcare Cost and Utilization Project (HCUP). Agency for Healthcare Research and Quality, Rockville. HCUP National Inpatient Sample (NIS) 2012–2017. Published online December 2019. [Google Scholar]
- 14.Houchens R, Ross D, Elixhauser A, Jiang J. Nationwide Inpatient Sample (NIS) Redesign Final Report. U.S. Agency for Healthcare Research and Quality, Rockville MD; 2014. [Google Scholar]
- 15.Elixhauser A, Heslin KC, Owens PL. Healthcare Cost And Utilization Project (HCUP) Recommendations for Reporting Trends Using ICD-9-CM and ICD-10-CM/PCS Data. U.S. Agency for Healthcare Research and Quality; 2017. [Google Scholar]
- 16.Centers for Medicare and Medicaid Services (CMS). 2017 ICD-10 PCS General Equivalence Mappings (GEMs) – Procedure Codes. Published online December 1, 2016.
- 17.Centers for Medicare and Medicaid Services (CMS). 2017. ICD-10-CM General Equivalence Mappings (GEMs) – Diagnosis Codes. Published online August 22, 2016.
- 18.Fung KW, Richesson R, Smerek M, et al. Preparing for the ICD-10-CM Transition: Automated Methods for Translating ICD Codes in Clinical Phenotype Definitions. EGEMS (Wash DC). 2016;4(1):1211. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Agency for Healthcare Research and Quality, Rockville. Website. HCUP NIS Trend Weights. Healthcare Cost and Utilization Project (HCUP). [Google Scholar]
- 20.Agency for Healthcare Research and Quality, Rockville. Website. Producing National HCUP Estimates - Accessible Version. Healthcare Cost and Utilization Project (HCUP). D. [Google Scholar]
- 21.Zhang Z Multiple imputation with multivariate imputation by chained equation (MICE) package. Ann Transl Med. 2016;4(2):30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Liu Y, Han H, Shah G, et al. Significant Nationwide Variability in the Costs and Hospital Mortality Rates of Autologous Stem Cell Transplantation for Multiple Myeloma: An Analysis of the Nationwide Inpatient Sample Database. Biol Blood Marrow Transplant. 2019;25(1):41–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Houchens R Missing Data Methods for the NIS and the SID. HCUP Methods Series. Published online 2015. [Google Scholar]
- 24.Thadewald T, Büning H. Jarque–Bera Test and its Competitors for Testing Normality – A Power Comparison. J Appl Stat. 2007;34(1):87–105. [Google Scholar]
- 25.Jarque CM, Bera AK. A Test for Normality of Observations and Regression Residuals. Int Stat Rev. 1987;55(2):163–172. [Google Scholar]
- 26.Le Strat Y, Le Foll B, Dubertret C. Major depression and suicide attempts in patients with liver disease in the United States. Liver Int. 2015;35(7):1910–1916. [DOI] [PubMed] [Google Scholar]
- 27.Jinjuvadia R, Jinjuvadia C, Puangsricharoen P, et al. Concomitant Psychiatric and Nonalcohol-Related Substance Use Disorders Among Hospitalized Patients with Alcoholic Liver Disease in the United States. Alcohol Clin Exp Res. 2018;42(2):397–402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Ukai W, Ishii T, Hashimoto E, et al. The common aspects of pathophysiolgy of alcoholism and depression. Nihon Arukoru Yakubutsu Igakkai Zasshi. 2009;44(6):704–711. [PubMed] [Google Scholar]
- 29.Badawy AA, Morgan CJ, Lovett JW, Bradley DM, Thomas R. Decrease in circulating tryptophan availability to the brain after acute ethanol consumption by normal volunteers: implications for alcohol-induced aggressive behaviour and depression. Pharmacopsychiatry. 1995;28 Suppl 2:93–97. [DOI] [PubMed] [Google Scholar]
- 30.Ewusi-Mensah I, Saunders JB, Wodak AD, Murray RM, Williams R. Psychiatric morbidity in patients with alcoholic liver disease. Br Med J. 1983;287(6403):1417–1419. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Koob GF. Focus on: Neuroscience and treatment: the potential of neuroscience to inform treatment. Alcohol Res Health. 2010;33(1–2):144–151. [PMC free article] [PubMed] [Google Scholar]
- 32.Wrase J, Makris N, Braus DF, et al. Amygdala volume associated with alcohol abuse relapse and craving. Am J Psychiatry. 2008;165(9):1179–1184. [DOI] [PubMed] [Google Scholar]
- 33.Marvel CL, Paradiso S. Cognitive and neurological impairment in mood disorders. Psychiatr Clin North Am. 2004;27(1):19–36, vii – viii. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Cederbaum AI. Alcohol metabolism. Clin Liver Dis. 2012;16(4):667–685. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Butterworth RF. Role of circulating neurotoxins in the pathogenesis of hepatic encephalopathy: potential for improvement following their removal by liver assist devices. Liver Int. 2003;23 Suppl 3:5–9. [DOI] [PubMed] [Google Scholar]
- 36.Bajaj JS, Schubert CM, Heuman DM, et al. Persistence of cognitive impairment after resolution of overt hepatic encephalopathy. Gastroenterology. 2010;138(7):2332–2340. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Elbekai RH, Korashy HM, El-Kadi AOS. The effect of liver cirrhosis on the regulation and expression of drug metabolizing enzymes. Curr Drug Metab. 2004;5(2):157–167. [DOI] [PubMed] [Google Scholar]
- 38.Lau-Walker M, Presky J, Webzell I, Murrells T, Heaton N. Patients with alcohol-related liver disease--beliefs about their illness and factors that influence their self-management. J Adv Nurs. 2016;72(1):173–185. [DOI] [PubMed] [Google Scholar]
- 39.Mikkelsen MR, Hendriksen C, Schiødt FV, Rydahl-Hansen S. Coping and rehabilitation in alcoholic liver disease patients after hepatic encephalopathy--in interaction with professionals and relatives. J Clin Nurs. 2015;24(23–24):3627–3637. [DOI] [PubMed] [Google Scholar]
- 40.Patel K, Maguire E, Chartier M, Akpan I, Rogal S. Integrating Care for Patients With Chronic Liver Disease and Mental Health and Substance Use Disorders. Fed Pract. 2018;35(Suppl 2):S14–S23. [PMC free article] [PubMed] [Google Scholar]
- 41.Ho SB, Groessl E, Dollarhide A, Robinson S, Kravetz D, Dieperink E. Management of chronic hepatitis C in veterans: the potential of integrated care models. Am J Gastroenterol. 2008;103(7):1810–1823. [DOI] [PubMed] [Google Scholar]
- 42.Dieperink E, Fuller B, Isenhart C, et al. Efficacy of motivational enhancement therapy on alcohol use disorders in patients with chronic hepatitis C: a randomized controlled trial. Addiction. 2014;109(11):1869–1877. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.