Factors associated with comorbidity in hospitalized and non-hospitalized inflammatory bowel disease patients: A single-center, preliminary study

Abstract

Background

Inflammatory bowel disease (IBD) may be burdened by other comorbid conditions. We herein sought to assess comorbidity in hospitalized and non-hospitalized IBD patients.

Methods

This was part of the San MAtteo Complexity (SMAC) study (2017–2019). Data of hospitalized IBD patients were compared to gender- and age-matched IBD outpatients in a 2:1 fashion. The association of comorbidity in relation to hospitalization and clinical and socioeconomic factors was assessed.

Results

We included 104 patients, 27 hospitalized (median age 49 years, IQR 32-70) and 77 outpatients (median age 54 years, IQR 37-68). Comorbidity was reported in 63/104 patients (60.6%), of whom 45 (58.5%) non-hospitalized and 18 (66.6%) hospitalized. Patients with comorbidity were older (median 60 years, IQR 44-61 vs median 40 years, IQR 28-50 vs; p<0.001), had a higher Cumulative Illness Rating Scale severity index (median 1.85, IQR 1.5-2.5 vs median 1.31, IQR 1.2-2.5; p=0.002), were more commonly female (37, 60.7% vs 14, 35.0%; p=0.01), frailer (14, 22.2% vs 1, 2.4%; p=0.004), and had a lower educational level (13, 20.6% vs 2, 4.9%; p=0.04). In a bivariable analysis, factors associated to comorbidity were age >65 years (OR 5.30, 95% CI 1.81-15.55; p=0.002), female sex (OR 2.92, 95% CI 1-27-6.71; p=0.012), income <1000€ (OR 3.04, 95% CI 1.09-8.44; p=0.033), schooling <8 years (OR 5.09, 95% CI 1.08-23.96; p=0.039), frailty (OR 12.56, 95% CI 1.48-106.45; p=0.020), and polypharmacy (OR 10.41, 95% CI 1.85-59.38; p=0.008).

Conclusion

A high prevalence of comorbidity was found in IBD, possibly related to low socioeconomic status and poor educational level.

Introduction

Inflammatory bowel disease (IBD), including Crohn’s disease (CD) and ulcerative colitis (UC), is a chronic immune‐mediated disorder of the gastrointestinal tract with relapsing-remitting course and heterogeneous clinical manifestations. ¹ At present, IBD cannot be cured and represents a lifelong disease burden ² and even patients with long periods of remission may experience a disease flare or a complication requiring surgery. ³ IBD complications are several and they may potentially be ascribed in a comorbidity framework,^4,5 where IBD represents the index disease, and all its complications and extra-intestinal manifestations represent the comorbid conditions. ⁶ Despite the impact of comorbid conditions on IBD clinical outcomes, there are few studies looking at this issue both in outpatient and hospitalized settings.

Indeed, since novel therapies have been made available, a reduction of the risk of IBD flares and hospitalization has been noticed, and the majority of patients can now be managed in an outpatient setting.^7,8 Accordingly, a large multinational study looking at hospitalization rates for IBD in 34 countries, showed that there is a stable or decreasing trend, possibly due to the shift in the care from hospitals to outpatient clinics. ⁹ Table 1 reports the studies focusing on IBD hospitalization trends and its risk factors.^9–27 Of note, the impact of associated conditions and socioeconomic factors have been poorly addressed.

Table 1.

Available evidence reporting on hospitalization trends in IBD patients.

