Introduction

Inflammatory bowel disease (IBD), encompassing Crohn’s disease (CD) and ulcerative colitis (UC), is a chronic inflammatory disorder of the gastrointestinal tract associated with substantial morbidity and healthcare utilization [1]. Estimates of the global IBD prevalence range from 4.9 to 6.8 million individuals, reflecting differences across datasets, with the largest patient populations observed in China and the United States [2,3]. The prevalence of IBD continues to rise and is projected to approach 1% of the population in several industrialized regions, such as Europe and North America, within the coming decade [4]. Annual direct healthcare costs for IBD, driven by outpatient care, hospitalizations, and pharmacotherapy, are considerable in high-income countries, averaging between $9,000 and $12,000 per patient [5].Optimal management of CD and UC focuses on inducing and maintaining remission, which may in turn reduce the need for medical interventions and overall healthcare costs [6].

Rehospitalization following an initial IBD-related admission is increasingly recognized as a critical indicator of disease course, quality of care, and healthcare utilization. High rates of rehospitalization not only reflect ongoing disease activity but also contribute substantially to healthcare costs and resource allocation [7]. Internationally, hospitalizations affect a significant proportion of IBD patients and account for a disproportionate share of healthcare expenditures [8]. While most patients experience only occasional hospitalizations, a subset of high-need patients undergo frequent admissions, many of which may be preventable, highlighting the burden of recurrent hospitalizations [9]. Although rehospitalization patterns in IBD have been described in individual international studies, comparable data from Germany are limited. Accordingly, this multicenter retrospective cohort study aimed to describe 12-month rehospitalization rates after IBD-related inpatient treatment in Germany and to identify administrative and clinical factors recorded in hospital ICD-10 coding that are associated with rehospitalization.

Methods

Results

Baseline characteristics

A total of 3,202 patients were included (1,642 with Crohn’s disease and 1,560 with ulcerative colitis, Fig 1). The mean age was 44.2 years (SD 21.4), and 50.1% of patients were female. The median length of hospital stay was 5 days. Overall, 47.0% of patients were treated in internal medicine, 19.3% in gastroenterology, 10.0% in pediatrics, 21.7% in surgery, and 1.9% in other departments. The most common comorbidities were hypertension (20.4%), gastritis/duodenitis (18.0%), and hypokalemia (14.9%, Table 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Baseline characteristics of the study sample.

https://doi.org/10.1371/journal.pone.0346270.t001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Selection of study patients.

Selection of patients hospitalized between January 2019 and December 2023 in 49 German hospitals. From 1,706,874 hospitalizations, 3,214 patients with a primary diagnosis of inflammatory bowel disease (ICD-10: K50 or K51) were identified. After exclusion of patients with in-hospital mortality during the index admission, 3,202 patients were included in the final analysis.

https://doi.org/10.1371/journal.pone.0346270.g001

Discussion

The 12-months rehospitalization rate in this multicenter German cohort is comparable to findings from other Western healthcare systems, although reported rates vary considerably depending on study design and follow-up duration. In a large meta-analysis conducted in the United States and Australia, the 30-day readmission rate was 18.1%, while the 90-day rate was notably higher at 26.0% [10]. The observed increase in the readmission rate within the 1- to 3-month interval is not unexpected and is consistent with the findings of the present study, which demonstrated a median time to rehospitalization of 92 days. A Canadian population-based study reported an annual readmission rate of approximately 20% among patients with IBD, which is comparable to the findings of the present study, where the 12-months rehospitalization rate was 22.6% [11]. In contrast, another North American cohort reported a substantially higher rate of unplanned readmissions, with 39.2% of patients rehospitalized within one year [12]. These findings highlight the heterogeneity in reported rehospitalization rates and show the need for region-specific analyses to support individual targeted healthcare strategies.

With regard to potential factors associated with rehospitalization in patients with IBD, our statistical analyses using univariable and multivariable logistic regression identified several variables of interest, some of which are consistent with findings from previous studies (Table 2). Surgical treatment emerged as one of the strongest factors associated with an increased 12-months rehospitalization. Admission to a surgical department during the index hospitalization was associated with rehospitalization, suggesting that patients with complicated disease courses represent a clinically vulnerable subgroup. In line with our results, other studies have similarly reported an increased risk of rehospitalization in patients with fistulating or perianal disease, the need for surgery, ileostomy in CD, colostomy in UC, and large intestinal resection in general [13–16].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Association between demographic and clinical variables and 12-months rehospitalization in patients hospitalized for IBD (multivariable logistic regression).

https://doi.org/10.1371/journal.pone.0346270.t002

The association between surgical treatment and rehospitalization likely reflects the burden of advanced or complicated disease, which often entails recurrent clinical instability. In this context, electrolyte disturbances are common sequelae in IBD, particularly in patients with high intestinal output, active colitis, infectious complications, or short bowel syndrome following resection. Consistent with these mechanisms, hypokalemia was independently associated with an increased rehospitalization rate in in both univariable and multivariable regression analyses. This finding underscores the clinical importance of early detection and correction of electrolyte imbalances as part of post-discharge management in IBD.

