Inpatient outcomes of rheumatoid arthritis in hospitalized patients using cannabis: data from the National Inpatient Sample

Abstract

Background

Despite advancements in rheumatoid arthritis (RA) treatment, effective pain management remains challenging, with many patients also experiencing insomnia, anxiety, and depression. Cannabis, with its analgesic properties, offers a promising nonopioid alternative. This study evaluated outcomes in hospitalized RA patients using cannabis through the National Inpatient Sample (NIS) database.

Methods

This retrospective study analyzed NIS data (2016–2021) using ICD-10 codes to identify RA patients. Demographic and clinical characteristics and inpatient outcomes were compared between cannabis users and nonusers.

Results

RA patients using cannabis (n = 42,415) were younger (mean age 51.8 vs 67.8 years), less likely to be female (60.65% vs 73.71%), and more likely to be African American (24.02% vs 12.86%) and Native American (2.25% vs 0.86%). Cannabis use was associated with lower mortality (0.98% vs 2.71%) and hospital charges ($57,773 vs $63,681). After adjusting for age, gender, race, and comorbidities, cannabis use was linked to decreased mortality (odds ratio [OR]: 0.50), depression (OR: 0.47), chronic pain (OR: 0.45), and anxiety (OR: 0.55). Conversely, cannabis use increased the risk of opioid use (OR: 1.10), nicotine dependence (OR: 1.35), and alcohol use (OR: 1.35).

Conclusion

RA patients using cannabis had lower mortality, depression, chronic pain, and anxiety, but higher risks of opioid, nicotine, and alcohol use. Further research is needed on the long-term effects of cannabis in RA management.

Rheumatoid arthritis (RA) is a chronic autoimmune disease affecting 0.5% to 1% of the global population and about 1% of adults in the United States, with a higher prevalence among women.^1–4 Currently, various options are available for the treatment of RA, which include glucocorticoids, conventional synthetic disease-modifying anti-rheumatic drugs (DMARDs), targeted synthetic DMARDs, and biological DMARDs.⁵ Despite various treatment options, a significant proportion of patients still complain of inadequate pain control.⁶ Chronic pain in RA is often accompanied by other distressing symptoms, such as insomnia, depression, and anxiety, which further compromise patients’ quality of life.⁷ Thus, besides prescription medication for RA, patients often seek alternatives for pain relief that include acupuncture/acupressure, hypnosis, herbal therapies, or cannabis.^8–11 Among these, cannabis has been the most popular.

Cannabis is one of the oldest cultivated crops by man but has always been mired in controversy.¹² According to the Centers for Disease Control and Prevention, 19% of Americans have used cannabis at least once, and it’s the most commonly used illegal drug.¹³ Cannabis users among patients with rheumatologic disorders are also in a similar range, at 18.3%.¹⁴ Studies have demonstrated the anti-inflammatory properties of cannabis, prompting researchers to conduct more research to look for the potential use of cannabis for decreasing pain and preventing joint destruction in rheumatic diseases.^15,16 While some studies suggest cannabis may have immunomodulatory properties, there’s insufficient evidence to confirm its role. Cannabinoid receptors, in synovial tissue, play a role in mediating chronic pain in arthritis and, thus, cannabinoid receptor agonists can help to alleviate the pain.^16,17

Despite the growing interest in cannabis as a treatment for rheumatic conditions, its effects on hospitalized RA patients remain largely unknown. With cannabis becoming increasingly legalized for medical use across the United States, it is important to examine its impact on key inpatient outcomes such as mortality, hospital length of stay, and overall treatment costs. Using data from the National Inpatient Sample (NIS), this study aimed to compare the clinical outcomes of RA patients who used cannabis with those who did not. The findings contribute to a better understanding of the potential risks and benefits of cannabis use in the hospital setting, offering valuable insights for clinicians managing RA patients.

METHODS

A retrospective observational study was conducted utilizing data from the NIS database from 2016 to 2021. NIS is the largest publicly accessible database maintained by the Agency for Healthcare Research and Quality. The database uses the International Classification of Diseases, Tenth Revision (ICD-10) coding system to record inpatient and outpatient diagnoses and procedures (Table 1). Since the dataset is publicly available and deidentified, the study was exempt from institutional review board approval.

Table 1.