First author and year	Study type	Aim	Main findings
Sonnenberg, 2009 20	Retrospective, administrative data-based	To study the characteristics of hospitalized patients	Female predominance in CD, male predominance in UC; IBD patients tended to be white, and live in metropolitan areas
Button, 2010 12	Retrospective administrative data-based	To establish hospitalized prevalence of IBD	Higher prevalence of hospitalized CD in more deprived areas; no association between social deprivation and hospitalized prevalence in UC
Costa, 2013 13	Systematic review	To study the impact of infliximab on hospitalization	Infliximab reduced risks of hospitalization and surgery in 27 studies
Vester-Andersen, 2014 26	Retrospective population-based cohort	To evaluate probability of recurrence and admission rates	Cumulative risk of recurrence rates in 2003-2011 were reduce, but there was no change in surgery and hospitalization rates
Vegh, 2015 24	Retrospective web-based cohort	To analyze hospitalization rate after 1-year diagnosis	Surgery and hospitalization rates were significantly higher in CD patients in Eastern compared to Western Europe/Australia
Annese, 2016 10	Meta-analysis	To analyze the impact of novel therapies on hospitalization	In Europe hospitalization had a reduction trend after the introduction of biologics, though rates of hospitalization varied among countries
Mao, 2017 15	Systematic review and meta-analysis	To analyze trends of hospitalization	Anti-TNFα reduced surgery and hospitalization by 50%; azathioprine was not associated with improvement
Nguyen, 2018 17	Retrospective cohort	To estimate annual burden for hospitalization in older vs younger patients	Older IBD patients had higher burden and costs of hospitalization. Attributed to serious infections and cardiovascular complications
King, 2019 9	Retrospective population-based	To assess hospitalization trends	Hospitalization in Western countries is either stable or decreasing; in developing countries there is an inverse trend
Nguyen, 2019 18	Retrospective cohort	To estimate the burden for hospitalization in obese vs non-obese patients	Obesity related to longer length of stay and costs
Murthy, 2020 16	Retrospective administrative data-based	To analyze trends of hospitalization after infliximab introduction	Infliximab did not reduce rates of IBD-related surgeries and hospitalizations
Faye, 2021 14	Retrospective administrative data-based	To assess the prevalence of frailty in the hospital setting	Frailty was associated with a higher risk of readmission and mortality
Scaldaferri, 2021 21	Retrospective, single-center	To evaluate trends of hospitalization	Hospitalizations have decreased in recent years, especially in younger patients
Verdon, 2021 25	Retrospective administrative data-based	To assess the hospitalization trends in 20-year follow-up	Anti-TNFα therapy was associated with low risk of hospitalization but overall trend remained unchanged; surgery risk increased even in biological patients
Zhao, 2021 27	Systematic review and meta-analysis	To analyze trends of diagnosis, phenotype, hospitalization	Surgery rates declined while hospitalization had no clear trend
Tsai, 2022 23	Systematic review and meta-analysis	To assess the risk of hospitalization	Risk of hospitalization is declining in patients diagnosed in the biologic era
Nguyen, 2022 19	Retrospective web-based cohort	To evaluate whether PROMs inform risk of unplanned healthcare utilization	PROMs may have a modest effect on modifying risk of unplanned healthcare utilization
Buie, 2023 11	Systematic review	To analyze global trends of hospitalization	Western countries have a stable incidence and progressive increased prevalence, newly industrialized countries have increased incidence
Rasmussen, 2024 22	Systematic review and meta-analysis	To assess the impact of sex, income, education on hospitalization	Male patients were more at risk of surgery with no differences in gender for hospitalization

Abbreviations: Inflammatory Bowel Disease (IBD); Crohn’s Disease (CD); Ulcerative Colitis; (UC); Anti-tumor necrosis factor alpha (anti-TNFα); Patient-Reported Outcome Measures (PROMs).

In this regard, the aim of this study was to evaluate the prevalence of comorbidity and its demographic, social, and clinical characteristics in a prospectively enrolled cohort of IBD patients in both a hospital and an outpatient setting.

Methods

In this single-center study, data from the San MAtteo Complexity (SMAC) study, a large ongoing prospective research project on clinical complexity (NCT03439410) were analyzed. In the SMAC study, adult patients admitted to our internal medicine unit have been included, regardless of the cause of admission. This academic unit also comprises a tertiary referral center for the diagnosis and treatment of IBD. The details of the enrolment have already been described elsewhere. ²⁸ Denial of informed consent and a prognosis <24 hours were the only exclusion criteria.