Age showed an association with rehospitalization rates in univariable analysis, with patients aged ≥51 years demonstrating higher overall rehospitalization rates. However, this effect was not sustained after adjustment for comorbidities. These findings suggest that age alone is not an independent factor associated with rehospitalization in IBD. This observation is consistent with previous studies reporting inconsistent age effects, with some cohorts indicating lower rehospitalization risk in older patients, while others identified younger age as a risk factor in CD [13,14]. Taken together, these findings indicate that age alone may be less relevant than multimorbidity and overall disease complexity in explaining increased rehospitalization rates, highlighting the need for stratification models that incorporate comorbidity indices rather than patients’ age.

Nutritional and metabolic status have gained increasing attention as determinants of outcomes in IBD. Alterations in body composition, on the one hand muscle wasting, on the other hand excess obesity, are frequently observed in patients with malnutrition and sarcopenia and are associated with impaired quality of life [17]. Furthermore, both underweight/sarcopenia and obesity, have been identified in US studies as being associated with increased rehospitalization rates, poorer clinical outcomes, and higher annual healthcare costs [14,16,18,19]. Although anthropometric and body composition data were not available in our dataset, metabolic comorbidities such as type 2 diabetes and hypertension were associated with an increased rehospitalization rate in adjusted analyses. These findings indirectly support the hypothesis that metabolic dysfunction may contribute to recurrent hospitalization among IBD patients.

Neither sex nor the IBD subtype was independently associated with an increased rehospitalization rate in our cohort and evidence from the literature on these factors is inconsistent. Only few studies have reported an impact of sex on the risk of rehospitalization. For example, one study found that male patients had a higher risk of readmission within 30 days after discharge. The same study also reported an increased risk for patients with CD within the same observation period [13]. Consistent with most previous studies, our analysis showed that neither sex nor the IBD subtype is associated with higher rehospitalization rates, suggesting that rehospitalization is more driven by comorbidities and disease complications.

This study has several limitations that should be acknowledged. First, although the Section 21 dataset includes an anonymous patient identifier, this identifier is hospital-specific and cannot be linked across institutions. Consequently, patients without a recorded rehospitalization in the participating hospital could not be distinguished from those who may have been hospitalized in non-participating hospitals or who died after discharge. For patients discharged in 2023, follow-up was truncated at the end of the observation window, resulting in potentially shorter observation periods for this subgroup. Second, although the dataset covers 49 hospitals across different levels of care, clustering by hospital could not be implemented because the data were provided in aggregated form. Nevertheless, all hospitals submit data under standardized national coding and reimbursement regulations, reducing the likelihood of systematic inter-hospital variability that would materially bias regression estimates. Third, important clinical variables such as disease activity, medication use, surgical history, nutritional status, and laboratory parameters were not available in the administrative dataset. This limited the ability to adjust for disease severity or to conduct meaningful subgroup analyses. Given these constraints, additional stratified analyses—for example by IBD subtype—were not performed to avoid over-interpretation of incomplete data. Fourth, although multicollinearity was assessed and excluded (all VIF values <1), residual confounding cannot be fully ruled out in observational research. Finally, the inability to differentiate between planned and unplanned rehospitalizations, as well as the absence of outpatient data, may have influenced the interpretation of rehospitalization patterns.

This study also has notable strengths, including a large sample size, multicenter design across different levels of care, and standardized data acquisition, which increase the robustness and generalizability comorbidities of the findings within the German healthcare system. Adjustment for relevant comorbidities allowed identification of clinically meaningful factors which are associated with higher rehospitalization rates. Importantly, our results highlight the clinical relevance of identifying factors associated with rehospitalization during the index hospitalization.

Using administrative hospital data, we identified several ICD-10–coded comorbidities and treatment characteristics associated with higher rehospitalization rates. These findings highlight subgroups that may benefit from improved post-discharge management, although clinical interpretation is limited by the administrative nature of the dataset.