ICD-10 codes used to extract the variables

Diagnosis	ICD-10 CM codes/PCS codes
Rheumatoid arthritis	M05, M06
Cannabis use and dependence	F12
Opioid dependence	F11
Depression	K50
Anxiety disorders	F41
Schizophrenia	F20
Nicotine dependence	F17
Alcohol use disorder and dependence	F10
Fibromyalgia	M79.7
Hypertension	I10, I15,
Hyperlipidemia	E78.1, E78.2, E78.3, E78.4, E78.5
Heart failure	I50
Diabetes	E10.1, E10.9, E11.1, E11.9, E13.1, E13.9, E14.1, E14.9, E10.2, E10.4, E11.2, E11.4, E13.2, E13.4, E14.2, E14.4
Overweight and obesity	E66
Coronary artery disease	I25.1, I25.2
Myocardial infarction	I21
Chronic kidney disease	N18.1, N18.2, N18.3, N18.4, N18.5
End-stage renal disease	N18.6
Severe sepsis with/without septic shock	R65.21, R65.20
Acute kidney injury	N17
Cardiac arrest	I46
Cardiogenic shock	R57.0

All adult patients (>18 years of age) with a diagnosis of RA were included in the study. We compared the demographics, baseline clinical characteristics, comorbidities, and inpatient outcomes, including mortality, hospital length of stay, and total hospital charges, in patients with and without cannabis use.

Data analysis was performed using Stata version 17. Chi-square and Student’s t tests were used to compare categorical and continuous variables, respectively. Frequencies, confidence intervals, and P values were reported for all outcomes, with statistical significance set at a P value <0.05. Univariate and multivariate regression analyses were conducted to estimate the risk of various complications. The desired outcomes were analyzed individually across various clinical and demographic characteristics, and variables found to be significant in univariate analysis were included in the multivariate analysis.

RESULTS

There were 3,347,284 RA patients hospitalized during the study period for various clinical reasons. Among them, 1.27% (n = 42,415) had a concomitant diagnosis of cannabis use at the time of hospitalization, while the remaining 98.73% were nonusers (n = 3,304,869). Of the cannabis users, 60.65% were female, whereas 73.71% of nonusers were male (P < 0.001). Cannabis use among RA patients is associated with distinct racial patterns: fewer Caucasians (63.52% vs 74.13%) and Asians (0.52% vs 1.58%), but more African Americans (24.02% vs 12.86%) and Native Americans (2.25% vs 0.86%), with similar proportions in other categories (P < 0.001) (Table 2). Cannabis use was more common in the lowest-income quartile (38.76% vs 29.07%). The proportion of cannabis users was similar in the 25th to 50th and 50th to 75th income quartiles (27.52% vs 27.05% and 21.50% vs 24.23%, respectively). However, cannabis use was lower in the highest-income quartile (12.22% vs 19.65%) (Table 2).

Table 2.

Baseline characteristics of rheumatoid arthritis patients with and without cannabis use

Variables	Rheumatoid arthritis patient		P value
	With cannabis use (n = 42,415)	Without cannabis use (n = 3,304,869)
Age (years), mean (95% CI)	51.80 (51.50–52.10)	67.77 (67.71–67.83)
Female, n (%)	25,724 (60.65%)	2,436,019 (73.71%)	<0.001
Race, n (%)			<0.001
Caucasian	26,942 (63.52%)	2,449,899 (74.13%)
African American	9890 (24.02%)	425,006 (12.86%)
Hispanic	3065 (7.44%)	278,931 (8.44%)
Asian	215 (0.52%)	52,216 (1.58%)
Native American	925 (2.25%)	28,421 (0.86%)
Other	925 (2.25%)	70,063 (2.12%)
Median household income, n (%)			<0.001
0–25th percentile	16,440 (38.76%)	960,725 (29.07%)
25th–50th percentile	11,673 (27.52%)	893,967 (27.05%)
50th–75th percentile	8885 (21.50%)	800,769 (24.23%)
75th–100th percentile	5050 (12.22%)	649,407 (19.65%)
Comorbidities, n (%)
Acute myocardial infarction	4110 (9.69%)	358,909 (10.86%)	<0.001
Congestive heart failure	6550 (15.44%)	838,445 (25.37%)	<0.001
Peripheral vascular disease	2895 (6.83%)	315,945 (9.56%)	<0.001
Cerebrovascular disease	2320 (5.47%)	236,298 (7.15%)	<0.001
Dementia	615 (1.45%)	233,984 (7.08%)	<0.001
Chronic obstructive pulmonary disease	16,185 (38.16%)	1,160,009 (35.10%)	<0.001
Peptic ulcer disease	850 (2.00%)	64,775 (1.96%)	0.78
Mild liver disease	3515 (8.29%)	135,169 (4.09%)	<0.001
Diabetes	4845 (11.42%)	518,864 (15.70%)	<0.001
Diabetes with complications	4280 (10.09%)	469,291 (14.20%)	<0.001
Renal disease	5230 (12.33%)	707,572 (21.41%)	<0.001
Cancer	1405 (3.31%)	176,149 (5.33%)	<0.001
Moderate/severe liver disease	770 (1.82%)	40,319 (1.22%)	<0.001
Metastatic cancer	730 (1.72%)	86,257 (2.61%)	<0.001
Chronic pain	7885 (18.59%)	369,814(11.19%)	<0.001
Depression	12,928 (30.48%)	657,999 (19.91%)	<0.001
Opioid use	5145 (12.13%)	85,926 (2.6%)	<0.001
Fibromyalgia	5895 (13.90%)	283,557 (8.58%)	<0.001
Autoimmune thyroid disease	180 (0.42%)	12,558 (0.38%)	0.48
Schizophrenia	1075 (2.53%)	17,185 (0.52%)	<0.001
Obesity	6970 (16.43%)	657,999 (19.91%)	<0.001
Anxiety	13,950 (32.89%)	629,908 (19.06%)	<0.001
Outcomes
Mortality, n (%)	415 (0.98%)	11,494 (2.71%)	<0.001
Hospital length of stay (days), mean (95% CI)	5.27 (5.14–5.41)	5.16 (5.14–5.18)	0.1
Total hospital charges ($), mean (95% CI)	$57,773 (55,857–59,690)	$63,681 (63,044–64,318)	<0.001