For the purposes of the present study, we only included patients with either a new diagnosis or a history of IBD (both CD and UC), regardless of the cause of hospitalization (that was either related to IBD activity or not). These patients were selected through the ICD9 codes pertaining to IBD (i.e., 555.0, 555.1, 555.2, 556.0, 556.5, 556.2, 556.3, and 556.6). Non-inflammatory bowel diseases were excluded. In all cases, the diagnosis was confirmed by reviewing the discharge letter. By applying these criteria, 27 consecutive hospitalized IBD patients were included. The second sample consisted of patients affected by IBD but managed as outpatients. The diagnosis was based on clinical, radiological, and endoscopic/histological criteria, according to internationally agreed guidelines. ²⁹ Disease location and extent were confirmed by endoscopy in UC, and by endoscopy plus imaging in CD. In order to avoid a statistical unbalance and potential biases, due to the small sample size of hospitalized patients, we prospectively enrolled 77 IBD outpatients, hence in a 2:1, age- and sex-matched, fashion, who had been diagnosed at least three months prior to study enrollment. Outpatients were consecutively enrolled in the dedicated IBD outpatient clinic of the same Internal Medicine Unit, hence randomly including both recently diagnosed patients, and those with a longer disease history, over a 1-month time span, following the age and sex matching rules. By doing so, we have enrolled a sample representative of the general IBD outpatient population.

As a primary aim, we assessed the prevalence of comorbidity in both hospitalized and non-hospitalized IBD patients, along with relevant sociodemographic and clinical characteristics, namely age, gender, age of IBD onset, therapy (steroid, immunosuppressive, biological), Harvey-Bradshaw Index (HBI) and Mayo score for CD and UC, respectively (dividing patients into active IBD or inactive IBD), Cumulative Illness Rating Scale (CIRS) severity and comorbidity indexes, Barthel index, Edmonton Frail Scale, 25-item Connor-Davidson Resilience Scale, and the Short Blessed test. An expert physician classified all patients, both from the hospital and outpatient settings, as having comorbidity according to current definitions (i.e., any condition in addition to the index disease under study, that is IBD in our population).^30–32 Under the variable “number of chronic conditions”, we simply enumerated the total number of conditions, also including IBD and their complications or extra-intestinal manifestations, if any. As a secondary aim, we assessed potential sociodemographic and clinical factors associated to comorbidity in the whole cohort of IBD patients, in relation to the hospitalization status.

Statistical analysis

Continuous data were described with the median and interquartile range (IQR) and compared between groups according to the type of cohort (outpatients vs hospitalized) and the presence of comorbidity vs no comorbidity with the Mann Whitney U test. Categorical data were reported as counts and percentages and compared with the Fisher exact test. Logistic regression was used to measure the association of the type of cohort with the presence of comorbidity. Odds ratios (OR) and 95% confidence intervals were computed. Though no multivariable analysis could be performed due to the sample size, this association was adjusted in turn in a bivariable analysis, for a series of candidate confounders, which were considered the most relevant patient clinical characteristics in relation to the hospitalization status. The modifying effect of the confounders was assessed by including an interaction term with cohort in the model. The software Stata 17 (StataCorp, College Station, TX, USA) was used for all computations. A 2-sided p-value <0.05 was considered statistically significant. The study was approved by the local Ethics Committee, as an extension of the SMAC study (Protocol number NCT03439410). The study follows the STROBE recommendations for quality assurance. No patients, nor members of the public were involved in the design or interpretation of the study. All data relevant to this study have been here reported; additional raw data can be shared upon reasonable request to the corresponding author.

Results

Overall, we included 104 patients, of whom 27 were hospitalized (median age 49 years, IQR 32-70) and 77 outpatients (median age 54 years, IQR 37-68) in a 2:1, age- and sex-matched, fashion.

Table 2 reports the sociodemographic characteristics and main clinical features of the non-hospitalized vs hospitalized patients. As per protocol, there were no age and sex differences. The bivariable analysis is reported later in another Table. Of note, there were no differences regarding Barthel index, CIRS comorbidity, and polypharmacy. Conversely, hospitalized IBD patients tended to have a lower BMI (p=0.056) and to be frailer (p<0.001), more socioeconomically deprived (p=0.05), more cognitively impaired (p=0.02), and were more likely to have a CIRS score >3 (p=0.019).

Table 2.

Sociodemographic and main clinical features of the non-hospitalized vs hospitalized patients.

	Non-hospitalized, median (IQR)	Hospitalized, median (IQR)	p-value
Overall population	77	27	/
Age	54 (37-68)	49 (32-70)	0.549
Number of chronic conditions	3 (2–4)	2.5 (1–5)	0.765
CD-RISC25	63.0 (48-78)	69 (63.5-77)	0.521
Barthel index	100 (100-100)	100 (96-100)	0.598
BMI	24.1 (22-27)	22.1 (19.4-26.2)	0.056
CIRS comorbidity	1.0 (1-2)	1.0 (1-3)	0.461
CIRS severity	1.69 (1.3-2.5)	1.38 (1.2-1.7)	0.021
Number of medications	3 (2–4)	4 (2–6)	0.148

Abbreviations: 25-item Connor Davidson Resilience Scale (CD-RISC25); Body Mass Index (BMI); Cumulative Illness Rating Scale (CIRS); Short Blessed Test (SBT); Edmonton Frail Scale (EFS); Short Blessed Test (SBT).

Table 3 reports the clinical characteristics of IBD in the two populations studied. Of note, hospitalized patients had a more active IBD, both UC and CD, as compared to outpatients (p<0.001), and fewer of them were treated with biologics (2, 7.4% vs 32, 41.6%; p=0.001) or mesalazine (13, 48.1% vs 53, 68.8%; p=0.05). Conversely, systemic steroid therapy was more commonly prescribed in the hospital setting (8, 29.6%, vs 10, 13.0%; p=0.05).

Table 3.

Clinical characteristics of IBD in non-hospitalized vs hospitalized patients.

	Non-hospitalized, n (%)	Hospitalized, n (%)	p-value
Crohn’s disease	43 (55.8)	16 (59.3)	NA
Active disease	0	6 (37.5)	0.001
Ulcerative colitis	34 (44.2)	11 (40.7)	NA
Active disease	6 (17.6)	8 (72.7)	<0.0001
Treatment
Biologics
Never/Past	45 (58.4)	25 (92.6)	0.001
Current	32 (41.6)	2 (7.4)	0.001
Systemic steroid
Never/past	67 (87.0)	19 (70.4)	0.07
Current	10 (13.0)	8 (29.6)	0.05
Oral mesalazine
Never/past	24 (31.2)	14 (51.8)	0.06
Current	53 (68.8)	13 (48.1)	0.05
Immunosuppressants
Never/past	66 (85.7)	27 (100)	0.06
Current	11 (14.3)	0	0.03

Abbreviations: Inflammatory Bowel Disease (IBD).

Table 4 reports the sociodemographic and main clinical features of patients with no comorbidity vs comorbidity. Overall, comorbidity was reported in 63/104 patients (60.6%). Factors associated to comorbidity were older age (median 40 years, IQR 28-50 vs median 60 years, IQR 44-61; p<0.001), CIRS comorbidity index (median 1, IQR 1-1 vs median 2, IQR 1-3; p<0.001), CIRS severity index (median 1.31, IQR 1.2-2.5 vs median 1.85, IQR 1.5-2.5; p=0.002), female sex (14, 35.0% vs 37, 60.7%; p=0.01), CIRS>3 (0, 0% vs 10,16.1%; p=0.006), Edmonton Frail Scale (EFS) (1, 2.4% vs 14, 22.2%; p=0.004), schooling <8 years (2, 4.9% vs 13, 20.6%; p=0.04), and polypharmacotherapy (11, 64.7% vs 38, 95.0%; p=0.006). Conversely, there were no significant differences with regard to clinical characteristics of IBD in relation to comorbidity (data not shown).

Table 4.

Sociodemographic and main clinical features of patients with no comorbidity vs comorbidity.

	No comorbidity, median (IQR)	Comorbidity, median (IQR)	p-value
Population	41 (39.4)	63 (60.6)	NA
Age	40 (28-50)	60 (44-61)	<0.001
CD-RISC25	61.5 (48-75)	65.0 (49-79)	0.34
Barthel index	100 (100-100)	100 (100-100)	0.30
BMI	23.1 (21.1-26.1)	24.2 (21.6-26.9)	0.31
CIRS comorbidity	1 (1–1)	2 (1–3)	<0.001
CIRS severity	1.31 (1.2-2.5)	1.85 (1.5-2.5)	0.002
Number of medications	2 (1–3)	4 (3–6)	<0.001

Abbreviations: Multiple Chronic Conditions (MCC); Connor Davidson Resilience Scale (CD-RISC25); Body Mass Index (BMI); Cumulative Illness Rating Scale (CIRS), Edmonton Frail Scale (EFS); Short Blessed Test (SBT).