Among the RA patients using cannabis, comparisons were made among common comorbidities, including chronic obstructive pulmonary disease (COPD) (38.16%), anxiety (32.89%), depression (30.48%), chronic pain (18.59%), and obesity (16.43%). In nonusers, the most common comorbidities were COPD (35.10%), congestive heart failure (25.37%), renal disease (21.41%), depression (19.91%), and obesity (19.91%) (Table 2).

This study examined three primary outcomes for hospitalized RA patients: mortality, total hospitalization costs, and length of stay. The mortality rate was significantly lower among cannabis users (0.98% vs 2.71%, P < 0.001). Cannabis users also incurred lower total hospitalization costs compared to nonusers ($57,773 vs $63,681, P < 0.001). However, there was no significant difference in the length of stay, with cannabis users staying an average of 5.27 days compared to 5.16 days for nonusers (P = 0.1) (Table 2).

In the multivariate logistic regression analysis, after adjusting for age, gender, race, and comorbidities, cannabis use in RA patients was associated with a significant decrease in mortality (OR: 0.5, 95% CI: −0.72 to −0.27), depression (OR: 0.47, 95% CI: 0.42 to 0.52), chronic pain (OR: 0.45, 95% CI: 0.39 to 0.51), and anxiety (OR: 0.55, 95% CI: 0.50 to 0.60). However, cannabis use was also associated with an increased risk of opioid use (OR: 1.1, 95% CI: 1.02 to 1.17), nicotine dependence (OR: 1.35, 95% CI: 1.30 to 1.40), and alcohol use (OR: 1.35, 95% CI: 1.29 to 1.42) (Table 3).

Table 3.

Adjusted odds ratios for rheumatoid arthritis patients who were cannabis users versus nonusers

Outcome	aOR	95% CI	P value
Mortality	−0.5	−0.72 to −0.27	<0.001
Depression	0.47	0.42–0.52	<0.001
Opioid use	1.1	1.02–1.17	<0.001
Chronic pain	0.45	0.39–0.51	<0.001
Anxiety	0.55	0.50–0.60	<0.001
Nicotine dependence	1.35	1.30–1.40	<0.001
Alcohol use	1.35	1.29–1.42	<0.001

aOR indicates adjusted odds ratio; CI, confidence interval.

DISCUSSION

This study was conducted to evaluate the inpatient outcomes of RA in hospitalized patients who use cannabis. Data for this study were obtained from the NIS from 2016 to 2021. Among more than 3 million patients with RA, about 1.27% used cannabis. This percentage is less than the general population who use cannabis in the US as well as among other patients with rheumatologic disorders or other medical conditions.^13,14,18 This is contrary to the finding that people with medical conditions are more likely to use cannabis daily for analgesia or other medical reasons.¹⁸ One of the reasons for the lower incidence of cannabis use in our study could be underreporting by hospitalized patients.

Our study showed that the African American population had a high prevalence of cannabis users, which supports similar finding by Hasin et al and Patrick et al, who studied the trends of cannabis use in the US.^19,20 Similarly, among RA patients, only 60.65% of cannabis users were female, compared to 73.71% of non–cannabis users. This shows that female patients with RA were less likely to use cannabis compared to male patients. The trend of cannabis use among the general population and patients with RA is comparable in terms of age, gender, and race.^18–20

Our analysis found a low mortality among RA patients who used cannabis, with a mortality rate of 0.98% for cannabis users compared to 2.71% for nonusers (OR: −0.5, 95% CI: −0.72 to −0.27). Furthermore, cannabis users had lower total hospital charges, averaging $57,773 compared to $63,681 for nonusers. There was no statistically significant difference in the length of hospital stay between the two groups.