Table 5 reports the bivariable analyses assessing potential factors associated to comorbidity in relation to the hospitalization status. The factors significantly associated to comorbidity, while accounting for hospitalization, were age >65 years (OR 5.30, 95% CI 1.81-15.55; p=0.002), female sex (OR 2.92, 95% CI 1-27-6.71; p=0.012), income <1000€ (OR 3.04, 95% CI 1.09-8.44; p=0.033), schooling <8 years (OR 5.09, 95% CI 1.08-23.96; p=0.039), EFS >5 (OR 12.56, 95% CI 1.48-106.45; p=0.020), and polypharmacotherapy (OR 10.41, 95% CI 1.85-59.38; p=0.008). In no cases hospitalization was an independent factor.

Table 5.

Bivariable analyses assessing potential factors associated to comorbidity in the whole cohort of inflammatory bowel disease patients in relation to the hospitalization status.

	Odds ratio (SD)	95% CI	p-value
Non-hospitalized	Base
Hospitalized	1.49 (0.75)	0.56–3.99	0.43
Age group 18–40 years	Base
Age group 41–65 years	2.53 (1.30)	0.92–6.91	0.071
Age group >65 years	5.30 (2.91)	1.81–15.55	0.002
Non-hospitalized	Base
Hospitalized	1.53 (0.75)	0.59–3.98	0.38
Sex (M)	Base
Sex (F)	2.92 (1.24)	1–27–6.71	0.012
Non-hospitalized	Base
Hospitalized	1.95 (0.97)	0.73–5.18	0.18
BMI <18.5	Base
BMI 18.5-24.9	0.44 (0.40)	00.07–2.57	0.36
BMI ≥25	0.84 (0.78)	0.14–5.20	0.852
Non-hospitalized	Base
Hospitalized	1.42 (0.67)	0.57–3.56	0.45
CD	Base
UC	0.97 (0.39)	0.44–2.14	0.934
Non-hospitalized	Base
Hospitalized	1.19 (0.58)	0.46–3.09	0.73
Income ≥1000€	Base
Income <1000€	3.04 (1.58)	1.09–8.44	0.033
Non-hospitalized	Base
Hospitalized	1.44 (0.69)	0.56–3.68	0.45
Schooling ≥8 years	Base
Schooling <8 years	5.09 (4.02)	1.08–23.96	0.039
Non-hospitalized	Base
Hospitalized	1.61 (0.78)	0.63–4.14	0.32
Smoke ≥ 4 cigarettes/daily, no	Base
Smoke ≥ 4 cigarettes/daily, yes	0.38 (0.26)	0.10–1.46	0.158
Non-hospitalized	Base
Hospitalized	0.83 (0.43)	0.29–2.31	0.72
EFS ≤5	Base
EFS >5	12.56 (13.69)	1.48–106.45	0.020
Non-hospitalized	Base
Hospitalized	1.37 (0.65)	0.54–3.49	0.50
SBT ≤9	Base
SBT >9	4.08 (4.50)	0.47–35.45	0.203
Non-hospitalized	Base
Hospitalized	0.72 (0.45)	0.21–2.48	0.59
Polypharmacotherapy, no	Base
Polypharmacotherapy, yes	10.41 (9.25)	1.85–59.38	0.008
Non-hospitalized	Base
Hospitalized	2.99 (2.97)	0.43–20.95	0.27
Active UC, no	Base
Active UC, yes	1.33 (1.09)	0.52–2.22	0.855
Non-hospitalized	Base
Hospitalized	2.07 (1.79)	0.38–11.23	0.39
Active CD, no	Base
Active CD, yes	0.14 (0.17)	0.01–1.44	0.099
Non-hospitalized	Base
Hospitalized	1.25 (0.60)	0.48–3.21	0.65
Oral mesalazine, never/past	Base
Oral mesalazine, current	0.51 (0.22)	0.21–1.20	0.124
Non-hospitalized	Base
Hospitalized	1.47 (0.70)	0.58–3.76	0.42
Immunosuppressants, never/past	Base
Immunosuppressants, current	1.29 (0.34)	0.22–0.83	0.22
Non-hospitalized	Base
Hospitalized	1.32 (0.65)	0.50–3.49	0.57
Biologics, never/past	Base
Biologics, current	0.81 (0.36)	0.34–1.95	0.642
Non-hospitalized	Base
Hospitalized	1.24 (0.60)	0.48–3.20	0.65
Systemic steroids never/past	Base
Systemic steroids, current	2.52 (1.55)	0.76–8.43	0.133
Non-hospitalized	Base
Hospitalized	0.84 (0.44)	0.30–2.34	0.74

Abbreviations: Standard Error (SE); Confidence Interval (CI); Body Mass Index (BMI); Crohn’s disease (CD); Ulcerative Colitis (UC); Edmonton Frail Scale (EFS); Short Blessed Test (SBT).

Discussion

We have herein focused on an overlooked topic in the field of IBD, namely the relevance of comorbidity in both outpatients and hospitalized patients. To corroborate common knowledge, hospitalized IBD patients had a more clinically active disease and needed systemic steroid therapy. Instead, among the novel findings, although our study should be considered as a preliminary observation, we showed that older age, female sex, low income, low educational level, frailty, and polypharmacotherapy were all more likely associated with comorbidity in IBD patients, regardless of hospitalization. Hence, IBD plays a key role in shaping the health profile of patients with multiple conditions, and even outside of severe cases requiring hospitalization, IBD contributes significantly to the occurrence of other health issues. Consistently, in our hospitalized patients IBD was active in many cases, with a greater severity in CD than UC, thus justifying the reason for hospitalization. The significance of these data is even stronger when considering that we prospectively collected a large amount of information per individual patient, including both clinical data and performance status indices, without the use of administrative databases. Indeed, a study including a larger cohort of IBD patients is needed to confirm our results.

As it appears clear from Table 1, there are little data about the impact of socioeconomic factors in determining hospitalization in IBD patients. In addition to the predictable associations between comorbidity and age, CIRS severity index, polypharmacotherapy, EFS >5, an association was also observed with female sex, income <1000€/month, and low educational level (<8 years). Although some of these factors may be associated with comorbidity per se, regardless of the underlying specific disorders, studies addressing these factors in IBD are largely lacking. In general, the increased prevalence of multimorbidity in females has been attributed to a higher prevalence of autoimmune diseases, mental disorders (e.g., depression, anxiety), and functional disorders (e.g., irritable bowel syndrome) compared to males. ³² In addition, women under 60 years old are more likely to have a contact with their general practitioner than men of the same age, thus making it easier to be diagnosed. ³³

Overall, our data showed that individuals with IBD frequently experience a heavy burden of comorbidity, potentially associated with lower socioeconomic status and limited education levels. A low income (<1000€/month) is not only associated with a comorbidity status ³⁴ but also increases the risk of it by about three times. Being low-income made people more likely to fall into a group of increased rates of mortality and illnesses and, at the same time, adult chronic illness may increase the likelihood of poverty. Indeed, a risk factor for every pattern of multimorbidity was material deprivation ³⁴ and a possible explanation may be the association between low income and eating an unhealthy and low-cost diet.^35,36 Although we did not specifically look at this issue in our study, an inappropriate diet has been found to be associated to malnutrition with all its complications, and it plays an even more fundamental role in gastrointestinal diseases, particularly in IBD. ³⁷ Additionally, low income and malnutrition are also strictly tied to the level of education which is per se a risk factor for developing comorbidities.^27,29,38,39 In this regard, it has already been shown that fewer patients with IBD achieved post-secondary education as compared to a control population. ⁴⁰ There are few inconclusive studies looking at the significance of socioeconomic deprivation in IBD. According to a review, it appears that socioeconomic, racial, and ethnic disparities in IBD are increasingly being identified, where both upstream (e.g., poverty, racism, underinsurance) and midstream (e.g., lack of social support, food insecurity, lack of access to IBD healthcare, psychosocial stressor) socioeconomic determinants cause downstream health outcomes (e.g., delayed IBD diagnosis, increased IBD severity, more hospitalizations, more disability and mortality). ⁴¹