After adjusting for potential confounders such as age, gender, race, and comorbidities, our analysis demonstrated that cannabis use was associated with a lower prevalence of depression (OR: 0.47, 95% CI: 0.42 to 0.52), chronic pain (OR: 0.45, 95% CI: 0.39 to 0.51), and anxiety (OR: 0.55, 95% CI: 0.50 to 0.60). While these findings suggest potential benefits of cannabis use for symptom management, it is important to recognize that the current evidence is observational, and there is a need for more detailed research to assess the risks and long-term effects, particularly in patients with rheumatologic conditions. A study conducted to identify the increasing cannabis use in patients with rheumatologic conditions demonstrated that those who use cannabis have high scores for pain, sleep, fatigue, anxiety, and depression.¹⁴ Users also had more comorbid conditions.¹⁴ Thus, they had overall worse disease activity. Fitzcharles et al indicated that physicians are unsure about the use of cannabis for the treatment of medical conditions due to the lack of evidence and formalized training.²¹ Thereby, they have drawn the attention of governments and regulators from all over the world to support scientific research on the use of cannabis so that patients can benefit from its use, if there is any. Otherwise, unregulated use may increase the risk of a cannabis overdose crisis resulting in another potential problem similar to the “opioid epidemic.”

Conversely, our study also revealed that cannabis use was associated with an increased risk of opioid consumption (OR: 1.1, 95% CI: 1.02 to 1.17), nicotine dependence (OR: 1.35, 95% CI: 1.30 to 1.40), and alcohol use (OR: 1.35, 95% CI: 1.29 to 1.42). These findings further strengthen the fact that it is debatable to justify the use of cannabis merely based on a handful of studies that show the benefit in some aspects of clinical outcome. A systematic review conducted in Brazil by de Carvalho et al concluded that the use of cannabis as a treatment of rheumatologic disease is an interesting topic for discussion; however, adequate studies need to be conducted.²² We agree that cannabis can be an alternative to opioids if its use for medical conditions is scientifically backed up through high-quality research. The potential motivation for this could be the fact that no evidence is available about fatal overdoses with cannabis as opposed to opioids.²³ Also, the risk of physical dependence and accidental overdose in the case of cannabis is significantly lower compared to opioids. Thus, the role of cannabis and its potential benefits over opioids warrant further exploration through well-designed studies.

Our study on inpatient outcomes in RA patients using cannabis has several limitations. The dataset lacked detailed clinical information, such as laboratory results, medication specifics, and treatment interventions, which is crucial for a deeper understanding of patient outcomes and treatment contexts. Treating joint pain and inflammatory arthritis differs, as chronic pain in RA is often linked to fibromyalgia or secondary osteoarthritis. Cannabis’ role in managing this pain, considering these comorbid conditions, should be recognized as a confounding factor in the study. Another limitation is the challenge of studying inpatient RA outcomes, as most flares are managed in an outpatient setting, and hospitalized patients often have non-RA conditions. Additionally, we lacked data on disease severity, which could influence cannabis use decisions, and correlating hospital stay and mortality with cannabis use is difficult since patients do not receive cannabis during hospitalization. The accuracy of the data is contingent on proper coding by healthcare providers and institutions, and any misclassification or coding errors may result in incomplete or inaccurate data. Further data collection would need to improve reporting and differentiate casual use from habitual use, which could also help explain the disparity between our findings and epidemiological data on cannabis use. Lastly, the applicability of our findings may be restricted to the population represented in the NIS. Despite these limitations, our study contributes valuable insights into the inpatient outcomes of cannabis use in RA patients. By utilizing a population-based dataset, we enhance the generalizability of our findings across diverse healthcare settings.

In conclusion, this study found that RA patients using cannabis were younger, more often male, and more often African American or Native American. Patients with RA using cannabis were found to have a low rate of mortality, depression, chronic pain, and anxiety. However, they also had increased risks of opioid use, nicotine dependence, and alcohol use. Findings suggest that while cannabis may offer potential benefits in reducing mortality and certain mental health conditions among RA patients, it is also linked to an increased likelihood of substance use disorders, highlighting the need for a balanced approach to managing patient care. Further research is warranted to explore the underlying mechanisms of these associations and to develop strategies that maximize the benefits of cannabis use while minimizing the risks.

Disclosure statement/Funding

The authors report no funding or conflict of interest.