Some differences also emerged between hospitalized and non-hospitalized patients. Regarding the prevalence of comorbidity, this was roughly 58% in non-hospitalized patients and 66% in hospitalized patients. The low difference in the two settings stresses the role of IBD in being associated with several complications and other comorbid conditions, where the hospitalization status has no influence. While this figure denotes a high prevalence, we might have expected an even higher prevalence in hospitalized patients, since our setting is that of internal medicine, where more than 90% of patients usually have this feature. However, the median age of hospitalized IBD patients was markedly lower (roughly 50 years old) compared to that of other patients admitted to our internal medicine ward (roughly 75 years old) ³⁸ ; age is indeed one of the most important factors associated to multiple chronic conditions.

As far as the clinical pattern of our outpatients compared to inpatients is concerned, it may appear paradoxical that more outpatients were on biologics. However, a large meta-analysis of 27 studies ¹³ highlighted that infliximab reduced the risk of hospitalization and surgery in IBD patients, so we can deduce that all new biologics and small molecules will reduce it even more significantly and hospitalization could be not strictly related to disease activity, but to other factors.

Among our results, frailty was also found to be rather common in hospitalized patients as previously shown in a retrospective and administrative database study. ¹⁴ Indeed, patients with IBD may be particularly exposed to both biological and social frailty, due to the impact of their disease on need for treatments (e.g., immunosuppressants, biologics), surgery, extra-intestinal manifestations, social stigmatization and low resilience.^42–46 This finding is of particular importance when considering the relatively young median age of our inpatients. IBD and related factors may therefore enhance frailty even in younger patients.

The present study has indeed some limitations that must be stressed. First, due to the small size of the sample, that did not allow us to differentiate between UC and CD, the inferential analysis must be generalized with caution. For this reason, this should be considered as a preliminary study that needs to be confirmed by further evidence. In addition, the low heterogeneity of the sample did not allow a deeper statistical analysis. Further, we could not make a distinction between comorbidity and multimorbidity,^6,30,31 not only due to the small size, but also because IBD certainly represents a prototype of index disease mostly affecting young adults and not older adults where multimorbidity is instead a common feature. On the other hand, our study has some strengths, such as the large amount of information collected per individual patient, including both clinical data and performance status indexes, and all data were collected prospectively and without the use of administrative databases. Finally, since we have particularly focused on socioeconomic factors, we believe that studying IBD as a whole, without splitting into UC and CD, is appropriate, because those factors are certainly shared between diseases. In fact, it is expected that social determinants, including financial strain, health literacy, and access to education, impact disease management and outcomes in a similar way in UC and CD. ⁴¹

To conclude, patients with IBD displayed a high prevalence of comorbidity, not only because of disease severity and its complications, but also possibly related to low socioeconomic status and poor educational level, in both the hospital and outpatient settings. Future studies addressing these issues to confirm our preliminary results are warranted.

Appendix.

Abbreviations

anti-TNFα

Anti-tumor Necrosis Factor alpha

BMI

Body Mass Index

CD Crohn's Disease

CD-RISC25

Connor Davidson Resilience Scale

CI

Confidence Interval

IBD Inflammatory Bowel Disease

CIRS

Cumulative Illness Rating Scale

EFS

Edmonton Frail Scale; Inflammatory Bowel Disease

HBI

Harvey-Bradshaw Index

IQR

Interquartile Range

OR

Odds Ratio

PROMs

Patient-Reported Outcome Measures

SBT

Sort Blessed Test

SMAC

San MAtteo Complexity

UC

Ulcerative Colitis.

ORCID iDs

Marco Vincenzo Lenti https://orcid.org/0000-0002-6654-4911

Gino Roberto Corazza https://orcid.org/0000-0001-9532-0573

Consent to participate

All patients provided a written informed consent to participate in the study.

Consent for publication

Obtained from all patients.

Data Availability Statement

All data relevant to this study have been here reported; additional raw data can be shared upon reasonable request to the corresponding author.*